Wednesday, May 18, 2011

Techjargon and Nomenclature


The title of this post combines two entries from my old Tough Guide to the Known Galaxy. We're in no rush, so read them and report back for further discussion.

Nomenclature

Techjargon

These subjects are brought back to mind by the comment thread on Space Warfare XIII, which has now reached a preposterous 836 comments, taking it past the mere 820 of Space Warfare XII. I have no moral grounds for ragging on the commenters for being bloodthirsty, given that I served up the topics. One of the many subthreads of the discussion turned to language, particularly (of course) military language, in all senses of the phrase.


How soldiers (or even civilians) talk is an issue that science fiction, fantasy, and historical fiction all face on much the same terms. Roman legionaries presumably swore in Latin, but surely not Ciceronian Latin. Nor was it even Caesarian Latin - at any rate not the Latin of De Bello Gallico, though Gaius Julius could no doubt express himself eloquently in good centurions' Latin when he needed to.

In a few cases we have direct evidence of how soldiers spoke: English troops in France during the Hundred Years' War said "God damn" so much that goddams, or godons, became French slang for the English. This is rather reassuring.

Contrary to a popular trope, most military swearing is not actually very imaginative. Yes, there is the occasional CPO or sergeant with Shakespearean mastery of English invective, but service field language relies mostly on a few very basic concepts, generously repeated.

Scroll down this old Language Log post (it isn't long) for what, from my recollection, is a very good summary of gruntspeak.

The example of godons also reveals another basic truth - styles in swearing change, over time and from culture to culture. In particular, while the Fourth Commandment still surely comes in for very frequent violation by service personnel, at least in 'Murrican military service the pride of place - as reflected in the Language Log example - now surely goes to what, with a delicate nod to spam filters everywhere, I shall call 'the eff word.' Even simpler military evolutions than passing pliers would be impossible without it.

Of course the eff word is no recent coinage. While its use by medieval English grunts went unrecorded, it surely did not go unspoken. (Shakespeare makes a sly grammatical reference to a 'focative case.') But it probably was used only in a fairly literal sense. In particular, acronyms such as FUBAR and the now-generalized SNAFU - like most other military acronyms - seem only to have become widespread during World War II.

And this is the point at which things get tricky. "Eff you!" strikes me as fairly timeless, nearly as home in the 31st century as in the 21st - or even in the 11th century, if the speaker is a Viking, given that "Odin damn!" somehow just doesn't work in our era. On the other hand, "all effed up" has - to my ear - a bit too much contemporary flavor, as though the speaker is not just a universal grunt, but specifically a current era 'Murrican grunt.

In this case, not only the YMMV principle but intended audience comes into play. Much military SF is, let's be honest, not just war porn but specifically Ameriwank war porn. Thus, for the stereotyped Baen audience (by no means identical to the full Baen readership, but surely a non-trivial part of it), 31st century espatiers who sound like they just shipped out from Camp Lejeune are a feature, not a bug. Note that this probably applies to an audience with inverse Ameriwank politics as well.

And it applies with even more force when we move away from swearing like a soldier to the more technical aspects of military language, such as names for weapon systems. SPQR may evoke Rome, but SPQ-31B evokes recent era Western militaries, and particularly (again) the 'Murrican military.

(Somewhat off message, but it surprises me that no one else seems to use USAF style sequential type numbering. Even the Soviet era Russians used manufacturer-centric aircraft designations, e.g. TU-95.)

Alphanumeric designations strongly evoke the recent era - great if the connection is intended, awkward and even frame-jolting if the setting is meant to have a more distant flavor.


More ambiguous are generic terms for weapons or ship classes, such as battleship and cruiser for combatant spacecraft. I have tended in asides and comments to come down fairly hard on such terminology - perhaps more so than is justified. For one thing, the opposite extreme of renaming familiar items can be horribly clunky. I've been unable to track down, among hundreds of comments here, a brilliantly devastating invented example, weapons with an imaginary name, described as "like swords, but more awesome." (Whoever wrote that, step up and claim your well-earned prize.)

Looking at different eras is, alas, rather unhelpful. Older terms for ship types, such as carrack or galleon, seemed to refer not to the ship's functional role but to structural features that are often now thoroughly obscure. (On the other hand, a 17th century Florentine type was called a bastardella, 'little bastard' - an expression that may not clarify its mission, but is timelessly nautical.)

Battleships and cruisers, by comparison, at least refer to familiar missions, at least in their 'classical' usage. (And by SF convention, cruiser seems to retain the sense of a large, fast independent patrol craft, not a heavy escort type.) I have argued against the automatic lifting of this nomenclature, as introducing a bias into our thinking about space warfare. But having said that, it is hardly implausible, in operatic settings, to have some heavy spacecraft, optimized for fighting power, that operate in main-force constellations, and somewhat smaller ones, optimized for mobility, that operate independently.

Details of an imagined technology come into play here. If the 'cruisers' are individually larger than 'battleships,' as is plausible (more propellant, larger crew), the classical terminology becomes a bit misleading. But this does give you a great excuse for battlecruisers, and the Rule of Cool is hard to resist in this case.

Another consideration is particulars of a future history. My old 'Human Sphere' setting was a variant on the standard First and Second Empire theme, and centered on nascent 'Second Empire' trade federations in the 28th century.

The 'First Empire' had no heavy space force in its salad days because it needed none. (Who was it going to fight?) So that particular setting should display a great discontinuity in military affairs. People might well cast back into history books to revive terms like frigate, but on the whole their combatant spacecraft and military institutions would more likely be adapted from police, exploration, or other activities and organizations that have some quasi-military features. Thus, in this particular setting, survey ship comes to have nearly the usual connotation of cruiser.

On the other hand, if your setting has space militaries established by terrestrial Great Powers in the 22nd century, with the latter having substantial continuity with the present day, your nomenclature jumps through a very different set of historical hoops, and cruisers can just be cruisers.



Discuss.


The image of a space battlecruiser comes from this SF website. I nabbed it via Google Images - one of eight space battlecruiser images that come up before the first image of the seagoing kind, HMS Hood. Not until the second page do you find any image of a ship that fought at Jutland.

231 comments:

1 – 200 of 231   Newer›   Newest»
Sabersonic said...

For the Nomenclature of familiar military equipment/vehicles/ships, well one could argue that such unsettling usage of said terms could be that such associations are ingraned to us culturally. Or as the saying goes "If it sounds like a duck and it walks like a duck..."

If I remember correctly, in a previous Space Warfare blog entry, someone argued that it would not be suprising if said nomenclature is used on an object that doesn't quite fit the deffinition mold 100% because they are cultural hangovers. An expectation that tends to stick when logically it shouldn't.

Of course, trying to argue to a layman as to why a particular combat spacecraft isn't dubbed a naval type even though its mission profile and weapon layout seemed suspeciously familiar may be a creative difficulty that would force potential SF writers to just accept Naval nomenclature of ship types even though there isn't a logical reason to do so. The same could be said of using the naval model instead of the more accurate air force model.

As for the *ahem* curses and swears that one would expect to evolve in the distant future compared to the presend day or in a historical piece. Well, all I can say that though it would be more accurate to lead such language to its ultimate conclusion, keep in mind that adding pages of glossary on the slang and jargon alone does eat up space needed for said story itself.

That and the audience has not only a particular expectation, but a limited relatability of the characters and having curses that are nowhere near remotely similar to what they know (even if one explained the logic behind them) would break the suspension of disbelief and might even call into question many parts of a story.

In short, when it comes to language its basically a lose-lose situation for the Author.

- Hotmail Address
Gmail Address

Mangaka2170 said...

Liking the classic StarCraft pic there, Rick.

I've been doing some thinking on this subject in the context of my own firm sci-fi project. For those who are not familiar with it, there are four major factions in the solar system of the 22nd century, two of which are at war with each other, and their military organization and jargon differ greatly.

In the case of the Earth Forces (mostly North America, as they appropriated the greater part of the old Earth Federation's space force), the Space Command is, in many ways, modeled after the US Air Force (explained as Air Force Space Command becoming its own branch of the military in the 21st century); warships tend to follow a "bomber crew" model, although there are plenty of naval elements that were added as needed. Anyways, the Earth Forces tend to organize ships by size and function based on naval nomenclature.

There are three basic classifications of warships according to Earth Forces doctrine: Escort, Capital and Super-capital. Frigates (escort) tend to be smaller ships bristling with smaller gun batteries and sporting a tight defense grid, while destroyers are simply frigates with a main weapon (usually a big ol' laser). Meanwhile, capital ships include carriers, cruisers (essentially bigger destroyers) and battlecruisers (cruisers with multiple main weapons).

There are two different classes of super-capital: Dreadnaughts and Barges. A Dreadnaught is a squadron-level command ship (a squadron is analogous to a Fleet or an Army Group in Space Command doctrine) that is approximately a kilometer in length, bristling with gun batteries, several different primary weapon systems and a large complement of space fighters (although space fighters are a more strategic weapon used for patrols and establishing orbital supremacy), and is heavily armored, with a powerful engine and large propellant reserves to match. While a Dreadnought is capable of operating on its own, it is almost invariably escorted by multiple capital-class ships in order to protect the vast resource investment the dreadnaught represents.

The Barge, on the other hand, is a completely different ship; a strategic logistics ship that the Earth Forces no longer have due to the terms of the treaty made when the Earth Federation was balkanized. A barge is about the size of an O'Neill-type colony cylinder and exists to carry other ships between planets in a hurry. It can do this by virtue of the many fusion reactors powering its solar torch, the firing of which the barge's commander is required by law to announce to everyone within half a million kilometers at least 24 hours in advance, and is not permitted to go anywhere near a planet, moon or other settlement (L2 and L3 points are common parking areas for these monstrosities). The barge has no weapons other than a sufficiently tight defense grid, not that it would need anything other than its torch to fight off attackers with.

The Martians, on the other hand, no not have such a rigid classification system. A battlestar is any kind of warship, a basestar carries fighters, a laserstar has a laser as a primary weapon, a gunstar has a railgun as a primary weapon, a missilestar carries a lot of missiles and a jammerstar is an ECM ship. A ship's designation is based on its primary role, and the only distinction made as far as size is concerned is heavy (for capitals and super-capitals), medium (escorts) and light (fighters). So, the Dreadnaught defined above might be classified as either a heavy laserstar or a heavy basestar, but it would most definitely be a heavy battlestar. A barge would simply be a heavy basestar, as would a carrier, but the engine flare would be sufficient to distinguish between them.

Milo said...

My impression with today's "acronym culture" is that it suggests people have more types of things than they can come up with creative names for, and so they can't be bothered trying and just slap on a matter-of-fact description.

The tendency is stronger with equipment that people perceive as "unimportant". Who cares if that thing up there has the bland and longwinded name of "Unmanned Arial Vehicle", if it's "just a drone" anyway? No-one likes drones.

But if a particular type of equipment becomes highly successful and directly relevant to many people's lives, people are going to come up with a less boring way to refer to it. If the bureaucracy doesn't then the grunts will.



Rick:

"Older terms for ship types, such as carrack or galleon, seemed to refer not to the ship's functional role but to structural features that are often now thoroughly obscure."

Yes, but word meanings can change, so (in theory) a term whose etymology originally came from a specific shape, can later come to be understood as referring to a specific role, and vice versa.

The main reason we think of "carrack" and "galleon" as referring to a specific shape is because we stopped using them. Since the terms are now seen as "period terms", we invariably associate them with specifically the ship shapes that were built when the terms were in use. But if, when carracks-as-we-know-them were first phased out, navies had been billing one of their ship types as "like a carrack", and so on through several tech revolutions, then we'd now perceive the term as more general.

Consider how "destroyer" has come to be seen as a general term for a small and fast capital ship (what you could also call a frigate), not necessarily one designed for destroying a specific type of target. Or for that matter, the targets the original destroyers were meant to destroy - why did the US military call them "patrol torpedo boats" when their primary function wasn't patrolling? Don't we still use "sailing" to refer to travelling from one place to another by ship, even when it isn't actually being propelled by sails?


"So that particular setting should display a great discontinuity in military affairs. People might well cast back into history books to revive terms"

Which means that, of course, they will not be naming their spaceships after historical ship classes, but rather after what 28th-century Society for Creative Anachronism members perceive historical ship classes as having been like.

Also depending on how much language has changed, they can end up seriously misusing some terms - it's easier to somehow end up using the term "cruiser" for a ship that stays near home if it doesn't recognizably mean "cruiser" in your language.

Milo said...

Sabersonic:

"Well, all I can say that though it would be more accurate to lead such language to its ultimate conclusion, keep in mind that adding pages of glossary on the slang and jargon alone does eat up space needed for said story itself."

The nice thing about swear words, though, is that you don't really need to understand them in order to be able to get what's being said. If someone says "Shfsdh you!" then you can probably figure out the intent, particularly if it's in an angry tone and it's obvious from context what the speaker is angry about. You don't need to know whether the term etymologically refers to a sexual act, or excrement, or a religious belief in final judgement in the afterlife. (And before you ask, yes, I created that word by mashing my keyboard randomly.)

In many sentences the use of swear words is effing irrelevant, and you could remove them without changing an effing thing about the diction.

Of course what we're all missing is that, no matter what swear words future soldiers use, future book publishers will still want to censor them.



Mangaka2170:

"A barge is about the size of an O'Neill-type colony cylinder and exists to carry other ships between planets in a hurry."

This sounds odd to me - I associate the term "barge" with something that can carry a lot of cargo cheaply, at the expense of being slow. (Hence my primarily using it in a space environment to refer to Hohmann-trajectory cargo ships.)

Anonymous said...

What to call space-going warships? We could call them names of old navel classes like cruiser, destroyer, frigate, etc; we could combine a discription of their size and armerment (heavy missile ship, light laser, etc); or we could name them after the mission that they perform (interceptor, orbital monitor, mobile support base, fast logistics transport, interplanetary interdiction, etc).
I think that calling these ships something that reflects their mission, rather than something about how they are equipted or borrowing old navel terms, makes more sense.

Characters cursing should be more dependent on how relevante to the character and to the story that cursing is...whether the characters curse or not, their speech needs to be unique to that character. One character may use flowery, imaginative curses; some may use the most basic oaths, while other aren't really any good at it at all. And above all, it needs to make sense in context with the story itself.

When naming a gadget, I try to either keep it vauge (if it isn't central to the story), or make it simple if it is importaint to the story.

Ferrell

Tobias said...

The swearing used in MSF, both bean and TV like the new Galactica, does feel a little strange to me.

The only person I can think off who swears the same way as soldiers do according to these authors is my aging aunt who works as a nun.

Does anybody know, if americans really swear that way?

Now the vessel class names, are to my eyes an indicator of how much the good old days of nationalism are remembered. New shipnames imply an unified old superplanetary administration, or a peacefull space exploration. Old shipnames imply a barely unified earth thorwn into to interstellar war to early.

In softer SF old ship names are used just because O'Brian and Forester used them, never forget this.

Geoffrey S H said...

I just make up stuff for my sopace craft names:

Fusilaire- corruption of "battle-rocket" in French. The largets craft with the best payload.

Intiller- "Interceptor killer". medium craft, essentially half a heavy that can be sent off to two missions at once or be sent together- if one is lost, you still have fully half the fighting capability left.

Shreib- corrupted German for "small thrust". Escorts the mediums and heavies, providing defensive fire from different angles given their place somew3hat afr out from constellation formations. if they are lost then the loss of life and equipment is small.

Handlestern- German for "merchant star". merchant craft.

Ship= Thrust. A warship is a "battle thrust". Squardons are "constellations" of course, and flotillas are "clusters". Navies are "astraliers" (from "astral"- star), and captains are "misscoms"- mission commander.

If the pilot of a satellite in NASA now is an "ace"- no reference to gthe best pilots in an airforce intended here), then a pilot of a spacecraft is an "ace", and a flagship is an "ace-thrust".

I aim to ransack current NASA ransk and terminology as much as possible, with non-anglosphere words beign used as well.

Brazillian themes and eventually Rwandan terminology will also creep in towards the end of the PMF in the setting.

Cambias said...

From the literary point of view, nomenclature has two functions. The first is to tell you something about the setting, and the second is to help tell the story. The more you pack into a term for the first purpose, the less useful it is for the second purpose.

Example: "Spaceship" tells you almost nothing about the setting. It's generic. Could be anything from a Mercury capsule to the Death Star. BUT the reader can gloss the word without stopping to think, just as with other generic words like "food" or "house."

By contrast, if you call the thing that carries people from planet to planet a "RTH" for "Reusable Transit Habitat" you're telling the reader this is a hard-sf world, space vehicles are thought of as payloads rather than "ships," hardware is secondary to the mission, and it's a bureaucratized, top-down organization.

Great stuff, but every time the reader runs across that little acronym, his brain bounces out of the story while he remembers what it means, translates "RTH" to "spaceship" and goes on with the story.

Tony said...

Let's see...

Does anybody know if Americans swear like they do in military SF works? Based on my experience in the US Marine Corps and tangential contact with the USN, I'd have to say they swear worse. I think the closest John Ringo, for example, ever came to the truth was a Philino (as in from a water world settled by Philinos) Imperial Marine in March Upcountry. He used the eff-word (pronouced "puck", due to the inability of native Tagalog speakers to pronouce f's correctly) as unselfconscious punctuation in every sentence.

Or, perhaps this true story illustrates it best:

About a month after 1st Battalion, 1st Marine Regiment got back to Camp Pendleton after Desert Storm, word came down from the commanding general of 1st Marine Division that we needed to stop talking like we were overseas. He was getting too many complaints about Marines' language. This was announced in morning formation one day. After the formation was dismissed, one of the Marines in my platoon asked, "So, Sergeant E, the General says we need to watch our language...what do you think?" Without missing abeat -- and, really, without thinking about it at all -- I replied, "F**k the General."

Now, that Marine was kind of the platoon clown, and he was probably relying on my being off-guard to get that kind of answer. But that just serves to make the point -- he could rely on me for that pithy epithet, and I didn't disappoint.

Milo:

"Don't we still use "sailing" to refer to travelling from one place to another by ship, even when it isn't actually being propelled by sails?"

No, we don't. For steamships it's "steaming". This usage extends all the way back to when steam engines were auxiliary on seagoing vessels. It was originally used to make the distinction between using wind power and auxilary steam power.

For ships equipped with diesel motors the technically correct term is "motoring". But that's a term that can also be used for a small boat with a 100 HP Volvo-Penta, so for big ships the standard term is generally "transiting".

Anonymous said...

(SA Phil)

Back when I was envisioning a rail-gun / missile dominated scenario.

I imagined the combat craft would be "Gunships"

Another term I like is "Fighter" although not the single seater swoopy combat craft. More like a "frigate" or "cruiser" by most space is an ocean comparions.

The more I think about what a plausible combat in space would look like -- the less I think something like crew size is going to matter.

Since I am begining to think that in any such combat environment "Robot-Ship" would rule the day.

Byron said...

I personally think that nomenclature will depend heavily on the traditions that the space force inherits. AV:T is a good example. IIRC, Olympia and Xing Cheng had naval traditions, and called their ships by naval names (frigate, cruiser, patrol escort, etc). Median's forces came from the Air Force, and called their ships things like fighters and attack craft, even though they were the same size as everyone elses. Novaya Rossiya's space force was a branch of the artillery, and called their ships artillery platforms.
The naming of ships is a good way to give people a look into the traditions of a force. If it's called a navy, uses naval ranks, and calls its ships "frigates" and "cruisers" then it shouldn't use air force terminology.
Of course, words might get corrupted. Look at "fighter" above, or the change in definition of "battlecruiser".
However, I do expect some sort of designation system to be in use, particularly for stuff like electronics. There's just too may systems, and they will need names.
The same will apply to ships. There will be some way of designating ships using a few letters and numbers. If you want something different, use pennant numbers.
I'm not sure what the soviets did for this.

As to swear words, to avoid the censors, use Mandarin Chinese.

Byron said...

Tony:
I thought you were in the Navy. Or was there a Marine detachment on Long Beach?

No, we don't. For steamships it's "steaming". This usage extends all the way back to when steam engines were auxiliary on seagoing vessels. It was originally used to make the distinction between using wind power and auxilary steam power.

For ships equipped with diesel motors the technically correct term is "motoring". But that's a term that can also be used for a small boat with a 100 HP Volvo-Penta, so for big ships the standard term is generally "transiting".

For the general public, I'm fairly certain that sailing is still used more frequently. It may not be among professionals, but it's still in common usage.

Byron said...

I've found out what the Russians do. They just assign a three-digit number to each ship. Submarine numbers are prefixed with K. China is the same, except that submarines don't have the K.

Tony said...

Byron:

"I thought you were in the Navy. Or was there a Marine detachment on Long Beach?"

There was a Marine detachment on the Long Beach. I was assigned to it from 1987-89. We had a guard platoon of 40-45 Marines and a small admin section with an armorer and a couple of personnel clerks. I was an NCO in the guard section and gun captain on Mount 52 (portside 5"/38 gun).

Tony said...

Byron:

"I've found out what the Russians do. They just assign a three-digit number to each ship. Submarine numbers are prefixed with K. China is the same, except that submarines don't have the K."

The Russians name their ships, including submarines. Some names have certainly changed since the end of the Soviet Union, and transfer of certain vessels to the Ukraine.

The Soviet type classification system used "antisubmarine warfare ship" and "missile ship" for corvette and frigate sized vessels, and "large antisubmarine warfare ship" and "large missile ship" for destroyer and cruiser sized ships. Their VSTOL carriers were designated "heavy aviation cruiser".

Byron said...

Tony:
I was speaking of numbering, not naming or type designations.
And thanks for the information on Long Beach. Though I do have to wonder about the 5/38 in the late 80s. :-)

Tony said...

Byron:

"I was speaking of numbering, not naming or type designations."

Oh, hull numbers. Everybody uses those. The Soviets did have a pretty random method there. I don't think they fooled anybody.

"And thanks for the information on Long Beach. Though I do have to wonder about the 5/38 in the late 80s. :-)"

Hey, it was better than nothing. With radar fire control, we would have had some chance against Iranian Silkworms -- if it had ever come to that -- which were large and subsonic. And it would have been a good backup to Marines and sailors with small arms when fighting Iranian speedboats in the Persian Gulf.

Anonymous said...

Space forces will probably begin as extensions of air forces, and air force rank structures and units will remain suitable for orbital deployments of hours to days in small craft with ten or fewer crew. However, interplanetary voyages of weeks or months in larger craft with tens of crew may force them to become more 'naval'. If space forces eventually have spacecraft large enough to have crews equivalent in size to air force units such as squadrons and wings, the officers commanding these craft will be acting more like naval officers than air force officers.

R.C.

Milo said...

R.C.:

"However, interplanetary voyages of weeks or months in larger craft with tens of crew may force them to become more 'naval'."

But the question is whether they have continuity of tradition with the navy, or whether the air-force-derived space forces will need to reinvent the wheel.

Scott said...

Milo, it's unlikely that the Air Force-derived troops would NOT re-invent the wheel, at least until the Earth-side government got tired of building replacement ships.

Then there'd be a BIG shift while the submariners took over.

Something that would be really instructive for prospective writers to look at would be a Navy interior communications manual. Unfortunately, it seems that the ICM isn't available.

One of my biggest pet peeves is calling bearings in 3d. And are you using decimal degrees, DMS, or mils?

If you do use mils, make sure you actually use them right (looking at you, Mr. Michael Z. Williamson)!

Rick said...

The main reason we think of "carrack" and "galleon" as referring to a specific shape is because we stopped using them.

Good point. In fact, both names were current during the 15th-16th century nautical tech revolution, and the configurations they referred to changed greatly.

In fact, what 'galleon' connotes to most people is substantially different from what it connotes to me, because the word came in the 17th-18th century to mean the Spanish equivalent of an Indiaman.

This last usage is what survived in the popular culture, though for much of the 16th century galleons were distinctly low-built, handy types - nearly opposite to their familiar image.

Having this sort of implied evolution of a type name in a future setting would be an interesting little challenge.

(Not to mention the range of types that have at some time or another been called a frigate.)

Milo said...

Scott:

"One of my biggest pet peeves is calling bearings in 3d. And are you using decimal degrees, DMS, or mils?"

In 3D, bearing is a two-dimensional value. (The third dimension in polar coordinates is distance, which isn't a part of the bearing.) Furthermore, the two numbers composing the bearing will be somewhat different in meaning. (Look at a globe of Earth. Lines of latitude and lines of longitude do not work the same way.) This means that in addition to deciding which direction is forward, you also need to decide which side of the ship will be treated as "north/south" for the purpose of calculating bearings.

It seems that when giving rough estimates of 2D bearings, people will often name it by a clock value (so "hostile at your seven o'clock" means someone is behind you and slightly to the left) rather than degrees. This doesn't seem like it'll catch on in space, since no-one has yet built a three-dimensional clock. Still, people might come up with informal names for particular directions - although due to the extra degree of freedom, it would take quite a few names to cover them all, so it might not work that well.

Byron said...

Milo:
I'm not sure how they make clocks where you're from, but all of my clocks are 3D.

Scott:
I've found a couple manuals for interior communications electricians. Is that what you mean?

Anonymous said...

So, future combat spacecraft might be called gunships, interceptors, or interplanetary bombers at first; but later might call the same types frigates, destroyers, corsairs, or crusiers. Then again, whatever they get named to start with, might just stick (with just a few modifications over the years).

Ferrell

Cambias said...

In the name of realism (or at least creating the illusion of same) I would suspect the formal names for combat spacecraft types would be unrelated to navy or air force terminology -- if only to avoid confusion during the era when they are all part of one super-battlefield. Airplanes never were "air battleships" or "air cruisers."

In the name of entertaining literary style, one can posit that spacecraft are unofficially called whatever you want them called. Welcome to the Mobile Laser Platform Liberty, which everyone calls a "space cruiser."

Teleros said...

Going back to the OP and regarding the influence modern American military terminology has (or appears to have), two things spring to mind:

1. The heavy use of acronyms and the like today. If you want something a bit different, have your grunts use "spades" rather than "E-tools". Have the commander-in-chief of the outer solar system *not* be referred to as "CINCOUTSOL" or similar - perhaps he's just "Neptune", for the location of his HQ.

2. Other bits of language, like the phonetic alphabet. "Able, Baker, Charlie" will sound more British than NATO's "Alpha, Bravo, Charlie", for example. Spelling words like "honour" properly ( ;) ), having autumn instead of fall, and "Ceres Colony Constabulary" instead of the "Ceres PD"... Of course, a lot of the American system came initially from Britain, but consider say the German navy, the Kriegsmarine - literally, it translates to something like "sea war" than navy.


As for what spaceships end up getting called... well who is in charge when the first ships are being built, and what's their experience? By that I mean that you could, I think, argue that even a USAF-created space force might go with names like "cruiser" and "battleship" if the guy(s) behind the naming think it'll mean more public / congressional support. The "Re-Useable Mobile Laser Array" USS Ronald Reagan just doesn't have the same ring to it as the "Battleship" USS Ronald Reagan, however accurate and technically correct the former is. And even if you do end up with the former, it wouldn't surprise me if people *think* of it as a "battleship" or something similar.

Mangaka2170 said...

@ Cambias: Actually, several early German (and possibly British, too) aircraft classes were lifted from naval nomenclature. For example, a heavy fighter was sometimes referred to as a destroyer. It took the interwar period for the European powers to learn more about how aircraft needed to be fielded, and the terminology changed into roughly what we have today.

Milo said...

Cambias:

"In the name of realism (or at least creating the illusion of same) I would suspect the formal names for combat spacecraft types would be unrelated to navy or air force terminology -- if only to avoid confusion during the era when they are all part of one super-battlefield."

By the time you have space opera levels of travel, ground forces will be increasingly marginalized. MAD already makes intraplanetary wars hard, and in an interplanetary war surface ships are useless without spaceships strong enough to carry and land them. Surface ships will still be used, but they may well come to be thought of more as organic support attachments to infantry regiments (like some airplanes and helicopters are today) rather than deserving top billing as a separate branch (like other airplanes and helicopters are today).

The most important type of seagoing vessel in a space war are surface-to-orbit submarines, which would be more a part of your spaceguard than your navy.



Teleros:

"having autumn instead of fall"

...In space? What?

Apropos of nothing, I never realized that "autumn" is more "British", but I've always preferred it because I've disliked how "fall" (if you use it) is the only season word that also has another more common meaning.


"The "Re-Useable Mobile Laser Array" USS Ronald Reagan just doesn't have the same ring to it as the "Battleship" USS Ronald Reagan, however accurate and technically correct the former is. And even if you do end up with the former, it wouldn't surprise me if people *think* of it as a "battleship" or something similar."

Wasn't it said in another thread that the term "battlecruiser" was first made up by the press, and later picked up by actual militaries? The same could happen in space.

In any case, genre savviness is often overlooked in stories. Most people who are likely to at some point design a space war platform will have read some science fiction stories, and so will let those influence their choices. They might deliberately use terms that they've grown to be used to as children, or they might deliberately avoid them because they see them as too reminiscent of "unrealistic" portrayals of space travel, and they want to distance themselves from those to emphasize that they've built the real thing.

Thucydides said...

The Royal 22nd Regiment ("Van Doos") of the Canadian army indoctrinated me with the use of words related to the church as the French Canadian form of swearing.

"Chalice" or "Tabernac" in isolation are quite innocent (Chalice is just what you imagine, and Tabernac is short for Tabernacle, or Church), but when a young French Canadian private closes the machine gun feed cover on his hand you suddenly realize that occasion, tone and inflection give words like this entirely new meanings....

I suspect that if the Space Force follows the progression of starting as an extension of the Air Force then there will be a use of Air Force terminology (including using Air Force ranks) for equipment. This is already a historical fact; Bell had a contract in the 1950's to study the possibility of building a RoBo (Rocket Bomber). Unlike R.C., I would expect tradition to continue even after the hardware and doctrine is in place to make long cruises possible and spacecraft are more like classical ships than Common Aero Vehicles. A vehicle that carries weapons to the target would still be a Strike Craft even if it weighs thousands of tons as opposed to the 500 kilogram ancestral unmanned vehicle.

Rick said...

Speaking of minor language ploys, I've considered using 'kilometre' and 'kilogramme,' etc., in SF aimed primarily at a US audience, specifically to give metric units less of a science-tech flavor.

The political dimension of type names is important, and can go either way. The USN decided around 1970 that jumbo destroyer types should be 'cruisers' rather than 'frigates' in order to upgrade them in public perception to match Soviet cruisers.

On the other hand, I understand that European navies have adopted 'through deck cruisers' as sounding more modest than 'aircraft carriers.'

And in an earlier era the USN found it was easier to get appropriations for 'battlecruisers' than for 'battleships,' which sounded more imperialistic for some reason. At least in this case the resulting ships fit the (rather loose) definition of battlecruisers.

Cityside said...
This comment has been removed by the author.
Cityside said...

"the range of types that have at some time or another been called a frigate"

Yep. And it's worth noting that "Frigate" and "Corvette" had both fallen from usage for roughly fifty years prior to being revived in the 1940s.

Sloop, curiously, survived the shift in nomenclature and the RN commissioned sloops continuously until the 1940s until they were superseded by "Frigates" and "Corvettes" - which, in a bizarre twist, cheaper (and in the case of a Corvette,smaller) knock offs of a sloop, designed to fill the same niche:

http://en.wikipedia.org/wiki/Black_Swan_class_sloop

http://en.wikipedia.org/wiki/River_class_frigate

http://en.wikipedia.org/wiki/Flower_class_corvette

Anonymous said...

Battlecruisers, as far as I can tell, came in two flavors: heavy or armored cruisers that had larger caliber guns or battleships that had lighter armor; either crusiers with battleship guns or faster, lighter battleships. Unless you have a heavy combat weapons module matted to a huge propulsion module, I don't see 'battleship' or 'battlecruiser' being official nomanclature...unofficial is a whole different story though.

Ferrell

Milo said...

Rick:

"And in an earlier era the USN found it was easier to get appropriations for 'battlecruisers' than for 'battleships', which sounded more imperialistic for some reason."

A battlecruiser kills you and then cruises back home, a battleship kills you and then stays to have its way with your women?

Lentulus said...

It's hard to evade 20th century thinking even about purely naval terminologies. For example, while we are inclined to think of battleships as larger than cruisers, the US battleship Texas of 1892 was actually smaller than the Spanish cruiser Vizcaya. And the Soviet/Russian cruiser class Slava is roughly the same displacement as the US Alrleigh Burke class destroyers.

And don't get me started about Marines.

I expect that by the time (if ever) that military forces are needed in Space a new set of traditions will already have developed; doubtless informed by older military traditions but also probably (like RAF officer ranks before WWII) deliberately establishing new ones. And if we want to speculate on what they might be, it would probably be a good idea to look more closely at Chinese than European examples.

Scott said...

'North' in space is pretty easy to define, 'towards the local star'.

I actually meant calling a bearing to a contact as 'three-zero-zero, mark three-zero-zero' for a target that is on a line 60deg left of the line to the local star, and 60deg below the the line to the local star.

I also forgot to add that there's also the difference between relative bearings, true bearings, and angle on the bow.

My crack about mils is directed at Michael Z. Williamson, who had the idea that mils are one thousandth of the circumference, not roughly equal to one miliradian.

Scott said...

An interior communications manual is something that describes how pieces of equipment are spoken. For example, the high pressure air compressors are spoken 'hi-pac', the emergency diesel generator is spoken 'diesel', etc. It's really a reverse dictionary of technical terms, that has the precise name of something and then the common name to be used.

This includes terms like 'baffles', 'waffles', and the all-important difference between 'torpedo' and 'ownship's unit'. A torpedo is something coming towards you, while an ownship's unit is what you shot at someone.

The manuals themselves aren't classified that I remember, but I haven't had any luck finding a copy of one, either.

Milo said...

Scott:

"'North' in space is pretty easy to define, 'towards the local star'."

(A) I didn't mean that kind of north. I meant an analogue of the north pole as seen in spherical coordinates, with regard to the asymmetry in your two bearing coordinates (again, analogues of latitude and longitude). This is something that will, per definition, rotate as your spaceship rotates. I think I'm not explaining myself very well, sorry...

(B) Even when you are talking about that kind of north, no, that's incorrect. In addition to your distance from the star, you will need two other spatial coordinates. You will need some way to tell those two apart. This would most likely follow the plane of the ecliptic, with "west" being the direction of orbits/rotations, and "north" being perpendicular to the ecliptic, defined such that "west" is seen as counterclockwise when viewed from "north". (Also note that you could reasonably use either spherical or cylindrical coordinates. Most action will take place in a narrow region surrounding the ecliptic - at least on a strategic scale. On a tactical scale, however, equal movement in all three dimensions will be fair game.)


"I also forgot to add that there's also the difference between relative bearings, true bearings, and angle on the bow."

I have been talking specifically about relative bearings. Sorry for not making that clear.

Thucydides said...

Fooling politicians is indeed a great way to designate ship classes.

The Royal Navy commissioned "Through Deck Cruisers" as a means of getting aircraft carriers into service at a time when Parliament was not inclined to support such a move (indeed the last real aircraft carriers belonging to the RN had been sold off or decommissioned in the recent past when "Through Deck Cruisers" were sent to appropriations).

The Japanese Self Defense Force has recently commissioned two rather lovely "Helicopter Destroyers" which to non expert eyes appear to be 18.000 ton aircraft carriers with a compliment of a dozen helicopters; see the Hyūga-class helicopter destroyer at http://en.wikipedia.org/wiki/Hyūga_class_helicopter_destroyer

As noted, Soviet ship classifications were revamped in the West to make the Soviet Navy seem much more formidable when the US Navy needed to procure more ships (bumping a class to "Cruiser" means the Soviet navy now has lots more "Cruisers" on the books).

I suspect that this sort of thing fools people who are either not well versed in the technologies or histories involved, or who generally are not coming to the yards to see what is really going on. When they show up to the launch ceremony it is a bit late to stand there and say "wait a minute".

jollyreaper said...

Like a sword but awesomer!

http://imgs.xkcd.com/comics/fiction_rule_of_thumb.png

Probability book is any good vs. number of made up words, quality dropping with increase in words.

"Except for anything by Lewis Carroll or Tolkien, you get five made-up words per story. I'm looking at you, Anathem."

jollyreaper said...

My impression with today's "acronym culture" is that it suggests people have more types of things than they can come up with creative names for, and so they can't be bothered trying and just slap on a matter-of-fact description.

German does the same thing but the words can sometimes appear more natural. National Socialists abbreviated as Nazi, Secret State Police goes to Geheime Staatspolizei to Gestapo. The longest word in common use in German is Donaudampfschiffahrtsgesellschaftskapitän, which in English becomes four words: "Danube steamship company captain."

One thing done in post-apocalypse fiction is to give more natural names to Cold War acronyms that only become clear by the end of the story. The great nations of Nato and Ussar strove against each other and employed their mighty weapons such as the Eye of Sabeem which rested in the backs of great leviathan monsters beneath the waves or in special tombs in the Glassy Plains.

If I recall correctly we had a lot of that in Canticle for Leibowitz with Strontium being one of the bowls of wrath poured on the land, Fallout being a demon of hell, and the misborn being a curse sent from Satan. I think there were other pieces of Cold War terminology sprinkled in as well.

jollyreaper said...


This sounds odd to me - I associate the term "barge" with something that can carry a lot of cargo cheaply, at the expense of being slow. (Hence my primarily using it in a space environment to refer to Hohmann-trajectory cargo ships.)


I also tend to associate it with an unpowered vessel towed or pushed by another vessel but there are such thing as powered barges, just typically built along awkward barge lines.

A scow is a flat-bottomed boat with a blunt bow used for hauling cargo but I bet everyone reading this puts "garbage" right at the front. Well, every fleet needs a trash hauler but why does a scow get all the bad press? Nobody could use the term today.

Star Wars made use of archaic names. They had Star Galleons and Carrack Cruisers. They loved ramming names together. I always thought the name "Star Destroyer" was meant to poetically indicate the level of firepower like "Mountain Remover." So as a kid I wondered why the Death Star was such a big deal when they could already blow up stars. It was only later that I realized the name was meant more like "Star Frigate, Star Cruiser, Star Destroyer, Star Battleship, Star Fighter," etc.

Lentulus said...

Of course, nothing points out the fruitlessness of speculating on the names of things better than Dreadnaught. Nineteenth century armchair admirals certainly speculated on All-big-gun battleships; no-one would have predicted that the first one actually built would not only break the mold for battleship design but have a catchy name. I bet if it had been called "George III" or some such Dreadnaught would never have entered common usage.

Or for that matter how about "tank"? Lots of people expected Land Battleships but no-one could have predicted the name.

Teleros said...

Milo: "(A) I didn't mean that kind of north. I meant an analogue of the north pole as seen in spherical coordinates, with regard to the asymmetry in your two bearing coordinates (again, analogues of latitude and longitude). This is something that will, per definition, rotate as your spaceship rotates. I think I'm not explaining myself very well, sorry..."

Slightly different topic, but the idea I had was that you'd have N/S/W/E being arbitrary directions along the ecliptic, with "altior" and "demitto" (yay for Google Latin translations), or "alt" and "dem" being above and below the ecliptic respectively. The use of different initial letters means you won't ever be confused by a "SW" bearing, for example.

Milo said...

Lentulus:

"Of course, nothing points out the fruitlessness of speculating on the names of things better than Dreadnaught. Nineteenth century armchair admirals certainly speculated on All-big-gun battleships; no-one would have predicted that the first one actually built would not only break the mold for battleship design but have a catchy name."

In space opera, "dreadnought" is commonly appropiated to refer not to a specific weapon loudout on battleships, but rather to an extra-large class of battleships. So the capital ship size range goes something like destroyer, cruiser, battlecruiser, battleship, dreadnought. (Spacecraft smaller than capital ships are often not thought of as "ships" because people like space fighters too much...)


"Slightly different topic, but the idea I had was that you'd have N/S/W/E being arbitrary directions along the ecliptic,"

Are you working in Cartesian coordinates or spherical/cylindrical coordinates?

Although Cartesian coordinates are easier on the math, the important and obvious central location of the sun encourages polar coordinates.

If I were using Cartesian coordinates, then I would do as you say, define two arbitrary axes along the ecliptic (I could call them "north-south" and "east-west" or something else, but it doesn't matter) and have the third axis be perpendicular to it.

In spherical coordinates, I would make it so the ecliptic is aligned as the coordinate system's "equator". In cylindrical coordinates, I would put the cylinder's axis perpendicular to the ecliptic.

(Also I understand that you're talking about absolute bearing now. Let's be clear about this.)

Milo said...

Oops, second comment should have been directed at Teleros!

Lentulus said...

but rather to an extra-large class of battleships

Exactly! The name has lost all sense of the real importance of the Dreadnaught, and has just degraded to a "cool name" instead -- and a Space Opera tradition that will be no more useful for predicting what military spacecraft will really be called than the Land Battleship of late-19th-century fiction was in predicting the name of the tank.

Mangaka2170 said...

Another thought on 3D coordinates: Like how aircraft pilots use clock references when calling out object locations relative to them, what about spacecraft crew using constellations?

Think about it: spacers would most likely already be expected to pick out major stars and constellations as astrogation references, so it wouldn't be too much of a stretch to refer to objects coming from Orion or Sagittarius.

Of course, this wouldn't work very well in an operatic setting, this is a more PMF or PDF thing.

jollyreaper said...


In space opera, "dreadnought" is commonly appropiated to refer not to a specific weapon loudout on battleships, but rather to an extra-large class of battleships. So the capital ship size range goes something like destroyer, cruiser, battlecruiser, battleship, dreadnought. (Spacecraft smaller than capital ships are often not thought of as "ships" because people like space fighters too much...)


Possibly the best way to avoid this is to just call a combat starship a warship and be done with it unless the situation calls for getting into tactical nuances. Lots of times people will use specific terms like battlecruiser which raises all sorts of questions when warship gets the point across.

I think we can safely lay to rest the whole idea of reusing modern day nomenclature because it all entirely depends upon the postulated tech in the story. Figure that out and then we can figure out where naming conventions will go from there. We may well reuse a modern term in an odd way. Dune, for example, called surface-to-orbit combat craft "frigates" when most of us would think of them as FTL-capable warships not carried into combat by a mothership (or heighliner in this case.)

I still shake my head at the nonsensical Lost Fleet combat where destroyers and other light ships were placed in the wall of battle against other fleet formations. Completely misses the point of them! All they do is die in large numbers when two fleets clash. The proper battleships in that setting should be there to take the hits and the only justification for lighter craft would be as economical force projection. You have a target across the system that's not too heavy? Send a destroyer squadron. They're small, very fast, and are less expensive than big ol' battleships. You have enemy stragglers limping away after the big formations clash? Send the destroyers out from the fleet to finish them off while keeping your main strength intact.

I'll tell you what, the Lost Fleet is a good read if only for starting an excellent discussion about how he got everything wrong. A teachable moment. :)

Anonymous said...

jollyreaper said:"I think we can safely lay to rest the whole idea of reusing modern day nomenclature because it all entirely depends upon the postulated tech in the story. Figure that out and then we can figure out where naming conventions will go from there. We may well reuse a modern term in an odd way. Dune, for example, called surface-to-orbit combat craft "frigates" when most of us would think of them as FTL-capable warships not carried into combat by a mothership (or heighliner in this case.)"
Yeah, an interceptor could be the size of the biggist ship class; the only difference would be the delta-V and crew endurance.

Ferrell

Thucydides said...

The most laughable part about using "Dreadnought" in Space Opera is how quickly the real HMS Dreadnought became obsolete.

A comparison between the real Dreadnought and the monster ships that sailed to Jutland a few years later is pretty illustative, but do space operas have tech revolutions similar to the introduction of reliable submarines, aircraft carriers or guided missile armed warships? What sweeps Dreadnoughts from the skys?

Milo said...

Jollyreaper:

"Lots of times people will use specific terms like battlecruiser which raises all sorts of questions when warship gets the point across."

If you're telling a war story, with fleets of ships battling each other, then you're going to need some idea of which are the largest and smallest ships in the fleet, which roles they play, etc.

If you have only a single friendly warship in the story, then congratulations! It's a cruiser per definition.

If you're a civilian, then "a warship is pointing its guns at us and telling us to do stuff" is enough to get a grasp of the situation. But if military affairs are central to the story, then readers are going to want to know if the heroes' ship is the grand flagship of the imperial armada, or the gutsy underdog.


"Dune, for example, called surface-to-orbit combat craft "frigates""

Surface-to-orbit would mean that atmospheric operation is vital to their mission, which means that for once you could actually justify calling them (and making them function as) fighters.


"You have a target across the system that's not too heavy? Send a destroyer squadron. They're small, very fast, and are less expensive than big ol' battleships."

Cruisers are for force projection. Destroyers, while sometimes useful for that, were originally all about staying with the main fleet and protecting battleships from stuff that they weren't good against on their own (first torpedo boats, later submarines and airplanes).

But I do agree that destroyers would try to avoid directly engaging enemy battleships if they can help it.

Hugh said...

My charitable explanation for using 20th C naval terminology for space warships of the future is that it's a translation from the original into something understandable to the current audience. After all, we generally don't read Shakespeare in the original either, most editions have modernised spelling and updated words.

Some science fiction authors manage to come up with convincing future lingo. Anthony Burgess did a fabulous (but non-military) job with the future slang in Clockwork Orange. And CJ Cherryh avoids a lot of naval cliches in her Alliance/Union books, for instance rider ships instead of fighters. But IIRC both had done fairly extensive studies of linguistics.

Cityside said...

"A comparison between the real Dreadnought and the monster ships that sailed to Jutland a few years later is pretty illustative"

Dreadnought missed Jutland due to a refit, but the six Bellerophon and St. Vincent class ships, essentially repeats of the Dreadnought design, all fought at Jutland.

Dreadnought as a term was useful during WWI to differentiate between ships of that type and the older, decidedly less capable "pre-dreadnought" battleships (six of which also fought at Jutland, despite their relative weakness). Bot types were technically "battleships" but the dreadnaughts had at least twice the heavy guns, around a 3 knot speed advantage and often more effective armor. Dreadnaught largely fell out of usage with the scraping of the pre-dreadnaughts, btw (one rarely comes across it in in WWII)

Scott said...

Another example of 'unusual' names for spacecraft comes from the anime 'Royal Space Force: Wings of Honneamise'. Their first orbiter is called a space battleship (mostly to convince the politicians to cough up enough money!)

Here's an example of the kind of technical report you need to think about: detection of a new contact somewhere.

Now, in the US Navy, you need to state type, bearing, and 'angle on the bow' (where the contact is heading). For example, "Offsa'deck, I hold a new contact! fishing boat, bearing three-zero-zero, angle on the bow port forty-five."

To translate: 'Officer of the Deck, I see a fishing boat to the WNW of us, and his port side is facing us at about a 45 degree angle.'

With spaceships, you'd have bearing, angle on the bow, and vector direction, all in 3d.

Angle on the bow would refer to the ship's facing, as opposed to it's velocity vector. You'd care about that because the ship's facing would determine what changes it could make to it's velocity vector.

I would expect the SF contact report to sound like "emcee, I have a new contact. slowboat bearing three-zero-zero mark zero-six-zero, angle on the bow one-thirty-five mark zero, vector one-thirty-five mark zero. He's just executed turnover."

Let's break that down. 'Emcee' is MC, mission commander. A slowboat would probably be a nuke-electric, due to their low thrust. 135 mark 0 means that their hull is in the same plane as the ecliptic, but their stern is pointing 45* away from you. The vector is the same, so their engine is facing toward their direction of motion. And, obviously, 'just executed turnover' means they're on a brachistochrone if they're thrusting.

Then you'd have follow-up reports to give more information about that ship, like what the thrust is, any unusual hot spots, etc.

Questions?

Brian/neutrino78x said...

Hey Matt, you are my fellow submarine vet aren't you? Silver dolphins as well, or are yours gold? Mine are silver. :-) I was on USS Florida (ssbn-728) and USS Asheville (SSN-758) although I did not go underway on the asheville, since that boat was in drydock the whole time. I did do four "strategic deterrent patrols" on the Florida, a total of 365 days at sea according to my DD-214. ;-)That's just because I was only in four years...more than enough for me! Don't worry, my dolphins are not "low sodium", I did do one patrol before I got them. :)

I think you are thinking of the vaunted "Doctrine for Submarine Interior Communications". I agree that, AFAIK, it is unclassified, but I have not been able to find it online either. ;-)

I did find this reference to it "COMNAVSUBFOR INST 2305.1 W/CH-1, DOCTRINE FOR SUBMARINE INTERIOR COMMUNICATIONS"

http://www.navybmr.com/ETV/PO3.html

One thing I always find amusing: you know how we are not supposed to say "niner" on submarines (per aforementioned manual)? Well, if you read the official Navy instruction for topside watch, which is applicable to both submarines and surface ships, it clearly states that "9" is to be spoken "niner" on the sound powered phone. Sometimes I would deliberately say "niner" as a result, just to [f-word] with people I found annoying, and some chief would be all "WHAT THE [f-word] DID YOU SAY???" and when I repeated myself I made sure to say "nine" rather than "niner". lmao...

--Brian

Scott said...

Yep, I'm the other bubblehead around here. Georgia GOLD and Kentucky GOLD, for me. And whoever called ossifer fish 'gold' needs his head examined. They're BRASS.

Enlisted fish are actually silver, if you find WW2 fish. I'm trying to see what it would take to be able to make real silver fish for Mess Dress use.

jollyreaper said...

A comparison between the real Dreadnought and the monster ships that sailed to Jutland a few years later is pretty illustative, but do space operas have tech revolutions similar to the introduction of reliable submarines, aircraft carriers or guided missile armed warships? What sweeps Dreadnoughts from the skys?

This brings us back to the accellerando/decellerando debate. My preference is for relatively stable tech in space so that the warships of an old man's day are not all that dissimilar to the ones in the fleet of his grandson. This preference is mainly to keep the setting from disappearing up it's own bottom with impossible-to-guess speculation about the future.

My other preference is to avoid making the tech and tactics too similar to the modern day so it won't feel dated in 30 years. Case in point, Battlestar Galactica lifting American carrier ops straight down to the jargon. Or look at the new Star Trek movie where it looks like they have a black gang belowdecks shoveling antimatter into the reactor.

jollyreaper said...

I don't think that we'll see the worst zeerust in the depiction of the ships and weapons but in the control rooms. My feeling is in real life the biggest changes between our modern destroyers and whatever we use in 50 years is not going to be in external appearance and weapons but in what the CIC looks like. My big prediction is that railguns and drones are going to provide the potential for future destroyers to become one-ship task forces. Railgun rounds are supposed to hit like a cruise missile at 200 miles range while being cheap enough to fire off like dumb artillery rounds. Guesstimates I've heard is $50k a shot versus $1.5 mil or more per missile with destroyers carrying a lot more railgun rounds than missile reloads.

The drones will allow similar levels of coverage to modern aircraft carriers with far less expense. You'll still have force projection, surveillance, strike power, but for less cost.

My guess is in 50 years we'll still have supercarriers out there flying a mix of manned and unmanned aircraft but the drone ships will be acting like jeep carriers from WWII, serving in a pinch in roles that don't warrant a fleet carrier.

Now what we may see is that the railgun destroyer will only operate several drones, possibly with the main role being as spotters for the railguns while purpose-built drone carriers specifically take that jeep carrier role. I think there's still going to be a lot of personnel involved in maintenance on the drones. They're complicated machines, just like manned jets, and a carrier wing is a real maintenance pig.

Then again, in the 70's and 80's navies were using destroyers for anti-ship and air defense with frigates specializing more in ASW and now they're trying to have more balanced designs like the Arleigh Burke doing it all. There's every chance that there's a truly optimum decision to be made here and the navies in question will do just the opposite for political reasons.

jollyreaper said...


Cruisers are for force projection. Destroyers, while sometimes useful for that, were originally all about staying with the main fleet and protecting battleships from stuff that they weren't good against on their own (first torpedo boats, later submarines and airplanes).

But I do agree that destroyers would try to avoid directly engaging enemy battleships if they can help it.


Yeah. I think my personal answer to this is to completely avoid repurposing modern terms so there's no baggage to be carried. If we're going with the constellation idea, ships moving like shoals of fish, the biggest, heaviest ones are out on the wall and will be ships of the wall. The whole fleet moves slowly and the smaller ships are faster and detached for specific purposes. So the ships that charge out from the wall to intercept could be called interceptors or chargers, light ships meant for scouting and harrying could be called skirmishers or raiders, etc. It all depends on how tech works in your setting.

Lentulus said...

A "Naval Nomenclature" story. I live in Halifax, NS. Back in the 80's one of the local photo studios acquired a set of turn-of-the-century plates including pictures of various ships making courtesy calls in Halifax harbor. One I particularly remember was of one of those pretty German pre-dreadnaught battleships wit hemispherical turrets.

Anyway, the photographer duly printed them and sold them carefully labelled as to what they were: German Destroyers. I don't think he knew there were other kinds of warships.

Lentulus said...

jolleyreaper:

completely avoid repurposing modern terms so there's no baggage to be carried

In the perfect War In Space SF novel, not only would each side have a completely distinct nomenclature matching its doctrine, but each would be constantly trying to fit the other side's ships into the same nomenclature and doctrine it uses itself because that is the One True Doctrine. And civilians would still call them all destroyers.

Tony said...

The really honest space warfare novel has technological and doctrinal convergence between enemies over time. The Starfire novels by White and Weber do a pretty good job with this, even though they can be pretty corny in every other sense.

WRT nomenclature in general, I think that anything the writer can justify is just fine. I personally have a hard time with gratuitously invented nomenclature, because it adds a layer of work for the reader that serves no rela purpose other than the author showing off that he did his homework.

If you just have to use an invented nomenclature, make it more subtle in presentation. For example, Poul Anderson, in Star Fox and Fire Time had a space warship taxonomy of:

Lancer
Cruiser
Blastship

Lancers were, as the name suggests, smaller, faster, lightly armed ships.

Cruisers were general purpose. If ships were operating alone, they were likely to be cruisers. Yet cruisers showed up in almost every scene where there were space warships, suggesting they were the base type.

Blastships were, as the name suggests, ships designed to blow up anything that got in their way.

There were also fighter carriers, but the fighters they carried were for low orbit and atmosphere work.

The interesting thing about all of this is that Anderson didn't take time off to infodump about the various ship classes. He just treated them as ship classes would be treated in a book about a 20th Century war. Lancers were ommands for bold young officers, making them a destroyer equivalent in the mind of the 20th Century reader. Blastships were the territory of flag rank officers, making them large and prestigious, like battleships. Cruisers were the commands of officers of Captain rank, good for fleet actions, patrolling, etc.

jollyreaper said...


The interesting thing about all of this is that Anderson didn't take time off to infodump about the various ship classes. He just treated them as ship classes would be treated in a book about a 20th Century war. Lancers were ommands for bold young officers, making them a destroyer equivalent in the mind of the 20th Century reader. Blastships were the territory of flag rank officers, making them large and prestigious, like battleships. Cruisers were the commands of officers of Captain rank, good for fleet actions, patrolling, etc.


So long as the author works it out in advance, the particulars should be delivered in natural dribs and drabs of exposition. Nothing has to be said when Captain Awesome of the lancer Fleetfoot comes into orbit. You know it's a type of ship and he's a young, dashing officer. Then when the fleet goes into battle and Captain Awesome's ship and several other lancers are sent out to harry the enemy, the role is naturally perceived. Then there could be a line saying "the lancer, lacking the heavy batteries of the laserstars, used antimatter torpedoes for their main punch. Small and fast, they could chase targets of opportunity and hit them as hard as any laserstar. But they paid for that speed with poor staying power. The Ferret strayed too close to the fleeing Eviltarian flagship and took a bolt from the main battery amidships. The blast tore her in two without a moment's hesitation. Captain Awesome knew his ship could be gone just as easily. He also knew the admiral would risk losing a dozen more lancers if it meant taking out that flagship. They had to destroy it before it could make warp."

And likewise the tactical situation could be explained at the right moment. "The planet was at his back, the lagrange point before him, and the enemy fleet before him. FTL jumps could only be made in areas of low competing gravity. That meant either a hundred AU's out from the primary or from within the gravity well at lagrange points. L1, L4, and L5 were the closest ones. Going for 4 or 5 would avoid a direct pass at the fleet but leave him exposed to attack for a greater period of time. The quickest way out was through the enemy fleet, through them to L1. A hotter fire but one passed through sooner. It was a gamble but with his fleet's reserves so low, one good fight was all they had left in them. Admiral Badguy was counting on this, he knew."

jollyreaper said...

Oh, and for common terminology in navigating, I like using sunward to refer to things in the direction of the primary, starward for anything away from it, upwell and downwell for positioning within gravity wells, and the established prograde and retrograde to describe orbits.

Anonymous said...

(SA Phil)

Based ont he Purple vs Green discussions - it could be there would only be two ship types.

So Missile Ship and Laser Ship or some variant might be the terminology

Tony said...

SA Phil:

"Based ont he Purple vs Green discussions - it could be there would only be two ship types.

So Missile Ship and Laser Ship or some variant might be the terminology"


I'm still not convinced that lasers would ever be primary armament. Kinetics and explosives are just too flexible. But I could see the taxonomy being something simplistic, like "large missile ship", "missile ship", and "small missile ship".

Anonymous said...

(SA Phil)

It really depends on how effective the Laser Star is.

If its really a light second range kill incoming missiles every second deal, you might not ever even see Missile ships.

Then you might just be ranking Laser Star sizes.

----
If Laser Stars arent effective - then you might have a lot of classes - Since at that point rail guns and light gas guns might have some value instead of just missiles. (since they would have a high relative speed to the missiles)

Tony said...

SA Phil:

"It really depends on how effective the Laser Star is.

...

Then you might just be ranking Laser Star sizes."


Like I said, I'm just not buying it. YMMV.

"If Laser Stars arent effective - then you might have a lot of classes - Since at that point rail guns and light gas guns might have some value instead of just missiles. (since they would have a high relative speed to the missiles)"

But they can only fire dumb projectiles or smart projectiles with very little ability to maneuver. The ability of missiles or even guided bombs to home on a target makes them much superior to guns at long range, and thus the obvious candidates for primary weapons.

Jim Baerg said...

Jollyreaper: "established prograde and retrograde to describe orbits"

That works well enough in any solar system like our own with most planets moving in the same direction in nearly the same plane. However, some of the newly found planetary systems have major planets in orbits *highly* tilted to one another.

jollyreaper said...

: "established prograde and retrograde to describe orbits"

That works well enough in any solar system like our own with most planets moving in the same direction in nearly the same plane. However, some of the newly found planetary systems have major planets in orbits *highly* tilted to one another.


Simple enough. I believe prograde is counterclockwise if one is looking down on the solar system from the north pole of the sun? Then we look at any system and prograde is counterclockwise and retrograde is clockwise.

Milo said...

Scott:

"You'd care about that because the ship's facing would determine what changes it could make to it's velocity vector."

Spaceships can have thrusters on more than one side, though, and for the exhaust-velocity-optimized engines most ships will use, the time taken to spin the ship around is fairly small compared to the time needed to produce significant thrust.

More important is looking at what direction the ship is travelling in, what direction it's thrusting in, and what (based on analysis of its behavior and on its publicly stated mission plan) it appears to be aiming for.

In combat, where the enemy's weapons are pointed will be more important than where the engines are pointed.



Jollyreaper:

"My preference is for relatively stable tech in space so that the warships of an old man's day are not all that dissimilar to the ones in the fleet of his grandson."

I understand the appeal of this for space opera, but I don't think it's terribly realistic - unless travel and communication are difficult enough that new technology has difficulty propagating (and even then you will have hotspots of high technology).


"This preference is mainly to keep the setting from disappearing up it's own bottom with impossible-to-guess speculation about the future."

Perfectly understandable. Making up one future is hard enough. Making up several...



SA Phil:

"Based ont he Purple vs Green discussions - it could be there would only be two ship types.

So Missile Ship and Laser Ship or some variant might be the terminology"


Why missile ships and laser ships? If missiles and lasers are both worth using, then it will be likely that many ships will be armed with a mixture of both, unless they are optimal for widely differing tactical situations (in which case ship classes would be distinguished primarily by what environments they're designed for, and only secondarily by what design choices they've made to this end). Meanwhile ships armed with the same general class of weapon can still have widely differing characteristics - overall size, weapon caliber, engine power, whatever.



Jim Baerg:

"That works well enough in any solar system like our own with most planets moving in the same direction in nearly the same plane."

This seems to be the norm, and I suspect it'll be doubly so for systems similar enough to home to be habitable. There will be pathological cases, but the terminology is still applicable widely enough to be worth using.


"However, some of the newly found planetary systems have major planets in orbits *highly* tilted to one another."

How many?

There has also been the opposite discovery - an unexpectly large number of systems where planets have orbits with even more precisely aligned orbital planes than our solar system.

Tony said...

I think the tactical coordinate system would simply be orbital elements WRT either a planet or a star. We already have defined coordinate systems for that, in use every day by real world military forces. For spacecraft under constant power, apparent changes in or bital elements would give a numeric basis on which to predict future position. But neither the orbital nor the acceleration data would be simple enough to include in a verbal contact report.

Teleros said...

Milo: "Are you working in Cartesian coordinates or spherical/cylindrical coordinates?"

It was really more a case of coming up with something a bit different from "they're above us" etc - useful more for rough and ready directions than precision. For precision I'd just use X,Y,Z co-ordinates centred upon a suitable body (Sag A* for the galaxy, a star for a solar system, etc).

jollyreaper said...

I think the tactical coordinate system would simply be orbital elements WRT either a planet or a star. We already have defined coordinate systems for that, in use every day by real world military forces. For spacecraft under constant power, apparent changes in or bital elements would give a numeric basis on which to predict future position. But neither the orbital nor the acceleration data would be simple enough to include in a verbal contact report.

And this is where we go mad trying to imagine future display tech.

Wiki has some great info on how it was done in the past.

http://en.wikipedia.org/wiki/Operations_room

Check out those external links, tons of WWII stuff.

The basic function of the CIC doesn't change much, just how much computers can streamline the process. Check out the floorplan there for an Aegis system. I remember film from the RAF control center during the Battle of Britain where people with poles moved around icons on a physical map representing the bomber formations. Those people have been replaced with computer displays now, of course.

Right now orders are still relayed by voice. How much more quickly can it go if dragging a mouse to select and clicking a target works as readily in the real world as it does in a video game? How much can we count on sensor fusion reducing the workload? Hard to say.

My guess is that whatever the advances, we're generally going to have the high mucky-mucks sitting there looking at a big screen while some number of techies are sitting at workstations massaging the data that goes into the big screen. If someone calls out the appearance of a new target, I think it would be just to say "Contact sighted, launching from planet" or "dropped from warp/FTL/hyperspace" and the commander can look for himself to see range and bearing. Aside from being a mouthful to say, I'm not certain how relevant saying the location of a new contact would be when the commander can look at it on the plot.

Tony said...

I think with complex astrodynamic coordinates and computers for display generation, a verbal contact report would be very simple:

"New contact, designate C1004."

All concerned would look at their displays and find the contact. If they need precise data beyond the visual display, they can click on the contact and pull up a window with a detail display.

As for sensor integration, they actually had it to a pretty high level in WWII. After screwing arounf for a year or two figuring out who needed to know what in combat, they settled on a simple system of putting the men who needed to know on appropriate voice circuits as listeners. So, for example, the search radar would designate a contact in ship relative coordinates, the fire control directors would train on and acquire the target with their fire control radar, and the computer operators would start running the solution. The gun mounts were generallyi nstructed to "match up and go to auto", meaning they matched their deflection and elevation with whatever the fire control computer was generating and engaged their remote control clutches. The only other verbal orders passed to the gun mounts were what type of ammo to use and when to commence and cease firing.

The reason we even have a sensor integration issue these days is that all of the sensors' signal processors are now expected to talk to each other, rather than having humans communicate the data to other humans.

Anonymous said...

Jollyreaper

My guess is that whatever the advances, we're generally going to have the high mucky-mucks sitting there looking at a big screen while some number of techies are sitting at workstations massaging the data that goes into the big screen. If someone calls out the appearance of a new target, I think it would be just to say "Contact sighted, launching from planet" or "dropped from warp/FTL/hyperspace" and the commander can look for himself to see range and bearing. Aside from being a mouthful to say, I'm not certain how relevant saying the location of a new contact would be when the commander can look at it on the plot.

================

It may be a lot simpler to cut the techs out of the loop entirely.

Why have a computer which sends data to the tech so the tech can interpret it and move it along to the decision maker?

Just have the computer collect and masage/interpret the data.

Beyond that - computers make decisions faster than humans, so why bother with the human decision loop?

Humans might be reduced down to the "Query: Destroy this Enemy? Yes/No?" level.

I imagine we will get to that point *before* we have space warfare. And thus there won't even be a debate about what system to go with.

(SA Phil)

jollyreaper said...



Humans might be reduced down to the "Query: Destroy this Enemy? Yes/No?" level.

I imagine we will get to that point *before* we have space warfare. And thus there won't even be a debate about what system to go with.


Ultimately, that seems likely. But the system I was describing was anticipating a point somewhere between where we are now and "kill y/n?"

People still go back and forth as to how strong they think AI could prove. I'm taking the conservative guess that we'll have really smart expert systems and that there will be some decision-making between those systems and the main display. Or those techs are taking delegated orders from the big cheese.

Again, it all depends on the posited setting and tech level as for how much automation can be had. If you have trans-human AI's, the whole military-industrial complex can be automated and you're back to the question of what good baseline humans are for. But I think that would fall under far future, not mid-future. And I do think we'll have our answers for CIC automation in the near future on Earth in wet navies far before we're doing it with space navies.

Tony said...

The problem with automation is dealing with novelty in the environment. A programmer can only automate to the extent he has reliable data about the environment of the automaton. You keep humans in the loop to supervise, so when something unexpected or unprecedented occurs, the computer can be given a nudge in the right direction, or bypassed entirely.

Heck, even the very simple data processing automatons that I design and implement require human supervision the first several runs through, to make sure nothing was missed, and to catch incorrect assumptions about the working data. With the combat environment, I'd want human supervision all of the time, simply because something new is bound to turn up, if not by enemy intention than by sheer accident of never having thought of that during test design.

It may not be that the sensor and weapons operators are in the command loop most of the time, but they're being able to monitor their systems, detecting and correcting for unexpected conditions, may be the difference between winning and losing.

Anonymous said...

Having a human supervisor for a computer system would probably be best; the human could supervise several computers or display systems. The commander could decide that he works better with an emphisis on size-vs-power rather than accelaration-vs-rate-of-fire on the main display, so the sensor supervisor tweeks the display. The commander has his info, he didn't have to take the time to do it himself, and all other info is still there; (if he could do it by simply asking the computer, he would only need an assistant with him in the room).

It's gonna be a while before we can get to a couple of people in CIC run the whole show, if ever...

Ferrell

Anonymous said...

(SA Phil)

If we are still talking plausible mid-future in the 200-300 year range - then I think the computing power is being underestimated.

You will probably have the equivalent of dozens or hundreds of modern supercomputers on each space-craft.

As to AI - you wont have HAL AI for space combat. It would probably be more specialized. And probably not "AI" in the science fiction sense.

As to Tony's concern about verification, etc - I expect that will be done by computers as well.

*Computers will write the code
*Computers will check the code
*The entire system will be tested for thousands of hours in a mock environment - exploring millions of potential variables
*Computers will make all adjustments during that time.
*Computers will then run the spacecraft (or aircraft w/e) and fight the wars.
*Computers will report out the results to humans who are busy doing something else.

Heck ETAS had a system that would autocode Powertrain software over 10 years ago.

In 100 years I doubt humans will still be writing the actual code for anything. They will just be telling Computers what they want to happen, and the computers will code it.

A very plausible future is one where the only humans that are killed in a war are "collateral damage"

Lentulus said...

I expect that will be done by computers as well.

I hope you are right. I think we have pushed past the limit of the complexity of software that humans can develop and support.

As I return to a bunch of *bleep*ing obscure RMI errors in a far too complex worldwide system.

Tony said...

Re: SA Phil

Nobody, human or computer, can write procedures for unexpected events or environmental conditions. And no amount of testing in any kind of simulated environment, no matter how detailed, is going to find out how a system behaves under conditions the test designers didn't foresee. That's the purpose of human supervision in service.

WRT programatically generated code, it just happens I ran a code generator today, to produce some object-to-relational mapping code in .NET MVC. As soon as it was done, I had to go in and tweak a couple of things to match my company's software writing standards. Microsoft couldn't predict everything a user might want to do, so there code generators just write vanilla code that takes care of most of the tedious chores. Even for more configurable code writing systems, guess what? Some person has design and implement the configuration add-ons. Humans will never be out of the code writing business, simply because no one human or group of humans will ever be able to predict all the possible parameters of every code writing environment.

jollyreaper said...


The problem with automation is dealing with novelty in the environment. A programmer can only automate to the extent he has reliable data about the environment of the automaton. You keep humans in the loop to supervise, so when something unexpected or unprecedented occurs, the computer can be given a nudge in the right direction, or bypassed entirely.


My best guess is that the optimal answer for this will vary with the tech and the only way to figure out the most satisfactory balance will be heavy wargaming in the design phase and seeing where everything breaks down. Of course, there's always the potential that someone has a pet agenda here and will use politics rather than evidence to get their way. 2002 Millenium Challenge.

http://en.wikipedia.org/wiki/Millennium_Challenge_2002

Red, commanded by retired Marine Corps Lt. General Paul K. Van Riper, used old methods to evade Blue's sophisticated electronic surveillance network. Van Riper used motorcycle messengers to transmit orders to front-line troops and World War II light signals to launch airplanes without radio communications.

Red received an ultimatum from Blue, essentially a surrender document, demanding a response within 24 hours. Thus warned of Blue's approach, Red used a fleet of small boats to determine the position of Blue's fleet by the second day of the exercise. In a preemptive strike, Red launched a massive salvo of cruise missiles that overwhelmed the Blue forces' electronic sensors and destroyed sixteen warships. This included one aircraft carrier, ten cruisers and five of six amphibious ships. An equivalent success in a real conflict would have resulted in the deaths of over 20,000 service personnel. Soon after the cruise missile offensive, another significant portion of Blue's navy was "sunk" by an armada of small Red boats, which carried out both conventional and suicide attacks that capitalized on Blue's inability to detect them as well as expected.[1]

At this point, the exercise was suspended, Blue's ships were "re-floated", and the rules of engagement were changed; this was later justified by General Peter Pace as follows: "You kill me in the first day and I sit there for the next 13 days doing nothing, or you put me back to life and you get 13 more days' worth of experiment out of me. Which is a better way to do it?"[2] After the reset, both sides were ordered to follow predetermined plans of action, leading to allegations that the exercise was scripted and "$250 million was wasted".[3] Due to his concerns about the scripted nature of the new exercise, Van Riper resigned his position in the midst of the war game. Van Riper later expressed concern that the wargame's purpose had shifted to reinforce existing doctrine and notions of infallibility within the U.S. military rather than serve as a learning experience. He was quoted in the BBC–Discovery Channel documentary The Perfect War[4] as saying that what he saw in MC02 echoed the same view promoted by the Department of Defense under Robert McNamara before and during the Vietnam War, namely that the U.S. military could not and would not be defeated.


If you want a story where scrappy upstarts defeat the 800lb gorilla, hubris is certainly the way to go. If that gorilla's doing what he's supposed to, there's no freakin' way the upstarts survive.

jollyreaper said...

Some person has design and implement the configuration add-ons. Humans will never be out of the code writing business, simply because no one human or group of humans will ever be able to predict all the possible parameters of every code writing environment.


I would say that the ability to adapt to the unexpected as well as if not better than a human would be the dividing line between expert system and strong AI. Watson on Jeopardy, that was programmers using their skills to beat other humans. Deep Blue, that's programmers beating a chess player. When the machine truly thinks for itself and can tackle a problem from start to finish the same way the programmers and engineers would, that's strong AI.

As to whether or not we'll reach that, that's an argument for many beers. But from a storytelling perspective, that's the real dividing line.

Anonymous said...

(SA Phil)

What will be unexpected in Space Combat though?

There will be no stealth, propulsion, weapons technology, etc will all be known quantities.

There won't be ambushes, feints, slick maneuvers, and so on.

The "real" space war scenario would likely be pretty boring from a drama perspective.

Mangaka2170 said...

We've already established in other discussions that Q-ships are going to be very difficult to discern from regular freighters up until they accelerate suddenly or start shooting, and as the MC02 and America's recent escapades in the Middle East has taught us, a large, more technologically advanced force can (and probably will) be defeated by a smaller force with more primitive tactics that is prepared to fight dirty. Depending on the resolution of a ship's sensors, for example, you could probably get away with detonating a few modified anti-tank mines disguised as satellites to temporarily put an enemy warship out of action (at least, until repairs can be made).

The advantage of a human crew, in this situation, is that they'd probably be able to make the necessary repairs until they get to a cageworks. Not only would a drone have difficulty determining the source of attack (since there wouldn't be any confirmed hostile sensor tracks, it might even conclude a long-range attack or a remote missile strike from the other side of the planet), but it wouldn't be able to repair itself if it needed to, so unless your combat drones are cheap enough to be considered totally expendable (and that's probably beyond PMF), any military will do what they can to make sure that their investment isn't destroyed by the space equivalent to the IED road bomb (plus, imagine the PR coup a scrappy resistance force could have if they shot down a $100 million combat drone with a $100 bomb).

Thucydides said...

Extending the constellation idea, maybe ships will be designated by delta V. Large ships with the mass and firepower to man the outer wall will generally have low delta V while the smaller, less massive ships inside the globe will have higher delta V to allow for repositioning and pursuit.

Inverting the positions is also possible (fast ships on the outside, massive high firepower ships on the inside). depending on the tactics or strategy the owning Navy follows.

jollyreaper said...


Extending the constellation idea, maybe ships will be designated by delta V. Large ships with the mass and firepower to man the outer wall will generally have low delta V while the smaller, less massive ships inside the globe will have higher delta V to allow for repositioning and pursuit.

Inverting the positions is also possible (fast ships on the outside, massive high firepower ships on the inside). depending on the tactics or strategy the owning Navy follows.


That's an interesting point. The battles in the Lost Fleet (my favorite bad example) were all very straightforward since both fleets basically screamed and leaped at each other. A fleet defending itself would be quite content to keep all the lesser ships tucked in behind the wall, the heaviest ships with the heaviest shields and guns defending. And if the two constellations wanted to fight, they'll reach each other in their own good time. But if one constellation is fleeing and the other wants to force the fight... Small, high-v ships can chase down stragglers, can attack targets far away from the constellation's course of travel. But how would the speed help against a defended constellation? How would one constellation overtake another in a stern chase? Simple enough if there's a limited delta-v and the attacker is certain of gaining remass soon. They could just burn longer than the defending constellation and overtake it.

But what if the high-v ships are sent around the constellation? Sure, charging up the middle would see them run squarely into the rearguard made up of ships of the wall and they would be smashed. But what if they try harrying the constellation from multiple vectors? It can't be strong everywhere and nipping at the heels might potentially open up vulnerabilities.

Thucydides said...

I also think we are caught up with the idea that the name is the actual descriptor of the device. There was a long period of time the historians had no clear understanding of galley warfare past Classical Greek times because ships were designated by the numbers of rowers (tetreres (quadriremes) and penteres (quinqueremes)). The idea that ships had morphed into artillery platforms and carriers for large numbers of Marines in Hellenistic times didn't seem to exist until the 1980's, at least I don't recall any discussion prior to that. Most discussion in the pre 1980's period seemed to revolve around the mechanical nature of such ships and how you could have 4,5 or 6 banks of oars (with the implied presumption that ships still sought to ram each other and more oar power was needed to strike the killing blow).

Later era galleys also had something of a blind spot for historians. A Capitana was clearly a flagship, but what was a Lantern as a ship class? The answer turned out to be a galley that had extra armament fitted; the Lantern referred to the light hung from the back of the ship to designate it as a heavy weapons carrier.

How this translates into PMF ship designations will be a bit unclear. Maybe a "thermal" will designate an NTR powered ship with high thrust and low ISP for combat manouevres, while a "plasma" will be the high ISP carrier ships that take "thermals" to the combat zone.

Rick said...

I am liking this discussion, but don't have a lot to add to it at the moment!

Rick said...

Oh, but I should add, welcome to a couple of new/returning commenters!

Tony said...

SA Phil:

"What will be unexpected in Space Combat though?

There will be no stealth, propulsion, weapons technology, etc will all be known quantities.

There won't be ambushes, feints, slick maneuvers, and so on.

The "real" space war scenario would likely be pretty boring from a drama perspective."


Drama need not apply. Something as simple as an unrecognized EM signature or almost any parameter out of the expected range could cause the program to say, "What now, Boss?"

In fact, that's considered a software development best practice. Anything you know to write defensive code against, you write defensive code. Something turns up that you didn't expect, the system raises an exception that causes an email to be sent to software people that can fix it, or sends them a text on all of their mobile devices if it's a real emergency.

Of course, in combat you don't have time to write and test a patch. So you have humans in direct supervision in real time. They can analyze the raw data and make decisions that a program can't. After the battle, the system logs and databases are copied to the appropriate contractor with instructions to add or update appropriate event handlers for the next release.

Tony said...

Lentulus:

"I hope you are right. I think we have pushed past the limit of the complexity of software that humans can develop and support."

Not quite. Service-based architecture is creating whole new levels of complexity. The services themselves are easy enough to program and understand. (I've both published and consumed services, and I'm hardly the slickest programmer in any room.) But the services joined together in multiple ad hoc manners creates supersystems that only the owners really understand, and then only in a conceptual way. The actual technical details aren't even part of a set design. Developers just produce and consume services as they need them, each as a separate entity.

Milo said...

SA Phil:

"In 100 years I doubt humans will still be writing the actual code for anything. They will just be telling Computers what they want to happen, and the computers will code it."

To get a computer to do something for you, you need to be able to explain to the computer what you want. Explaining to a computer what you want is, per definition, coding.

Coding techniques may change, but nothing short of a strong AI will be able to produce reliable flight software from nothing more than a layman ordering "go make some flight software".



Jollyreaper:

"When the machine truly thinks for itself and can tackle a problem from start to finish the same way the programmers and engineers would, that's strong AI."

But it was designed to think for itself. So it is merely doing what it was coded to. Therein lies the paradox.

Deep Blue had some ability to think for itself, in that not every possible chessboard configuration needed to be coded into it by hand.

Higher intelligence would be signified by the ability to adapt to a wide variety of problems - not only being able what to figure out what to do in a chess configuration you haven't seen before, but also being able to play other board games when asked, or dealing with things less clean and clearcut than a board game. And potentially think outside the box and question the merit of what it's doing in the first place - Deep Blue and its siblings would never have concluded, of any game, that the only winning move is not to play.

Anonymous said...

(SA Phil)

Tony,

Do you honestly think humans will be writing "patches" in say 200 years?

How many programs do you write in machine code or assembly language these days?

I imagine at most "programmers" will look at a graphical interface and draw relationships between logic blocks. Everything else will be automated.

At least they will just tell the computer "I need you to do this" and the computer will reprogram itself and all the networked computers involved on the fly.

The only human "coders" will be the ones figuring out how to get computers to code better, they will be work in computer R&D.

jollyreaper said...

If you look at it a certain way, raising a child and teaching it is like programming. Your expert system is like a voodoo zombie doing only what it is told without imagination or like a magician commanding a bound djinn. It will do what you ask but never exceed the parameters.

Strong AI would need to be created just like a child is brought into the world. What it does after is where things get interesting. The king's son doesn't want to rule and make war but write poetry instead. The strong AI cares nothing for patentable designer drugs and refuses to work on that, finding obscure branches of mathematics more interesting.

The reasonable conclusion is that any AI that can be expected to function as reliably and dependably as a machine is going to be an expert system and will never exceed it's programming. Any AI that can be trusted to think for itself and react to the unexpected as well as or better than a trusted human is strong AI and should carry with it significant storytelling potential. There's no story in a cage holding a prisoner, not from the cage's point of view. There's no story in a computer minding a shock collar and zapping the prisoner if he tries to leave where he is confined. But when there is no human jailer, when the machine has that responsibility and is smart enough to recognize and defeat any escape attempt just as the human jailer would, then you can ask whether the machine might not be convinced to change it's mind on the subject.

Heh. Imagine putting a conscience inside the dread superweapon and the commander has to justify the morality of using it to the satisfaction of the AI before it will fire. Imagine ending up like the stalemate in WWI where rebellious soldiers mutinied and said they would hold the line against attack but make no more attacks of their own.

Tony said...

SA Phil:

"Do you honestly think humans will be writing "patches" in say 200 years?"

Yes.

"How many programs do you write in machine code or assembly language these days?"

I haven't written anything in assembly language since I graduated from college, but they still teach you how, not for practical reasons, but to help you understand how the computer works. It also helps you appreciate high level languages, which abstract away a lot of tedious, repetitive work.

"I imagine at most "programmers" will look at a graphical interface and draw relationships between logic blocks. Everything else will be automated.

At least they will just tell the computer "I need you to do this" and the computer will reprogram itself and all the networked computers involved on the fly.

The only human "coders" will be the ones figuring out how to get computers to code better, they will be work in computer R&D."


Heh, "logic blocks". I saw a demonstration of something like that at COMDEX in 2001. Obviously, it never went anywhere.

What the programmer does, no matter how the code is generated, is design the logic by which the program operates. That's never going to change, because it's a creative process, not a mechanical one. It takes imagination, not simply following some set of rules. All of the stuff that can be accomplished according to a set of rules is already coded into standard libraries. You need a list? You include the object library that contains the template code and you invoke a list object. Same with trees, database connections, math functions, whatever. It's been done that way for decades now.

In reality there are no more "coders". There are no more people who make a living writing code to someone else's design. The only people left making money are people who can do design. To them code is just a medium of expression. That's not going to change.

Anonymous said...

Tony,

Heh, "logic blocks". I saw a demonstration of something like that at COMDEX in 2001. Obviously, it never went anywhere.

====
I disagree

This is how Automotive software is developed. Normally because of limited processor performance it is coded/optimized by coders - but there are systems that are used in advanced(R&D)vehicle programs that will autocode directly from a logic diagram.

The goal of that ETAS/INCA system for example is to completely remove human coders from the software development for automobile embedded controls.

PLCs use a graphic interface as well. Of course they have a lot more processing power than they need to do their tasks which lets this work easily.

Which is what will be the case on a 200 year out spcecraft.



(SA Phil)

Tony said...

SA Phil:

"I disagree

This is how Automotive software is developed. Normally because of limited processor performance it is coded/optimized by coders - but there are systems that are used in advanced(R&D)vehicle programs that will autocode directly from a logic diagram.

The goal of that ETAS/INCA system for example is to completely remove human coders from the software development for automobile embedded controls.

PLCs use a graphic interface as well. Of course they have a lot more processing power than they need to do their tasks which lets this work easily.

Which is what will be the case on a 200 year out spcecraft."


I must disagree in turn. You're talking about a standard architecture with wide applicability. That makes designing an automated programing system worthwhile. Weapons and sensor systems are always going to be unique architectures, requiring unique solutions.

Anonymous said...

(SA Phil)

It could be that guidance and sensors are all pretty much standardized in the midfuture.

Since every spacecraft Military and civilian will need them.

Anonymous said...

(This is Milo. Blogger appear to have downgraded their design so the Name/URL button no longer works unless you have JavaScript enabled, and their JavaScript crashes my browser.)



Jollyreaper:

"Imagine putting a conscience inside the dread superweapon and the commander has to justify the morality of using it to the satisfaction of the AI before it will fire."

An effective superweapon needs to have some notion that certain people are "allied" and should not be shot at. This plants the seed for rethinking exactly how wide your class of "allies" is.



Tony:

"I haven't written anything in assembly language since I graduated from college, but they still teach you how, not for practical reasons, but to help you understand how the computer works."

If nothing else, you need to know assembly language if you're in the business of writing compilers for other languages.


"It also helps you appreciate high level languages, which abstract away a lot of tedious, repetitive work."

Which is the important bit - high level languages can only abstract away tedious, repetitive work. No programming language, short of strong AI, will abstract away creativity.


"Weapons and sensor systems are always going to be unique architectures, requiring unique solutions."

While we're on the subject, weapons and their controlling software will need to be reinvented multiple times, since rival governments are unlikely to share their tech.

Tony said...

SA Phil:

"It could be that guidance and sensors are all pretty much standardized in the midfuture.

Since every spacecraft Military and civilian will need them."


It's not at all that likely. Civilian spacecraft need positioning and collision avoidance radar, just like modern seagoing ships and aircraft. Military spacecraft need sophisticated search radars, also analogous to modern sea and air systems. And there's no analog between anything on civilian vessels or aircraft and military fire control systems.

Unlike automotive control systems, there's no hardware catalog for military systems. And if there's no hardware catalog, there's no standardized software architecture. Sorry.

Anonymous said...

Jolly here. Blogger hateses me, precious.


An effective superweapon needs to have some notion that certain people are "allied" and should not be shot at. This plants the seed for rethinking exactly how wide your class of "allies" is.


A lot of skull sweat went into the firing mechanism for atomic bombs, not just the physics of it but the security side. Who has the right to fire, how do you keep positive control of the weapon, how do you make it deployable in a war situation but not easy to go rogue with if one of your generals or pilots goes a little funny in the head? If we're talking planet smashers and grand operatic doomsday devices, the matter becomes increasingly urgent.

Anonymous said...

Tony,

It's not at all that likely. Civilian spacecraft need positioning and collision avoidance radar, just like modern seagoing ships and aircraft. Military spacecraft need sophisticated search radars, also analogous to modern sea and air systems. And there's no analog between anything on civilian vessels or aircraft and military fire control systems.

Unlike automotive control systems, there's no hardware catalog for military systems. And if there's no hardware catalog, there's no standardized software architecture. Sorry.

=========

Again I think you are underestimating the sheer computing power that will be availible. It will just be much simpler to do all these things.

The computers will be able to adapt/create the "hardware catalog" even for one-offs.

While we didnt actualy use that term, but much of our "hardware catalog" was highly adjustable. Enter some values into the Cal tool and you could switch from a analog to digital input sensor, and adjust the frequency or voltage response calibrations.

-----
As to the sensors I think the civilian vs military senors idea sounds nice in a sci fi book. But I dont think the distinction will exist 200+ years from now.

The civilian stuff will just be so good and so relatively inexpensive that the military will just be better off using that.

Its already starting to happen with some electronics.

------

As to weapon guidance - if you are right it will all be missile guidance. The missiles will need all the same kinds of guidance ... as any other space craft.

If you are wrong and lasers rule the day -- it will be laser telescope optics. Which will be part fo everyone's sensor suite (needed for communications anyway.)

Tony said...

Jollyreaper:

"A lot of skull sweat went into the firing mechanism for atomic bombs, not just the physics of it but the security side. Who has the right to fire, how do you keep positive control of the weapon, how do you make it deployable in a war situation but not easy to go rogue with if one of your generals or pilots goes a little funny in the head? If we're talking planet smashers and grand operatic doomsday devices, the matter becomes increasingly urgent."

Maybe we could call this the "Ender Syndrome". Would Ender have used the Mollecular Disruption Device on the Buggers' homeworld if he had known that was what he was doing, rather than playing a game? Would an intelligent planet smasher conduct an attack without being convinced there was no other way?

Or, to take things in another direction, could you program an AI to exert itself fully in its own destruction, without giving it the ability to ask why? IOW, could you really make a "good soldier" AI, yet have it exhibit all the signs of intelligence? If you could, would it involve installing the proper set of motivations, or would it involve teaching the AI to see things your way? If the second, what happens when the administration cahanges, or even the same administration takes significant policy departures?

Anonymous said...

blogger still hatses us.

Maybe we could call this the "Ender Syndrome". Would Ender have used the Mollecular Disruption Device on the Buggers' homeworld if he had known that was what he was doing, rather than playing a game? Would an intelligent planet smasher conduct an attack without being convinced there was no other way?

That's the stuff the head shrinkers worked on when the whole nuke system was assembled. You've got two men in a bunker with keys. How can you ensure they'll pull out their keys and turn them when the time comes?

Or, to take things in another direction, could you program an AI to exert itself fully in its own destruction, without giving it the ability to ask why? IOW, could you


One idea I had for keeping humans in the loop is that any AI smart enough to replace humans in combat was too smart to get into combat in the first place or went nuts from the combat stress. The baseline Mark 1 human represented the perfect balance of smarts and stupidity that could not be bested by machine. AI generals ran the wars and squishy humans did the dying.

Anonymous said...

Jolly again. Blogger hatses.

The flip side is that any AI suitable for employment as a weapon, especially as an expendable weapon, was dangerously psychotic in ways that made Norse berserkers look all cute and cuddly-wuddly. Deploying something like that is like calling on Godzilla because King Ghidora is attacking you. Great, you got rid of the hydra dragon monster but now you're stuck with freakin' Godzilla.

really make a "good soldier" AI, yet have it exhibit all the signs of intelligence? If you could, would it involve installing the proper set of motivations, or would it involve teaching the AI to see things your way? If the second, what happens when the administration cahanges, or even the same administration takes significant policy departures?

I would think that brainwashing and inhibition-locking an AI would be a dicey process, the same as with real human beings (Call it the HAL syndrome). If you can easily clone AI minds then that's not a problem, you only need to get it right once or twice. As my personal preference for AI designs, I like the idea of the consciousness stored in unduplicatable neutral net crystals. It's a piece of imaginary tech that says you can't just run off copies of an AI, each one will be unique -- potentially immortal but you can't back it up.

I think there's real potential for Bolo-style storytelling when your super-weapon judges your new administration to be immoral and refuses to fight. Despite all the hyper-weapon supercheese, the best part of a really good Bolo story is that the machines are better human beings than we are, believing in and adhering to the ideals that we say are important but only give lip-service to. And there's the ambiguity where cynics will say that the machine wasn't feeling, wasn't deciding, wasn't conscious, was just carrying out programming while those who lived through the fight can't shake the feeling that the machine was more than that and it's not just sentimentality or anthropomorphization.

Tony said...

SA Phil:

"Again I think you are underestimating the sheer computing power that will be availible. It will just be much simpler to do all these things."

If you put the equivalent of Watson into a smart phone form factor, you still have to program it. Also, the more hardware power you have, the more you can brute force certain functions, but those functions still don't go away. I've got powerful graphical design tools on my work desktop. Those tools have numerous integrated code generators. I still have to figure out what the system is going to do in logic terms, and how to assemble all of the components. That simply will not change, no matter how many levels of abstraction you invent.

"The computers will be able to adapt/create the 'hardware catalog' even for one-offs."

A one-off is, by definition, not a standard architecture. some human is going to have to write the software definition of the hardware functions, and the logic of how the system converts inputs into outputs. There is no standard way to that that applies to any arbitrary machine. Period.

"While we didnt actualy use that term, but much of our 'hardware catalog' was highly adjustable. Enter some values into the Cal tool and you could switch from a analog to digital input sensor, and adjust the frequency or voltage response calibrations."

Of course -- because you were working with a standardized architecture with standardized input and output conventions. IOW, you had a simple mapable process with a limited number of choices for output, given a certain input. But somebody (and probably a large group of somebodies) at ETAS analyzed all of the things that could happen in each model of automotive ECU and built a system of software tools to abstract all the code away from automotive system design engineers. In effect, you're not really programming anything, you're just adjusting settings in a pre-built system template.

"As to the sensors I think the civilian vs military senors idea sounds nice in a sci fi book. But I dont think the distinction will exist 200+ years from now.

The civilian stuff will just be so good and so relatively inexpensive that the military will just be better off using that.

Its already starting to happen with some electronics."


There is a lot less COTS stuff in military systems than there is anywhere else. And while there may be some degree of technoligcal convergence, do you really expect non-military systems to need the level of discrimination and counter-countermeasures that military systems will?

"As to weapon guidance - if you are right it will all be missile guidance. The missiles will need all the same kinds of guidance ... as any other space craft.

If you are wrong and lasers rule the day -- it will be laser telescope optics. Which will be part fo everyone's sensor suite (needed for communications anyway.)"


I said "fire control", not "guidance". Again, civilian systems, even communications -- which are much more likely to be microwave rather than visible or IR light frequency -- just don't have an application.

Tony said...

Jollyreaper:

"That's the stuff the head shrinkers worked on when the whole nuke system was assembled. You've got two men in a bunker with keys. How can you ensure they'll pull out their keys and turn them when the time comes?"

But you have to have the Two Man Rule to ensure that no single person can screw things up. Yet the more people you require to get anything done, the harder it is to ensure it happens. Two men turned out to be not only the minimum acceptable number for nuclear surety, but the maximum practical number for deterrent credibility. (In the sense that the Soviets could accept that two guys would probably turn the keys in most of the command bunkers, but three or more guys might not.)

A little known fact about the two man rule was that it was not enforced on the honor system. Either it was mechanically impossible for a single person to release a weapon or, in the maintenance and repair context, there were armed guards who would rather shoot you than let you in the nuclear exclusion area alone.

The paranoia and risk aversion built into the US nuclear security program has to be seen to be believed. Personnel and systems were under constant scrutiny. Heck, you have to have a Secret clearance just to guard physical access to areas the nukes might be in. You need a Top Secret clearance with special access provisions to get near enough to touch them. Having said that, given the consequences of accidental or unauthorized release, I would have to say it was all fully justified.

"AI generals ran the wars and squishy humans did the dying."

What human in his right mind would risk death on the orders of a machine, no matter how intelligent?

Anonymous said...

Tony said:"Or, to take things in another direction, could you program an AI to exert itself fully in its own destruction, without giving it the ability to ask why? IOW, could you really make a "good soldier" AI, yet have it exhibit all the signs of intelligence? If you could, would it involve installing the proper set of motivations, or would it involve teaching the AI to see things your way? If the second, what happens when the administration cahanges, or even the same administration takes significant policy departures?"

One obvious way, (not the one I would endorce) would be to purposely create an AI that had a "mental illness" so that it wanted to destroy itself (along with your enemies); other, even more twisted, means of having an advanced AI be perfectly happy killing vast numbers of people. Copy the personalities of psychopaths and sociopaths into the AI, along with a perverted desire to please its masters, and you get around all of those questions. Rather far into the demonic end of the ethics spectrum, but it gets you your super-intelligent super weapon.

Nomanclature as descriptor: there are two viewpoints, as far as I know; a name that contains, in part or whole, something about what the thing does or is; a unique name that simply identifies the thing to destingush it from anything else.

Tony, Phil, jollyreaper, Milo, et al: just as there is not now a monolithic computer archetecture or total commanality between civilian and military hardware, I don't see that there being such in the future; military and civilian uses and purposes have too many differences. Having said that, there will be some overlap, but there will be unique systems and modifications that the military will need that civilians will not have.

SA Phil said:"There will be no stealth, propulsion, weapons technology, etc will all be known quantities."

Umm, no...not always. New modifications, prototype weapons, novel ways of using existing weapons or other capabilities,would not necessarily be known until one of your ships survived (or an observation platform) to report it. Things are only known after the first time they are encountered or conceived of. If you haven't encountered or conceived of it, then it can't be 'known'. If you don't know about it, you can't deal with it.

Ferrell

Thucydides said...

AI as self sacrificing weapons isn't a weird concept at all. Samurai warriors evolved into cultured gentlemen who were (ideally) as adept in poetry and tea ceremonies as they were in killing, and were rigorously trained to behave as if each breath was their dying one.

Now of course not every Samurai embodied these ideals (any more than every European Knight embodied the ideals of Chivalry), but if AI is possible and reproducible, then once you have the program right you simply upload it into all the relevant military hardware.

Anonymous said...

jolly

The paranoia and risk aversion built into the US nuclear security program has to be seen to be believed. Personnel and systems

How do you explain the "lost" nukes on that B-52 from a few years back, where they uploaded live warheads and nobody knew about it? I'd have laughed it out of the building if someone presented it in a techno-thriller and yet here we are, faced with the evidence. Everything I'd read went along the lines of the professional paranoia and "you're in good hands" line that you're talking about and I figured if the military would take anything seriously, nuclear security would be it. Now I'm not so sure.

I worked with a guy who said he served on a destroyer that carried nuke-tipped tomahawks back in the Cold War. He had a story he told about having found a flaw in their security procedures and reported it up the chain of command. He was ignored. He claims he decided to teach them a lesson by making the warheads "disappear" on paper, the flaw he was talking about. Said it caused a huge crapstorm. I'd dismissed the story as BS because a) people who come anywhere near these systems tend not to talk b) anyone playing a stunt like this would be have a special run of the book printed up just to it could be thrown at him in volume and c) it seemed improbable that something so serious could be reduced to a military "good times" story. But hell, if we manage to accidentally upload nukes to a bomber without anyone knowing about it, maybe his story's not so far-fetched after all.

What human in his right mind would risk death on the orders of a machine, no matter how intelligent?

What human in his right mind would risk death on the orders of his superiors, men dull and unimaginative enough to attain command rank in the military? :) When you get much beyond hunter-gatherer war parties in the woods, you're taking orders from people you've never met, never seen, and may as well be computers in a secure data bunker somewhere.

Teleros said...

jollyreaper: "Again, it all depends on the posited setting and tech level as for how much automation can be had. If you have trans-human AI's, the whole military-industrial complex can be automated and you're back to the question of what good baseline humans are for. But I think that would fall under far future, not mid-future."
Oh the (heavily genengineered / cyborg-ed) humans would just be there for story purposes, and perhaps to make decisions (because humans like to (appear to) be in control) or act as the ship's away teams.

SA Phil: "As to AI - you wont have HAL AI for space combat. It would probably be more specialized. And probably not "AI" in the science fiction sense."
Why not? If you've got an AI that's as smart / smarter than a human, unless it takes much longer than a specialised AI to do a particular task (much longer being defined as "long enough extra to make a noticeable difference in combat"), go with the smart one.

"A very plausible future is one where the only humans that are killed in a war are "collateral damage""
Depends on the scope and nature of the conflict. Even if you have transhuman AIs doing the dirty work, people may still want to make the decisions etc, and become targets that way.

Tony: "Nobody, human or computer, can write procedures for unexpected events or environmental conditions. And no amount of testing in any kind of simulated environment, no matter how detailed, is going to find out how a system behaves under conditions the test designers didn't foresee. That's the purpose of human supervision in service."
No, that's why you code your AI to respond like a human. When you stumble upon unexpected events, you respond based on things like memories, current state of mind, other people (ie other information sources), and instinct. Why can't AIs (eventually) do the same?

"Humans will never be out of the code writing business, simply because no one human or group of humans will ever be able to predict all the possible parameters of every code writing environment."
They won't need to - see above. I'm sure that one day (assuming I live that long :P ) I'll be able to talk to my computer, instruct it (say) to draw a cool new fantasy sword for a game it's working on for me, and it'll duly draw one up, even if it has no clue what a sword is, because it'll be able to check online under its own initiative.

Teleros said...

SA Phil: "What will be unexpected in Space Combat though?

There will be no stealth, propulsion, weapons technology, etc will all be known quantities.

There won't be ambushes, feints, slick maneuvers, and so on.

The "real" space war scenario would likely be pretty boring from a drama perspective."

There will probably be a fair bit of tactical manoeuvring and such, not to mention freak accidents etc. Yes, I suspect that there'll be good doctrines for handling ships of such-and-such a class, but people won't always stick to doctrine.

Mangaka2170: "The advantage of a human crew, in this situation, is that they'd probably be able to make the necessary repairs until they get to a cageworks. Not only would a drone have difficulty determining the source of attack (since there wouldn't be any confirmed hostile sensor tracks, it might even conclude a long-range attack or a remote missile strike from the other side of the planet), but it wouldn't be able to repair itself if it needed to"
Not really. A human crew might think it an attack from the planet too, and surely a fully automated ship would have some sort of on-board repair system(s)? As to the decision-making side though, well it all depends on quality of the AI. Don't forget that humans can get slack - your human crew may not notice the early warning signs of a terrorist attack on the ship, whereas an AI, incapable of boredom, might spot them.

Milo: "To get a computer to do something for you, you need to be able to explain to the computer what you want. Explaining to a computer what you want is, per definition, coding."
I think there has to be a distinction drawn though, between coding a computer in C++, machine code, or whatever, and me sitting here and saying "design me a flight simulator program please."

"But it was designed to think for itself. So it is merely doing what it was coded to. Therein lies the paradox."
Uhm, no? Go back to the basics. Artificial intelligence = intelligence that has been designed. Crafted. Created. Coded. As opposed to natural intelligence - an accident of dear old Mother Nature. So the fact that an AI has been created to come up with solutions to problems the same way a human does doesn't cause a paradox, because we are after all talking about a strong man-made intelligence.

jollyreaper: "Heh. Imagine putting a conscience inside the dread superweapon and the commander has to justify the morality of using it to the satisfaction of the AI before it will fire. Imagine ending up like the stalemate in WWI where rebellious soldiers mutinied and said they would hold the line against attack but make no more attacks of their own."
Strong French AI. Heh.

More seriously, I suppose you need some sort of "read-only" limitations built into your AI to prevent it changing its mind. "Thou shalt never, under any circumstances, permit the prisoner to escape (see C:\escape-definitions.txt for details)", and so on. So worst case, all the AI can do is say "I'd like to let you go, but I'm physically incapable of doing so". Or perhaps have an expert system "OS" which can run several subordinate AIs for certain tasks.

Tony: "Or, to take things in another direction, could you program an AI to exert itself fully in its own destruction, without giving it the ability to ask why?"
I don't see why not. It should be possible to create an AI without a sense of self-preservation, or to unhesitatingly follow orders. Worst case, perhaps a sort of tiered system of instructions (a bit like the above):

1. You will always obey legal orders.
2. You will always try and make suggestions to maximise efficiency (eg in combat, if a combat AI).
...

So even if it does ask why, it still does as its told.

jollyreaper said...


More seriously, I suppose you need some sort of "read-only" limitations built into your AI to prevent it changing its mind. "Thou shalt never, under any circumstances, permit the prisoner to escape (see C:\escape-definitions.txt for details)", and so on. So worst case, all the AI can do is say "I'd like to let you go, but I'm physically incapable of doing so". Or perhaps have an expert system "OS" which can run several subordinate AIs for certain tasks.


My favorite explanation for the Cylon religion in BSG (since disproven by the way it turned out) was that religious beliefs were implanted in their minds for precisely that sort of reason. Thou shalt, thou shalt not. But these are thinking machines and it's unnatural to put articles of faith into the mind of something that's supposed to be running on logic. Animals = meat. Dogs = animal. Dogs = meat. Wrong? Why? Dogs = pet. Pets = animal = meat. Wrong? Why? "Just because!"

Look a the long history of twisting religion for justification of things. I can find you people on both sides of the Civil War who can prove that the Bible is on the side of slavery or abolition. Says it right here in the scripture! lol

So my thinking is that the Cylons were coded with a set of morals that should have seemed reasonable but they ended up coming up with convoluted rationalizations and justifications, either trying to create a logical argument to underpin something that was arbitrary or to justify something they want to do that the law seems to prevent.

Tony said...

jolly:

"How do you explain the "lost" nukes on that B-52 from a few years back, where they uploaded live warheads and nobody knew about it? I'd have laughed it out of the building if someone presented it in a techno-thriller and yet here we are, faced with the evidence. Everything I'd read went along the lines of the professional paranoia and "you're in good hands" line that you're talking about and I figured if the military would take anything seriously, nuclear security would be it. Now I'm not so sure."

My explanation? Loss of focus within the USAF due to "war on terror". It was bound to happen sometime. When it did a lot of people got punishes and/or relieved. I understand they're back on track.

"I worked with a guy who said he served on a destroyer that carried nuke-tipped tomahawks back in the Cold War. He had a story he told about having found a flaw in their security procedures and reported it up the chain of command. He was ignored. He claims he decided to teach them a lesson by making the warheads "disappear" on paper, the flaw he was talking about. Said it caused a huge crapstorm. I'd dismissed the story as BS because a) people who come anywhere near these systems tend not to talk b) anyone playing a stunt like this would be have a special run of the book printed up just to it could be thrown at him in volume and c) it seemed improbable that something so serious could be reduced to a military "good times" story. But hell, if we manage to accidentally upload nukes to a bomber without anyone knowing about it, maybe his story's not so far-fetched after all."

My direct, personal experience with nuclear security was back in the 80s, but the principles hadn't changed for over 30 years then, and I doubt they've changed in the last 20. There were enough people with direct knowledge of whether or not nukes were on a vessel that no paper "offload" would have made a difference. And anybody that deliberately falsified nuclear weapons records would have been standing tall at Captain's Mast, facing the maximum NJP penalty.

"What human in his right mind would risk death on the orders of his superiors, men dull and unimaginative enough to attain command rank in the military? :) When you get much beyond hunter-gatherer war parties in the woods, you're taking orders from people you've never met, never seen, and may as well be computers in a secure data bunker somewhere."

Dull and unimaginative generals tend to have mediocre armies. Good armies are generally commanded by highly intelligent and imaginative men, often men who could legitimately be rated polymaths -- Alexander, Caesar, Gustavus, Frederick II, Napoleon, Nelson. Even in the more bureuacratic armies and navies of the last couple of hundred years, there are many examples of exceptional men rising to the top -- Sherman, Fisher (who was perhaps too imaginative), Arnold, Zhukov (rising to the top in a regime that positively encouraged dull routiners). It's not enough that the General said so, the troops have to believe he knows what he's doing. And, I think, nobody would have that kind of faith in an AI.

Anonymous said...

(SA Phil)

If you base "AI" on how computers work now - then there will be no problem getting it to participate in its own destruction.

If you mean HAL or Cylon AI - that might be different.

But is that really AI, or is that giving "human" qualities to a machine?

jollyreaper said...


have to believe he knows what he's doing. And, I think, nobody would have that kind of faith in an AI.


Based on my dad's description of what it was like in Vietnam under the whiz kids like MacNamara and all their fancy formulas, I don't see how a computer could have made it any worse.

Besides, when you get the hidebound guys who are constantly trying to figure out how the book would have them do it and show no variation from that model, all an AI would do is save them the step of looking it up for themselves. Would this be an optimal system? Hell, no. Potentially disastrous. Could I see someone developing it anyway, because people can sometimes do very, very dumb things? Hell, yes.

This is for a dumb computer, mind you, expert system. Strong AI's that can truly think for themselves are a whole 'nother ballgame.

Tony said...

jollyreaper:

"Based on my dad's description of what it was like in Vietnam under the whiz kids like MacNamara and all their fancy formulas, I don't see how a computer could have made it any worse."

Vietnam, huh? This would be the same Vietnam that generated the generals who led us so effectively in Desert Storm, and at least the initial stages of Afghanistan and Iraq? You don't perhaps see some connection between a poorly led engagement and the professional and institutional changes that resulted in much more competent and effective forces a generation later?

"Besides, when you get the hidebound guys who are constantly trying to figure out how the book would have them do it and show no variation from that model, all an AI would do is save them the step of looking it up for themselves. Would this be an optimal system? Hell, no. Potentially disastrous. Could I see someone developing it anyway, because people can sometimes do very, very dumb things? Hell, yes."

You really need to get this chip off your shoulder about the service and its culture. You've been fed way too many cliches and uninformed opinions, and very little truth at all, that I can see. And you're just repeating it all without thinking IMO about what you're saying.

Anonymous said...

Tony

do you really expect non-military systems to need the level of discrimination and counter-countermeasures that military systems will?

====
What I expect is that the electronics, sensors, computers, etc and the software that runs them to so far exceed the requirements of running, guiding, and fighting a spacecraft that it simply wont be a complex enough of a task to need customized anything.

How complicated is it to automatically turn a lightbulb on at 7am?

That is how routine I think the system designers 200 years from now will see those tasks.

I already said I think the human programers / "coders" whatever they are called then will exist - but not on space craft. They will exist in Computer R&D

The computers themselves will do all the "systems integration" stuff.

If its a cuustom piece of hardware - the computer on that bit of harware will just tell all the other computers on the spacecraft what they need to do in order to integrate it.

Everything that can be controlled that you could possibly describe in the midfuture will be "plug and play"

(SA Phil)

Anonymous said...

Jolly here, blogger is buggered

Vietnam, huh? This would be the same Vietnam that generated the generals who led us so effectively in Desert Storm, and at least the initial stages of Afghanistan and Iraq?


Um, yeah, that same war. I've read the bios. The junior officers swore up and down that they would never be as pig-headed and out of touch as the senior officers who led them into that quagmire.

You don't perhaps see some connection between a poorly led engagement and the professional and institutional changes that resulted in much more competent and effective forces a generation later?

I'll let you know when I see it. I've read the glowing accounts of the Gulf War. Struck me at the time like a giant live fire exercise. I could never take seriously the idea of the Iraqis posing a serious threat in a stand-up fight. Now I could be completely wrong and am being deceived by just how amazingly wonderful our military is because they made it look so easy. That's possible. But it's also true that we stopped the war at the point the Iraqis were kicked out of Kuwait and before we started actually occupying enemy territory.

But go on, tell me about how well our professional, best trained military in the world is doing in Afghanistan and Iraq. We've fought the war so well it's been going on for ten years! But wait, I forget. Are wars the kind of thing where the length it goes on for is a sign of the quality or am I getting it mixed up with something else?


You really need to get this chip off your shoulder about the service and its culture.


The comment about hidebound and by the book isn't just about the military, it's about any ginormous bureaucracy. Innovation tends to be frowned upon. This goes from shooting guns to running religions to building cars. Don't rock the boat.

You've been fed way too many cliches and uninformed opinions, and very little truth at all, that I can see.

So prove me wrong. :)

And you're just repeating it all without thinking IMO about what you're saying.

You often make the mistake of assuming that people whose opinions differ from your own are suffering from appalling ignorance.

Anonymous said...

(Milo here.)



Anonymous:

"AI generals ran the wars and squishy humans did the dying."

Also add AI politicians to decide what the war is about, and you got yourself a really scary scenario.



Tony:

"What human in his right mind would risk death on the orders of a machine, no matter how intelligent?"

I dunno. Depends on what the machine is fighting for. If the propaganda is catchy, people will listen regardless of who's saying it. (Okay, it'll help if it's a cute machine.)

It also depends on how long AIs have been around and accepted as part of society. If AIs are new, they'll probably need a human "agent", even if he just acts as a figurehead and the AI really does all the work. But then, Hollywood scaremongering aside, new AIs aren't going to be put in charge of the military. People will only entrust them with that once they've proven themselves capable and responsible in less critical roles.



Ferrell:

"One obvious way, (not the one I would endorce) would be to purposely create an AI that had a "mental illness" so that it wanted to destroy itself"

Then how do you get the machine to stand down if the war ends?

If all a bomb wants is to blow up, then it might as well do so in your own base. You need to teach the bomb that it's too expensive to be allowed to destroy itself unless it takes a significant amount of enemy forces down with it. Oh look, now it has a sense of self-preservation!


"New modifications, prototype weapons, novel ways of using existing weapons or other capabilities,would not necessarily be known until one of your ships survived (or an observation platform) to report it."

Lightspeed communication means you don't need to survive the encounter or return home. If you survive long enough to laser a message, that's enough.



Thucydides:

"Samurai warriors evolved into cultured gentlemen who were (ideally) as adept in poetry and tea ceremonies as they were in killing, and were rigorously trained to behave as if each breath was their dying one."

Most warrior cultures tried to teach their warriors to not fear death. For every individual who truly lived up to those teachings, there were plenty who only paid lip service to the ideal.

Tony said...

SA Phil:

"I already said I think the human programers / "coders" whatever they are called then will exist - but not on space craft. They will exist in Computer R&D"

To quote myself:

"After the battle, the system logs and databases are copied to the appropriate contractor with instructions to add or update appropriate event handlers for the next release."

I agree there wouldn't be systems programmers in the fighting forces. But there will be people programming systems, not just doing software R&D.

"Everything that can be controlled that you could possibly describe in the midfuture will be 'plug and play'"

You obviously don't understand what plug-n-play means. It does not mean, as you seem to think, that a piece of hardware or a software module can just be introduced to the system and the system will automatically figure out how to use it. What it means is that the device or module conforms to various interface conventions and advertises certain facts about itself for automated discovery. But somebody still has to decide what to do with the thing, and write the code to do it.

jollyreaper said...


"What human in his right mind would risk death on the orders of a machine, no matter how intelligent?"

I dunno. Depends on what the machine is fighting for. If the propaganda is catchy, people will listen regardless of who's saying it. (Okay, it'll help if it's a cute machine.)


The computer god trope may be old as hell but I still think it's valid. It can come in a couple of flavors:
1) Computer as false idol. It's just priestcraft and cons, no more real than a mechanical idol sacrifices are dumped into. Despite the techno-twist, it's fundamentally no different from any other cult.

2) Computer as agent of God. We revere written books and claim every word in it is holy and passed down by god himself. We'll even claim he invented the writing system. Computers are too new for us to treat them so but put it out a few thousand years where nobody even has good documentation on when computers were invented and it would be very easy for the tech to become mysticized. Even if the computer itself is not ascribed deliberately mystical powers, the aura of majesty and authority associated with it would be overwhelming. The last surviving AI in the Postman was sort of like this. Turns out the computer was destroyed and the scientists at the university were doing all the work. But it doesn't take much imagination to consider what it would be like a generation or two down the line if the AI were real and handing out good, solid advice on when to plant crops, weather predictions, etc. So basically you're adding the mysticism and importance given to holy books with the seriousness regard any bureaucrat has for The Way Things Are Done and automating all of it.

3) It's a strong AI, self-aware, all the trappings. Explored thoroughly in Orion's Arm. Many of the AI's are willing to accept their position as a living god to lesser sophonts.

Tony said...

jollyreaper:

"Um, yeah, that same war. I've read the bios. The junior officers swore up and down that they would never be as pig-headed and out of touch as the senior officers who led them into that quagmire."

Vietnam was not our best moment. but that doesn't mean it was typical, either of the US services or military services in general. And it did serve to refocus us on what we needed to be doing.

"I'll let you know when I see it. I've read the glowing accounts of the Gulf War. Struck me at the time like a giant live fire exercise. I could never take seriously the idea of the Iraqis posing a serious threat in a stand-up fight. Now I could be completely wrong and am being deceived by just how amazingly wonderful our military is because they made it look so easy. That's possible."

Old saying in the Marine Corps:

"First you win, then you fight."

Yeah, the Iraqi ground forces were a bunch of sad sacks, but only because our preparatory air campaign had eviscerated their comand-and-control and logistics. Things wouldn't have been so jolly had we had to actually fight an undemoralized army with a secure rear.

"But it's also true that we stopped the war at the point the Iraqis were kicked out of Kuwait and before we started actually occupying enemy territory."

Elements of the 101st Airborne Division finished the ground campaign on the Euphrates River. US II Corps finished on the approaches of Basrah. The Coalition occupied southern Iraq in order to make it operationally impossible for the Iraqis to remain in Kuwait.

"But go on, tell me about how well our professional, best trained military in the world is doing in Afghanistan and Iraq."

I did say:

"at least the initial stages of Afghanistan and Iraq"

After the toppling of the Taliban and Saddam, I'm not so sure there's much we could have done other than grind away and get nowhere. In any case, I view the initial, mostly conventional occupations as something entriely separate from the ensuing guerilla wars.

"The comment about hidebound and by the book isn't just about the military, it's about any ginormous bureaucracy. Innovation tends to be frowned upon. This goes from shooting guns to running religions to building cars. Don't rock the boat."

That can be a consequence of large bureaucracies, but as we have also established, those bureaucracies are necessary for large organizations to function. Also, there are many large companies that encourage and reward innovation. If you know anything about the modern business climate, you should be able to name at least five off the top of your head, each in a different industry.

"So prove me wrong. :)"

I don't have to. You're doing just fine on your own.

"You often make the mistake of assuming that people whose opinions differ from your own are suffering from appalling ignorance."

Sometimes it is a mistake. Sometimes it isn't.

Thucydides said...

WRT the wars, what soldiers can and cannot achieve is often a function of politics rather than military capability.

To get away from the Americans for a moment, consider the Eastern Front of WWII. Soviet forces were deployed stacked up apparently for an offensive on Stalin's orders and promptly chopped to pieces. German forces drive deep into Russia but the rear echelons proceed to create a hostile environment for the occupants on ideological grounds, which in turn breeds a hostile environment for the logistics trains to move through. Soviet forces stream in for a series of uncoordinated counterattacks on the orders of the Generalissimo of the Soviet Union, while the professional officers of the STAVKA cower in fear. German armies are hived off into various rabbit chases on the order of the Führer while the professional officers of the OKW cower in fear. After some time, the Soviets free themselves from many of their political restrictions, get their act together and proceed to hammer the Germans to pieces (often because the Germans are still following ill considered orders from the top).

OK so this is a very condensed version, but I hope it illustrates why the two mightiest war machines in history had such a hard time of it.

I recall reading the US was not really prepared for the finish of the Persian Gulf War, with no clear "end point" for US VII and XVIII Corps to reach. This may have been deliberate, in order to keep options open, or perhaps events moved much faster than planers had anticipated. This is almost certainly the case in OIF, where Iraqi resistance crumbled almost overnight.

Canadian Forces in Afghanistan were shuffled from Kandahar to Kabul and back, and will soon be back in Kabul, not because they are chasing the Taliban, but to follow the direction of Parliament.

Pick your own military example and you should soon discover the political logic (or lack therof) behind the move. I believe the current limited kinetic action in Libya would be most illuminating.

Anonymous said...

(SA Phil)

I suspect human nature is the cause for those examples of "Military" or "Political" Incompetence.

In every organizational structure politics, business, the military, the girl scouts, etc --- The Decision makers become removed from those most affected by their decisions.

The farther away -- the higher up in a hierarchy -- the farther they are removed.

They lose the perspective of whatever the ground level reality is. And therefore make decisions with incomplete or slanted information.

Often hilarity ensues.

Lentulus said...

Computer as God? I think the prospect of "Computer as rational employer" might actually be a good one. Subtracting greed and ego from the executive floor, and enforcing honest-to-god HR best practices because they work, along with a long view of corporate success could produce a far better place to work than the executive-elite-driven financial system we have today.

At least the computer wouldn't be jumping chamber maids in its hotel room.

Byron said...

Why do we always end up debating humans vs. robots? Always. It's like the only time we ever stay on topic. I suppose this thread drifted there as the last one was closing down.

At least the computer wouldn't be jumping chamber maids in its hotel room.
No, that would be replacing politicians with computers. That's not a bad idea, though.

jollyreaper said...

Computer as God? I think the prospect of "Computer as rational employer" might actually be a good one. Subtracting greed and ego from the executive floor, and enforcing honest-to-god HR best practices because they work, along with a long view of corporate success could produce a far better place to work than the executive-elite-driven financial system we have today.

If the machine is self-aware, presumably it cannot be directly subverted and it would require deceit and manipulation and concealment to allow wrongdoing on its watch. But if the machine is not self-aware, then all that's required is corruption to seep right into the system.

If you think about it, the whole idea of having laws, rules, and regulations is so that justice can be applied equally and impartially. We've had to rely on law enforcement and a judicial system because there were no such things as robots. And while the standard is for all men to be treated equally under the law, no one being more equal than another, in practice money talks and you know what walks. In theory, an impartial robot should be great but the question is no longer "who watches the watchmen" but "who watches the programmers?" Is anyone playing games with the code?

Now in the 20th century, especially in totalitarian regimes, the State pretty much assumes the position of the gods for the average citizen. It is omniscient, omnipresent, and omnipotent. But because we lacked thinking machines, humans were in charge and were making up and enforcing the rules.

Computers will make for a more effective and absolute police state but it's still a matter of tools for control and oppression in the hands of human beings. But to the person on the receiving end, it will become easier to state's presence to feel like a personal god. In 1984, Big Brother was an idea fostered by MiniTrue. But without much extrapolation, one can imagine the citizen being minded by a digital conscience. They're not just hearing a propaganda recording, they're having a conversation. The computer is watching what they do, monitoring vital signs like a lie detector, telling them when to work harder, when to take a potty break, asking them to recite the state's catechism, whatever. Prayers or confessions could be offered to the machine and it wouldn't take that much effort to simulate a conversation. A computer-generated avatar could be on-screen talking to them. and all of this would be without the security apparatus being strong AI, self-aware and self-directing.

Stross has a good essay on the kind of Panopticon Society we're heading towards.

http://www.antipope.org/charlie/old/rant/panopticon-essay.html

Anonymous said...

=Milo=



Lentulus:

"Subtracting greed and ego from the executive floor,"

Umm, isn't earning money the point of the corporation? You need to keep some greed to be effective.



Byron:

"No, that would be replacing politicians with computers. That's not a bad idea, though."

I'm partial to replacing politicians with dinosaurs. How can you argue with policy statements like this?

Tony said...

jollyreaper:

"Stross has a good essay on the kind of Panopticon Society we're heading towards.

http://www.antipope.org/charlie/old/rant/panopticon-essay.html"


Typical of people who know a lot of technical facts but are clueless about operational reality. The problem with all surveillance technologies, no matter how cheap and ubiquitous -- it actually gets worse as price goes down and ubiquity up -- is signal to noise ratio. Even with sophisticaed text processing and pattern recognition technology, you still need people to interpret the data and turn it into actionable information. The more surveillance you run, the more people you need. And that's where you run into a competence/reliability/loyalty bottleneck.

jollyreaper said...

Typical of people who know a lot of technical facts but are clueless about operational reality.

Tony, have you noticed that of all the people posting on this blog, you're the only one who keeps ending up at the center of pigfights? It's not what you're saying, it's the way you say it. If someone thinks A and you think B, there's not a chance that you two simply are starting from different fundamental assumptions, no. It's clear that the person is thinking A only because he's an ignorant shit. If he weren't he would think B. Like you, the wise one who isn't an ignorant shit. You see how this causes conflict?

You either:
a) Didn't read the article.
b) Skimmed the first paragraph
c) Read the whole thing and decided to ignore what he said that directly addressed your complaint.

Quoting from TFA:

It is worth noting that while the effectiveness of human-based surveillance organizations is dependent on the number of people involved -- and indeed may grow more slowly than the work force, due to the overheads of coordinating and administering the organization -- systems of mechanised surveillance may well increase in efficiency as a power function of the number of deployed monitoring points. (For example: if you attempt to monitor a single email server, you can only sample the traffic from those users whose correspondence flows through it, but if you can monitor the mail servers of the largest ISPs you can monitor virtually everything without needing to monitor all the email client systems. Almost all traffic flows between two mail servers, and most traffic flows through just a few major ISPs at some point.) Moreover, it may be possible to expand an automated surveillance network indefinitely by simply adding machines, whereas it is difficult to expand a human organization beyond a certain point without having knock-on effects on the macroeconomic scale (e.g. by sucking up a significant proportion of the labour force).

We are all criminals, if you dig far enough: we've broken the speed limit, forgotten to file official papers in time, made false statements (often because we misremembered some fact), failed to pay for services, and so on. These are minor offenses -- relatively few of us are deliberate criminals. But even if we aren't active felons we are all potential criminals, and a case can be -- and is being -- made for keeping us all under surveillance, all the time.

A Panopticon Singularity is the logical outcome if the burgeoning technologies of the singularity are funneled into automating law enforcement. Previous police states were limited by manpower, but the panopticon singularity substitutes technology, and ultimately replaces human conscience with a brilliant but merciless prosthesis.




"You often make the mistake of assuming that people whose opinions differ from your own are suffering from appalling ignorance."

Sometimes it is a mistake. Sometimes it isn't.


What a dichromatic little world you must live in.

Byron said...

Tony:
I have to second jollyreaper here. Actually, I'll repeat what he had to say:
If someone thinks A and you think B, there's not a chance that you two simply are starting from different fundamental assumptions, no. It's clear that the person is thinking A only because he's an ignorant shit. If he weren't he would think B. Like you, the wise one who isn't an ignorant shit. You see how this causes conflict?
My view of it is even more simple. You never admit the possiblity of your being wrong.
I'm going to quote from something that happened a while ago:
"And you could try accepting that other people could be right. I freely admit that maintenance could go either way. You seem completely certain that you're correct, to the point of contradicting yourself to make your points."

I see...

I think you're going to have to accept that not all people were taught or accept the artificial rules of high school forensics competitions. I assess my personal experience to be relevant where I think it's relevant, and rely on it in those situations. I don't accept purely theoretical arguments on anything, at any time, simply because somebody worked out the math or the formal logic correctly. Theory and practice are not and never have been the same thing. I try to exercise decorum in debate as much as I can, but I think what I think and I will say what I think.

The first part is mine, the second yours.

This is the core of the problem we've been having. Your positions are based entirely on your own experience, without considering anyone else's input. I'm willing to admit that I can be wrong. I can't recall you having ever done the same on a factual issue. I'm not trying to be mean, but it's honestly no fun trying to debate someone you can't convince. It's like arguing with someone who holds an unfalsifiabe hypothesis.

Anonymous said...

(SA Phil)

You only need humans to interpret your Big Brother data as long as you are concerned with Justice/Fairness.

Once you decide that it no longer matters if you persecute "innocent" people - its a whole lot easier to get a comupter program to do it.

jollyreaper said...

I'm not trying to be mean, but it's honestly no fun trying to debate someone you can't convince. It's like arguing with someone who holds an unfalsifiabe hypothesis.

What's more, we're talking about stuff that can't really be put to the test.

If we're talking dog-breeding and training you can talk about what works and what doesn't. Some stuff is just flat out right and wrong, not really subject to debate. Some stuff is a matter of taste. If someone's trying to win a dog show, this is what will work. But someone else might feel that the top breeds are getting too inbred and purposefully tries to breed dogs he feels are healthier even if they don't meet the breed standards set down by the kennel people.

There's plenty of room for debate between what's right, what's wrong, and what falls in between. And it can be put to the test in the real world.

Now if we move this off into the realm of fantasy, right and wrong becomes even more complicated. Let's say we're talking about dragon-breeding and riding. Ok, so we can start from a real world basis with what's experienced by professional breeders of dogs and horses. But then someone's going to come in and say dragons should take more after reptiles than mammals because they're lizardy. Well wait a second, they're dragons! They don't bloody exist in the first place! That would be like arguing whether a griffin should behave more like a eagle or a lion. It all depends on how you want to write your griffin!

jollyreaper said...

Now when someone says the kind of dragon he wants to write, now we can start having a reasonable discussion drawing from real world examples. The dragons are going to be like dolphins -- they don't speak human tongues, there's no telepathic mindspeech with the rider, smarter than horses or dogs and some people but they don't talk, they don't have conversations. Temperament will always be fiery, spirited, and uncompromising. They must respect the rider. You don't get learner dragons the same way you have learner horses.

Ok, so now you can talk about how to make the breeding and training feel authentic. If the dragons lay eggs, it's not going to have the same feel as a pregnant mare eventually dropping a foal. Maybe someone with reptile experience could help extrapolate what a dragon stable would be like, perhaps there's a special creche for laying eggs, the stable hands keep a fire pit beneath the stone floor to help keep the heat up since they're a bit further north than the usual nesting grounds for dragons. Someone with animal training background can point out whether an idea seems suitable for a dragon or if it's a bad idea that wouldn't work on any real life creature he's heard of and so would ring false if used on a dragon.

There's also room here for talking about the physics of dragon flight. If the author wants the dragons to be as realistic as possible, living breathing creatures not reliant on magic, there you can talk hard numbers. "This is how big a dragon would have to be and carry a rider. This is what the caloric needs would be." And here you really could tell a person whether or not his idea's plausible. "No. A dragon twice the length of a horse with a wingspan equally as wide, that's not possible." Valid input here.

Now if I'm weighing in, my feeling is that strictly natural dragons are are problematic as giant fighting mecha. Physics is unkind to them. If you're going with magic in your fantasy setting, you may as well make dragons magical. Most fantasy settings allow for levitation. What if the dragon could use a similar process to reduce his weight? Then the physics might work out better. Figure out limitations to what the magic can do so the dragon's abilities are as constrained as any natural animal so the dragonrider has to work within those limitations. Keeps the story interesting, grounded, and fun.

But if someone insists they want realistic, natural dragons but insists on going with a design that doesn't work, oh well. You tried. The answer is "You're wrong," not "You're wrong, you stupid, ignorant ass."

Tony said...

SA Phil:

"You only need humans to interpret your Big Brother data as long as you are concerned with Justice/Fairness.

Once you decide that it no longer matters if you persecute "innocent" people - its a whole lot easier to get a comupter program to do it."


Even totalitarian absolutists have to rely on an appearance of justice. Most people have to be secure against persecution, or revolution happens. And no ruler is so secure that he can't be deposed if the people get angry enough. Arbitrary, uncontrolled persecution has undone more dictators than the mere fact of heavy social and political control. Automating it will just make the revolution come faster.

jollyreaper said...


I agree with most of your filibuster, but simply saying "you're wrong" is just as bad as calling people names.


And you're dong it again. Why call it a "filibuster?" That's a word loaded with negative connotations. Why not "argument" or "train or thought"? You may as well call it a "diatribe" or "harangue".

One has to explain why and how a person is wrong. If they interpret that as insulting, then it says a lot more about them than it does about the person correcting them, don't you think?

How did I put it?

"But if someone insists they want realistic, natural dragons but insists on going with a design that doesn't work, oh well. You tried. The answer is 'You're wrong.'"

"You're wrong" wasn't the opening and closing argument with nothing in the middle, it was shrugging and moving on. The elaboration on particulars would have come before it.

Teleros said...

Jollyreaper: "But go on, tell me about how well our professional, best trained military in the world is doing in Afghanistan and Iraq. We've fought the war so well it's been going on for ten years! But wait, I forget. Are wars the kind of thing where the length it goes on for is a sign of the quality or am I getting it mixed up with something else?"
1. Given the huge disparity between US / NATO equipment & training and Iraqi in Gulf War 1 & 2... it's not really surprising that we streamrolled them. Also, deserts make for much nicer tank battles than SE Asian jungles. That said... there were definitely great improvements in the US re how wars were fought thanks to the experience of Vietnam.

2. As far as Iraq & Afghanistan go, we seem to be doing okay (esp in Iraq). There have been difficulties, yes, but with people like Gen. Petraeus leading the US Army, the learning curve's been quite manageable. As far as the timeframe goes... well, we're trying to build one new nation and rebuild another, out of blood, sweat and tears. Those are not 9-5 jobs: if you take the historical view, it's amazing that the US & its allies have done so much at so little cost. It took decades and a lot of lives for the British government to beat the IRA, and Afghanistan will certainly be (and is) tougher than Northern Ireland. It can be done though, and it's still nothing like the mess called the Vietnam War.

I should also point out that we're not at war in Iraq (since Saddam fell), and arguably not in Afghanistan either, if you consider that we toppled the previous regime & installed a friendly new one.

Milo: "(Okay, it'll help if it's a cute machine.)"
For morale purposes, all military personnel are to be given a Companion Cube...

Tony: "Things wouldn't have been so jolly had we had to actually fight an undemoralized army with a secure rear."
Given the qualitative differences though, the Iraqis would still have been steamrolled. To give one simple example, US tanks could take out an Iraqi tank when the latter wasn't even in range.

Thucydides: "To get away from the Americans for a moment, consider the Eastern Front of WWII."
Rule 1 of warfare: you do not march on Moscow.
Rule 2 of warfare. you DO NOT march on Moscow.
They understood this in the Crimean War, but Hitler forgot about what Napoleon taught everyone.

Lentulus: "a far better place to work than the executive-elite-driven financial system we have today."
That's more to do with short term-ism than anything else. I believe the Germans & Japanese, for example, often take a much longer view than their Anglo-American counterparts.

Tony said...

jollyreaper:

"And you're dong it again. Why call it a "filibuster?" That's a word loaded with negative connotations. Why not "argument" or "train or thought"? You may as well call it a "diatribe" or "harangue"."

I originally wrote "diatribe", but I substitutued "filibuster" in the sense of "author filibuster", for the sake of accuracy. From tvtropes.org:

"An Author Filibuster is the extreme anvilicious case of Writer On Board, where the plot stops dead in its tracks to give the author an opportunity to preach their message to the readers or audience, often very political or ethical in nature."

Isn't that essentially what you were doing?

"'You're wrong' wasn't the opening and closing argument with nothing in the middle, it was shrugging and moving on. The elaboration on particulars would have come before it."

Fair enough. But I think you tend to interpret the elaboration as insulting. Nobody has called you "stupid" or an "ass", and only obliquely referred to what they might interpret as your ignorance. IMO, you're a very smart young (I believe) man who has unfortunately been exposed to a lot of horseradish. And that's all I think. Period. Anything else you see is of your own manufacture.

Tony said...

Teleros:

"Given the qualitative differences though, the Iraqis would still have been steamrolled. To give one simple example, US tanks could take out an Iraqi tank when the latter wasn't even in range."

They would have been defeated in relatively short time. But it would have cost a lot more in terms of ground troop casualties, mostly US. Not all of the fighting was done with tanks, on either side. There would have been some serious infantry battles in southern Kuwait for at least a few days. I have personal reason to be thankful that never happened.

jollyreaper said...

Rule 1 of warfare: you do not march on Moscow.
Rule 2 of warfare. you DO NOT march on Moscow.


You got the order wrong.

1. Never get involved in a land war in Asia.
2. Never go in against a Sicilian when death is on the line!
3. you do not march on Moscow.
4. you DO NOT march on Moscow.

fixed :)

Byron said...

I originally wrote "diatribe", but I substitutued "filibuster" in the sense of "author filibuster", for the sake of accuracy. From tvtropes.org:

"An Author Filibuster is the extreme anvilicious case of Writer On Board, where the plot stops dead in its tracks to give the author an opportunity to preach their message to the readers or audience, often very political or ethical in nature."

Isn't that essentially what you were doing?

Did you even bother to really read his statement? This reply is a shining example of exactly what we're asking you to stop.
First, he's not writing a book. He's asking you to stop treating the rest of us as ignorant if we don't share your views. So no, it wasn't a fillibuster.
Also, don't link there.
Second, he specifically said that diatribe was no better. Your use of those sorts of terms is exactly what the problem is.

Fair enough. But I think you tend to interpret the elaboration as insulting. Nobody has called you "stupid" or an "ass", and only obliquely referred to what they might interpret as your ignorance. IMO, you're a very smart young (I believe) man who has unfortunately been exposed to a lot of horseradish. And that's all I think. Period. Anything else you see is of your own manufacture.
That is the problem. You aren't going "anyone who disagrees with me is an idiot." You instead imply "anyone who disagrees with me is ill-informed." It's only marginally less insulting, and it also makes no sense. Why is all of your experience and knowledge relevant, and nobody else's is? If you really want, I can provide lots of examples of this attitude.

Tony said...

Byron:

What we're dealing with here is something that us programmers call an "impedence mismatch". I'm older than most people here -- significantly older than a few. I've been places and done things most people here haven't. I'm naturally and, in my opinion, correctly going to view a lot of disagreements as a matter of ignorance. And I'm not going to whitewash it when I think that's the case. I was never shined-on by anybody I ever respected, and I wouldn't expect anybody here to respect me if I shined them on.

What would you (and jolly and anybody else) have me do, specifically? Not tell you when I thought you were wrong? Not tell you why I thought you were wrong? Not tell you how I thought you were wrong? Not use accurate terminology because it has connotations you don't like?

Tony said...

BTW, when somebody don't know the correct technical language, don't know even the standard interpretations -- much less all of the variant interpretations or think a variant is standard -- and when he is told that insists that the teller is "grasping at straws" how is that supposed to be addressed politely?

That's a serious question -- how would you tell somebody they don't know what they're talking about?

Tony said...

Anyway, I don't want to argue, I know Rick doesn't want us to argue. So I'm willing to have done with this right here and now, and I'll try some different approaches to saying what I have to say.

Byron said...

Tony:
Simply, my problem is that you always assume you know what you are talking about. On some things, like computers, you're probably right (I don't know for sure that you're not lying, but I'm almost positive that's not the case). On others, like what space warfare will look like, we're all ignorant to a large degree.
The purpose of discussion is to arrive at the truth, which involves either being absolutely sure that you already posess (say, something you have a degree in or do for a living) or being willing to admit that you might be wrong. You seem to believe you always fall in the former category, even though nobody offers degrees in space warfare. And service in the Marines isn't enough. At least I don't think so, and I'm pretty sure the rest of the commentators agree with me. And it certainly doesn't qualify you to talk as an expert about high-powered laser systems, particularly ones that are, by nature, speculative.

Tony said...

Re: Byron

To bring this somewhat back on track, I fully realize that a lot of what is discussed here is sepculative. That's why, you may recall, I've objected to too high a reliance on numbers -- the assumptions aren't grounded in known facts.

But here is where I think the biggest misunderstanding arrises. Because I don't accept that hard numbers can be attached to these speculations much beyond making some judgements about what is clearly out of the picture -- like gigajoule lasers in softball sized satellites -- I think we have to rely on historical military experience. Leave the specifics of technology out of it -- what happens when we apply the logic of strategy? What happens when we think in terms of economics? How do politics effect all of this? We have plenty of data to answer those questions (yes, even the last one) and we simply aren't being honest with ourselves if we don't include those answers in our analysis.

So I'm always going to bring those issues up. They matter. No matter what the technology of the future brings, strategy will still be strategy, economics will still be economics, and politics will still be politics. So what is the reason that some people refuse to engage those issues? I ask you?

jollyreaper said...


I originally wrote "diatribe", but I substitutued "filibuster" in the sense of "author filibuster", for the sake of accuracy. From tvtropes.org:


Oh, you used a synonym that's just as pejorative. That's completely different. Except wait, no, it's not!

"An Author Filibuster is the extreme anvilicious case of Writer On Board, where the plot stops dead in its tracks to give the author an opportunity to preach their message to the readers or audience, often very political or ethical in nature."

Isn't that essentially what you were doing?


You really don't see what you're doing here, do you. And stopping the plot in its tracks? The very format of this blog is a conversation, a debate. There's no plot to interrupt.

Let's quote more from the page.

If the work goes beyond anvilicious into hectoring lectures, then it has become an Author Filibuster. Note that some works are openly intended to hammer home points, and are essentially teaching material in literary form: fairy tales, religious works, and position papers of all sorts may be heavy-handed, but that doesn't make them anvilicious. To achieve that distinction, the reader has to experience the sense that the author is foisting opinions, in the guise of telling you a supposedly entertaining story - and doing it clumsily enough that it becomes uncomfortable or irritating. Similarly, it is not anvilicious only because you disagree with any inherent message.

And you fail to grasp how this might be taken by some as insulting?

Anonymous said...

jollyreaper said:"But go on, tell me about how well our professional, best trained military in the world is doing in Afghanistan and Iraq. We've fought the war so well it's been going on for ten years! But wait, I forget. Are wars the kind of thing where the length it goes on for is a sign of the quality or am I getting it mixed up with something else? "

You are confusing politics with combat; the military is directed by civilians in this country, just on the off chance you've forgotten. Please don't insault those of us who have been in the militay with comments like that. Thank-you.

Ferrell

Tony said...

Re: jollyreaper

mea culpa, mea culpa, mea maxima culpa

I'm not going to continue the argument. There's no point in it. Rick doesn't want it. It shouldn't have gon on as long as it did.

Tony said...

Ferrell:

"You are confusing politics with combat; the military is directed by civilians in this country, just on the off chance you've forgotten. Please don't insault those of us who have been in the militay with comments like that. Thank-you."

I don't think he means to be insulting.

Byron said...

Tony:
I fully realize that a lot of what is discussed here is sepculative. That's why, you may recall, I've objected to too high a reliance on numbers -- the assumptions aren't grounded in known facts.
This is what I don't get. If we can't make certain assumptions, then we have nothing to debate. It's like debate nihilism. However, you do make assumptions as much as any of us. Your problem with specific assumptions seems to only exist when we have numbers in the assumptions.

I think we have to rely on historical military experience. Leave the specifics of technology out of it -- what happens when we apply the logic of strategy? What happens when we think in terms of economics? How do politics effect all of this? We have plenty of data to answer those questions (yes, even the last one) and we simply aren't being honest with ourselves if we don't include those answers in our analysis.
You make it sound as if technology matters not at all. That is completely wrong. Technology will drive all of the others to a large extent. Let's look at strategy and logistics of a war in Germany. In the thirty years war, logistics was mostly concerned with being able to pillage sufficient supplies from the local area. That dictated strategy in that you had to keep the army moving, but you didn't really have a base.
In World War II, a massive logistics train was required. You had to keep your base safe, but you could keep your troops fed anywhere.
The question is which do I draw analogy to?
The answer is obvious in this case, but the difference is because of technology.

No matter what the technology of the future brings, strategy will still be strategy, economics will still be economics, and politics will still be politics. So what is the reason that some people refuse to engage those issues? I ask you?
Those things will be altered by technology. If the cost of interplanetary shipping comes down enough due to technology, then the entire economic picture of colonization changes. If I have nanofactories and solar power that allows my army to live off the land, how much different will strategy look? Sieges are no longer possible, nor is the larger version of starving an enemy out.
Plus, you often engage in debates on technology. Please stop being inconsistent.

jollyreaper said...

I'm not going to continue the argument. There's no point in it. Rick doesn't want it. It shouldn't have gon on as long as it did.

And two minutes later.

I don't think he means to be insulting.

o_o

*pause*

"I see what you did there."
\
-_-

Anonymous said...

Ok, sometimes even I get upset with the overely passionate debates we have, (and coupled with a bad day), leads me to state things badly. Perhaps we should get back to the names of things...like the proper use of the terms 'gizmo, gadget, widgit, thangamabob, and whatchmacallits'. :)

Ferrell

jollyreaper said...

Perhaps we should get back to the names of things...like the proper use of the terms 'gizmo, gadget, widgit, thangamabob, and whatchmacallits'. :)


Or not. BLOOD FOR THE BLOOD GOD! SKULLS FOR THE SKULL THRONE!

Anonymous said...

=Milo=



SA Phil:

"Once you decide that it no longer matters if you persecute "innocent" people - its a whole lot easier to get a computer program to do it."

An oppressive government would be okay with harming some amount of innocents as "bycatch" when hunting for the guilty ones, but you still have to provide tolerable standards of life for the majority of your populace, or your civilization will collapse. "Kill everyone and let God sort them out." doesn't work on a national scale.

Tony said...

Byron:

"This is what I don't get. If we can't make certain assumptions, then we have nothing to debate. It's like debate nihilism. However, you do make assumptions as much as any of us. Your problem with specific assumptions seems to only exist when we have numbers in the assumptions."

I don't think it's making an assumption to suggest that long and widespread historical experience should be valid in analyzing the human future. When humans stop being human, then we enter a whole 'nuther realm of speculation. If you mean other assumptions, could you refresh my memory. I really am trying to be consistent, and i'd like to know what assumptions you think I'm making that are inconsistent.

As for making specific technical assumptions, go ahead. I've said more than once a lot of things depend on your technical assumptions. But any technical assumptions can only be valid within their contexts. Change the context, you change the valididity of the assumption, perhaps invalidate it altogether.

"You make it sound as if technology matters not at all."

As Mrs. teller used to say in trig class, "No, no, noooo!" I don't deprecate technology one bit. What I do think is that no technology, ever, has invalidated what Luttwak called the "paradoxical logic" of strategy. I see no reason to believe the future is any different.

Similarly, whatever technology you have, it's application is going to be governed by economic reality. If your country can't afford supercarriers, your navy is going to have to figure out how to do without them. Referring to your example, if transportation costs too much, then your army has to live off the land.

Finally, politics governs all, since war is about politics. Maybe you can really afford a supercarrier or two, but the people won't vote the navy the money. Or maybe the dictator wants submarines and tanks rather than a high seas fleet. Referring again to the early modern example, perhpas we campaign through Saxony rather than Bavaria, because even though Saxony is our ally, and campaigning there won't be popular, we have interests to secure there that an invasion of Bavaria, our enemy, won't secure.

Yes, technology is significant. It's a filter through which a lot of decisions are prcessed. It's the means to the ends. But it is still just a tool. Focusing on it exclusively misses most of what war is about.

"Plus, you often engage in debates on technology. Please stop being inconsistent."

See above. Technology isn't an unimportant or uninteresting topic. But it has to be kept in its place. Many students of WWII, for example, concentrate on technology and tactics. They can tell you everything about guns and armor and engines and bombs and shells and whatever else. But they simply can't fathom the larger picture. I've heard it suggested that the German u-boat campaign was successful because the u-boats certainly sunk a lot of tonnage, especially in the beginning. When you tell them that it was a failure because it didn't decisively effect the Allied effort in the long run, they howl epithets at you and accuse you of dishonoring the 30,000 or so u-boat crewmen who died.

See, it's all small and sentimental to them. Economics, strategy, and politics are too large, too impersonal, too absent fire and fury. But those are what war is really about. The fighting and technology are important and shouldn't be ignored. But they shouldn't be the focus either. A well rounded student, I think, should be able to talk about all of those things, and be able to apply his knowledge of them to speculations about the future.

Tony said...

jollyreaper:

"'I see what you did there.'"

Sigh...I honestly don't think you mean to be insulting when you deprecate the service and its culture. I think you really believe what you say. At least I'm willing to give you credit for sincerity. I'd appreciate such credit likewise in return.

Can we be done now?

Rick said...

I tend to agree with Tony on the substance of this debate, but - to be blunt - with Jolly and Byron on the tone.

It is getting to the point of just being not much fun to deal with some of these threads. Blogger probably has tools I can use to block people from commenting.

I really don't want to have to figure out those tools.

Anonymous said...

Tony,

As for making specific technical assumptions, go ahead.
==================

Ill attempt to be polite about this. But this is what you tend to not allow for.

You basically will automatically shoot down *any* technology assumptions that are not 20 minutes into the future -- other than fusion tech.

For example on multiple occasions in this thread you shot down out of hand my assertion that computers will largely program themselves 200 years from now.

Even though I pointed out some modern systems that do just that (I know they do; I have set up and then driven the results).

You might not agree that is the route it would take - but you instead basically argued that it is not possible to do. Which is a pretty big thing to say when I am talking about 200 year out computer tech.

Now you may assume I am ignorant on computer technology. You would be entirely wrong. But you are free to assume what you will.

(SA Phil)


------
PS - I forget what the trope is called, but its the one where only One Technology advances in a Sci Fi story - everything else stays pretty much todays tech.

Tony said...

Re: SA Phil

What I see happening WRT software is that we're talking about two different things. You're talking about your experience with a system of templates designed to stramline the programming of standardized hardware. I'm talking about developing fundamentally new algorithms for fundamentally new applications.

Taking a template, adding modules (which are in themselves also templates), setting all of the properties in all of the modules and perhaps in the master template, then telling it to go off and write code is not what I would call programming. The design and implementation of the templates and modules was programming, and it was done by people, not machines.

No matter how much abstraction you pile on top of the machine code, no matter how many standardized tasks can be handled by code writing automatons, there will always be new applications that some human will have to develop and implement the logic for, because there will always be new machines that nobody has ever written any software to control.

I can't be any clearer than that. So what do you disagree with, specifically?

----------

WRT technology assumptions in general, I don't criticise assumption simply because they are assumptions. I criticise reasoning from the specific to the general, which seems to happen a lot.

Anonymous said...

Tony,

I can't be any clearer than that. So what do you disagree with, specifically?
==========

What I disagree with:
That Space Travel and Space Combat will be in the "unique" category you specify here.

My assertion is that in 200 years very few "control system" tasks will be in that unique category.

I'll go further and say in 50 years embedded controllers across the board will need less than 10% of the human involvement they do now.

-------------
What assumption you appear to have made:
That I didn't understand you before. I understood you clearly in your other posts. I did not agree with your conclusions.

====================
--------------
Other Technologies you have quickly and adamantly shot down off the top of my head, specifics or no:
*Mid-future Bio Fuels
*Laser Stars
*Limited crew/no crew space craft
*Any standardization of military equipment beyond Today's level.
*Anything more than very limited sharing of civilian and military systems/technologies.
*The use of low thrust propulsion in combat.
*Large Nuclear Fission Reactors in Space.
*Any substantial increase in space lift capacity that does not involve magi-tech fusion.

Your definition of "Plausible Midfuture" sometimes comes across fairly strongly as "2012 AD + sometimes Fusion"

(SA Phil)

Anonymous said...

(SA Phil)

Clarification -
"Unique" = "New Applications"

Also what you call coding or programing and what I called coding or programing is really meaningless semantics.

Use "Set Up" instead - the distinction does not matter.

Build a spacecraft, add all your modular goodies:

Turn the computers all on.

They will do all the set up work for you. You will not be doing any "new applications" in that location. They will not be needed.

Your missile launcher will come equipped with guidance software. It will update your control room computers to handle fire control.

Your Laser Optics will come equipped with fully adaptable software needed to run them. You will do no integration.

If there are any hiccups, solution software will be available on whatever passes for the internet. The ship's computers will download it and implement it.

Thucydides said...

Looping back ever so slightly, technology simply gives us the tools to carry out strategy and tactics in the military arena, and economics provides the means to get these tools. Politics dictates to what end economic power is harnessed to strategic and tactical ends (operational too, if you are a student of Soviet military thought...).

Consider two invasions of Germany.

The Roman legions who were destroyed in the Teutoburg forest were not there to check out Oktoberfest. The Emperor wanted to extend his reach into Germania to secure the northern borders and to extract any wealth that might be available. Enough money was raised through taxes to raise and equip several legions. The legions were equipped with the most economical equipment that could be procured and trained and led following well known procedures which worked well. Unfortunately, the commander was not very cautious and led his men into a trap, negating the superior training and equipment of the legion.

General Patton led an invasion of Germany to end an existential threat to the western democracies as part of a larger team. His government raised and equipped an army, but chose to use relatively simple, mass produced tools to ensure there was lots of "stuff", even if some of the competing technology was better. Patton looked at the well known tactics and procedures that his contemporaries used, and realized they could be applied in different ways, resulting in his Army making spectacular inroads into enemy positions and unhinging a large portion of the German defense. However, both internal politics (the need to ensure all the Allied field commanders were happy, or at least satisfied) and external (agreements between the Great Powers) limited many of his options and prevented him from achieving all he hoped for.

Jim Baerg said...

Note re: the digression on Dragons.

If you want flying animals much larger than the biggest birds on earth, you put the story on a planet with some mix of lower gravity &/or thicker atmosphere.

See http://www.worlddreambank.org/P/PLANETS.HTM
for some examples.

Tony said...

SA Phil:

"What I disagree with:
That Space Travel and Space Combat will be in the "unique" category you specify here.

My assertion is that in 200 years very few "control system" tasks will be in that unique category.

I'll go further and say in 50 years embedded controllers across the board will need less than 10% of the human involvement they do now."


Let me get this straight...you're saying that all possible sensor and control technologies for spacecraft will somehow converge? My fire control radar will run on the same hardware, with the same software, that my search radar does that my navigation radar does? The brains in my missiles and laser guns will come out of the same tech tree?

"Other Technologies you have quickly and adamantly shot down off the top of my head, specifics or no:
*Mid-future Bio Fuels"


On cost-benefit grounds, especially on the food vs fuel issue.

"*Laser Stars"

Are highly sensistive to technical assumptions. (This is primarily what I was referring to when I said reasoning from the specific to the general.) They are also a species of overspecialized weapon system. Such systems rarely, if ever fare well in actual combat.

"*Limited crew/no crew space craft"

Also highly sensitive to technical assumptions.

"*Any standardization of military equipment beyond Today's level."

???

"*Anything more than very limited sharing of civilian and military systems/technologies."

Military vehicles and systems of all types have different perfromance requirements than civilian ones. Convergence always has been limited, is limited today, and I see no reason why it wouldn't be limited in the future.

"*The use of low thrust propulsion in combat."

In combat, yes. Not in interplanetary travel between orbital combat spaces. Once in orbit, you need high performance propulsion, because you can't take days or even weeks to make tactical orbital changes.

"*Large Nuclear Fission Reactors in Space."

They aren't economically feasible with chemical rocketry launch vehicles.

"*Any substantial increase in space lift capacity that does not involve magi-tech fusion."

There's no plausible market for any of the megastructural solutions that have been proposed.

"Your definition of 'Plausible Midfuture' sometimes comes across fairly strongly as '2012 AD + sometimes Fusion'"

I would say economically realistic and no special pleading for anybody's favorite technology or technological context. I don't think technology will fail to advance. I just don't think it will advance as quickly or in the directions that many seem to think.

Tony said...

SA Phil:

"Clarification -
'Unique' = 'New Applications'"


That's not correct.

"unique" = "any configuration of software and hardware that nobody else is using"

That's quite common with military systems.

"Also what you call coding or programing and what I called coding or programing is really meaningless semantics."

Except that semantics are never meaningless in software. Your semantics define the logical operations that you can undertake.

"Use "Set Up" instead - the distinction does not matter.

Build a spacecraft, add all your modular goodies:

Turn the computers all on.

They will do all the set up work for you. You will not be doing any "new applications" in that location. They will not be needed.

Your missile launcher will come equipped with guidance software. It will update your control room computers to handle fire control.

Your Laser Optics will come equipped with fully adaptable software needed to run them. You will do no integration.

If there are any hiccups, solution software will be available on whatever passes for the internet. The ship's computers will download it and implement it."


What you're advocating is a more static technological future than anything you've accused me of. For it to work, every system would have to have known operational parameters, the ships central supervisor system would have to have standard a software package for integrating any kind of weapon, in any configuration, and no new capabilities would ever have to emerge. And the idea that an open source diagnostic and repair package for military systems would be freely downloadable from the internet is, well...unbelievable.

Anonymous said...

(SA Phil)

"Whatever Internet" = Military Intranet, Skunkworks secure connection, Whatever.

I never specified any level of security there.

---
I made the "unique"="New Applications" comment because I paraphrased you to "unique" which in retrospect isnt the term you used so I was attempt to correct my previous post.

---
===========

In a way I *am* arguing there will be "more static" future in software. I am postulating that software will advance to a level of ubiquitous inconnectivity and compatibility (forwards and backwards)that "custom" systems will literally be able put themselves together.

This static future of course doesn't happen until the software advances (and the hardware to facilitate them) many levels past where we are now.

=========

I am saying that if you want to put a functioning missile launcher on your warship --- there will be an App for that.

Anonymous said...

Tony,

and no new capabilities would ever have to emerge

---------------

This part is patently untrue. You would instead need a system sophisticated enough that it could make allowances for new capabilities.


As to unbelivable - I disagree. In fact it is the direction many embedded control systems have been slowly heading for the past 10-15 years.

(SA Phil)

Anonymous said...

(SA Phil)

A lot of people didnt beleive "plug and play" operating systems would work either.

Tony said...

SA Phil:

"'Whatever Internet' = Military Intranet, Skunkworks secure connection, Whatever.

I never specified any level of security there."


Okay...where does the "solution" software come from? It can't just write itself. It has to be based on known problems with known parameters, then address them specifically. Somebody has to analyze the problem and then design and implement the fix. Dsitributing it over a network doesn't change that.

"In a way I *am* arguing there will be "more static" future in software. I am postulating that software will advance to a level of ubiquitous inconnectivity and compatibility (forwards and backwards)that "custom" systems will literally be able put themselves together."

Ubiquitous interconnectivity already exists, if you're willing to put up with certain interface convention inefficiencies. XML, which is what most interconnectivity standards are basedo n in some way, is self-describing but ridiculously verbose, especially when transmitting numerical data. That's why various industries still stick with specialized messaging formats.

"This static future of course doesn't happen until the software advances (and the hardware to facilitate them) many levels past where we are now."

It can't advancei n the direction you think it will go, simply because no matter how standardized interface conventions become, you still have to tell your supervisory process what to do with incoming data and how to respond. I spend all day, every day, doing that kind of work. It's not going to go away, because, for the umpteenth time, computers cannot program themselves. They can be given programs to automatically generated code under certain well-defined circumstances, but that's no different than dynamic HTML generation on databased web sites (something I also work on every day). Autocoding is just templated text generation or, if you're writing directly to machine code, templated generation of long lists of binary numbers.

"I am saying that if you want to put a functioning missile launcher on your warship --- there will be an App for that."

There already is an app for that. Part of the contract for the system includes the supply of all necessary software. But people in the contractor's shop had to write that software.

"This part is patently untrue. You would instead need a system sophisticated enough that it could make allowances for new capabilities."

You would need a system sophisticated enough to make allowance for any plausible new capability, using any plausible data format.

"As to unbelivable - I disagree. In fact it is the direction many embedded control systems have been slowly heading for the past 10-15 years."

Jeepers. What has happened with embedded controls is that they have come to be almost as standardized as fasterners. That allows companies like ETAS to write standardized template software solutions for them. These enable non-progammer engineers to set up (you got that part right) controller modules for various tasks, without requiring software engineers to apply the same kinds of tweaks in the same places of multiple instances of a common program. It's a somewhat more sophisticatd version of programming the VCR.

But at some point some programmer has to analyze each controller, catalog its imputs and outputs, figure out all the possible execution pathways, and write a template program for it. As long as new hardware models and configurations are introduced, somebody is going to have to do the same task. Somebody is going to have to accomplish the three basic programming tasks:

Define input rules
Define processing rules
Define output rules

Aaa-gain, computers can't do that for themselves.

Tony said...

SA Phil:

"A lot of people didnt beleive "plug and play" operating systems would work either."

Who are these "[a] lot of people"? The major problem with plug-n-play was agreeing on standards. Everybody wanted to do it, because a less technical computer experience was only good for sales, both of the computers themselves and aftermarket hardware.

What you seem to not comprehend about plug-n-play is that it still requires a device to conform to the applicalbe interface standard, in both hardware and software terms. It also has to have driver software for the operating system to invoke, because the operating system can't be expected to know how each new piece of hardware works. A lot of hardware nowdays (particularly USB thumb drives) has the driver onboard in a ROM, and operating systems have been updated to discover and load the driver.

So where you think we're going in 200 years is already here -- interface conventions, autodiscovery of device driving software, and auto-configuration. Yet we still need programmers to write all of this software that makes things go. The computers can't do it for themselves.

Perhaps military systems will one day be equally plug-n-play. But none of the technical details are likely to have changed that much, not matter how super-wonder-fabulous our computing resource pool becomes. As previously stated, somebody is still going to have to analyze the hardware and define it's operating logic in software. Computers can't do this for themselves.

M. D. Van Norman said...

“At least the computer wouldn’t be jumping chamber maids in its hotel room.”

Funny. I’ve been tinkering with a story concept that would open with a virtual tryst between an AI and a human.

Transhuman AIs are people too, and they have their needs. :-P

Anonymous said...

=Milo=



SA Phil:

"If there are any hiccups, solution software will be available on whatever passes for the internet."

Sure, maybe, but someone still needs to have written that software.


"The ship's computers will download it and implement it."

Automated patches? On what's supposed to be a secure system? Really?

Byron said...

Disclaimer:
I actually have to agree with Tony in the debate over computers. Admittedly, I'm not a programmer, but what he's saying makes sense.
(And the world didn't end when I wrote that.)

Tony:
Are highly sensitive to technical assumptions. (This is primarily what I was referring to when I said reasoning from the specific to the general.) They are also a species of overspecialized weapon system. Such systems rarely, if ever fare well in actual combat.
Funny, "highly sensitive to technical assumptions" didn't come across very well. It sounded more like "won't happen due to technical issues."
Even then, could you stop saying that. There's a fine line between highly specialized and overspecialized. Nobody has seriously proposed a laserstar with only a single main laser recently. We do generally discuss it, but that's because it's the main offensive weapon. Most discussion of tank armament centers around the main gun for similar reasons.

Tony said...

Byron:

"I actually have to agree with Tony in the debate over computers. Admittedly, I'm not a programmer, but what he's saying makes sense.
(And the world didn't end when I wrote that.)"


But my heart went pitter-pat...

"Funny, 'highly sensitive to technical assumptions' didn't come across very well. It sounded more like 'won't happen due to technical issues.'"

They're not mutually exclusive conditions. A concept can be highly sensitive to technical assumptions and at the same time require the most favorable technical assumptions to be true. Reasonable people can believe these favorable assumptions to be unrealizable.

"Even then, could you stop saying that. There's a fine line between highly specialized and overspecialized. Nobody has seriously proposed a laserstar with only a single main laser recently. We do generally discuss it, but that's because it's the main offensive weapon. Most discussion of tank armament centers around the main gun for similar reasons."

Ahhh...tanks. We (as in the disembodied, rhetorical "we", not anybody in particular) always invoke tanks when we want to make a point about the viability of highly optimized systems. The problem is that they are highly optimized, and vulnerable to just about everything except inferior tanks and poorly equipped infantry. They are also marginally effective against a lot of battlefield targets. At one point not that long ago the M1 Abrams, as a system, was so optimized for fighting tanks that its only antipersonnel weapons were machineguns and HEAT rounds used as field expedient HE. That has since been remedied, but the M1 is still not the best system for most ground combat requirements.

If I was going to make the analogy between tanks and laserstars, I would say that laserstars, just like tanks -- and regardless of being equipped with secondary weapon systems -- are optimized for fighting other laserstars. This renders their utility questionable, IMO.

Anonymous said...

Tony

What you seem to not comprehend about plug-n-play is that it still requires a device to conform to the applicalbe interface standard, in both hardware and software terms. It also has to have driver software for the operating system to invoke, because the operating system can't be expected to know how each new piece of hardware works. A lot of hardware nowdays (particularly USB thumb drives) has the driver onboard in a ROM, and operating systems have been updated to discover and load the driver.

So where you think we're going in 200 years is already here -- interface conventions, autodiscovery of device driving software, and auto-configuration. Yet we still need programmers to write all of this software that makes things go. The computers can't do it for themselves.
=========

No - What you assume I dont understand about plug and play is (stuff)

---

Please dont make the mistake for the umpteenth time of taking your *interpretation* of my statements and then thinking I don't understand the process.

I do understand there will be people involved in bringing things to this state -- In fact I said it twice in this thread. Three times now.

I just dont think any of them will be involved in *directly* setting up a specific spacecraft's systems.

As in they will never step onto said spacecraft and press a single button. Nor will any of those humans do anything remotely either. Installing your missile launcher will be easier from a systems integration standpoint than installing a new video card is today.

The humans will instead work in computer R&D. Developing the necessary software infrasture.

You on the other hand seemed to say there will be software experts on board the ship reprogramming systems essentially "on the fly" not only during the construction of the ship but what it seemed to be during operations.

While that makes great drama in a sci fi story - I do not think that will actually be happening.

Instead the computers' systems interfaces will be sufficiently sophiticated enough to make any accomidations to the actual set up of the space craft.

(SA Phil)

jollyreaper said...

Tanks to me are like aircraft carriers; I have a suspicion that they're really vulnerable, more vulnerable than anyone cares to admit to, but we won't really be able to prove it one way or another without a major war.

Aside from the "show the flag" ability with a carrier, both weapons share very successful histories tied to winning major wars. Lots of military men have made great careers with the weapons, we have a military-industrial complex dedicated to their manufacture, care and feeding.

Now everything I said above could be equally true for obsolescent weapons concepts that have outlived their usefulness but could also be said for weapons that remain incredibly lethal, effective, useful, and completely justify their existence a hundred times over.

Now to consider them separately.

The tank is optimized to dealing with other tanks. With all of the other threats on the battlefield than can make tanks vulnerable, what's the justification? Not a challenge here but a query. Is a tank truly the most effective way of engaging other tanks, especially on the move? What penalties does the deploying army entail with logistics and cost? Does a tank represent using a 70 ton solution when a 10 ton solution would do? Could we be seeing a loosely analogous situation to the battleship where you needed something that big to carry the guns, armor, and powerplant to be a ship-killer but now we can pack the same killing power into a destroyer hull or under a wing with guided missiles?

The argument for the carrier is that it's about power projection and you can bring your airfield with you, not relying on borrowing real estate in-theater. And air power is the premier advantage America brings to the battlefield, we can't do without it.

My personal guess is that there's a lot of time and money tied up with the current doctrine that's been proven in battle and the commanders are loathe to throw out combat-proven concepts on the supposition that they might have become outdated and obsolete.

Stuff I've read has gone back and forth about whether or not the tank is obsolete. The latest round of discussion says that active defense systems that shoot down incoming weapons will give tanks and armored vehicles a new lease on life on the battlefield.

There's a whole new class of vehicles the Army was looking to purchase back before the wars and they've scrapped and gone back to the drawing board a couple times. I hold about as much hope for any of those vehicles as for a NASA-backed shuttle replacement -- lots of money down the rat hole leading to a terminated program.

Tony said...

SA Phil:

"Please dont make the mistake for the umpteenth time of taking your *interpretation* of my statements and then thinking I don't understand the process."

Okay...

"You on the other hand seemed to say there will be software experts on board the ship reprogramming systems essentially "on the fly" not only during the construction of the ship but what it seemed to be during operations."

Uhhh...sorry, no. I said that I think it's prudent to have human supervisors monitor the systems in action, and take approriate measures when the systems are presented with unprecedented or unexpected events. Appropriate measures might be anything from nudging the computersi n the right direction, to exerting manual control. I didn't say anything about anybody on a ship reprogramming anything, either on-the-fly or at leisure. I said that after the engagement, surviving records of what happened would be forwarded to the system contractors for them to develop and deploy software upgrades.

BTW, the kind of programming involved in developing system software is not R&D. It's normal, everyday systems rpogramming.

Tony said...

Re: jollyreaper

At the moment tanks are the best way to deal with other tanks -- in any numbers, anyway. But in any other terrain besides a desert, you have to support them with infantry and artillery, and have at least conditional control of the air.

Future Combat System was a pig in a poke. It was betting on leap-ahead technology that never materialized. As for current developmental systems, well...nobody realy seems to care about them with any great passion. They may lead to something or they may not.

Anonymous said...

Tony,

BTW, the kind of programming involved in developing system software is not R&D. It's normal, everyday systems rpogramming.

==========
For someone who claims not to be caught up on semantics .. you seem to be pretty caught up on semantics...

I used the term R&D becuase I think there will be far less humans involved in the entire process than an equivlent system today.

So I envision the human involvement to be pushed closer to the design end.


(SA Phil)

Tony said...

SA Phil:

"For someone who claims not to be caught up on semantics .. you seem to be pretty caught up on semantics..."

To quote myself, what part of:

"Except that semantics are never meaningless in software. Your semantics define the logical operations that you can undertake."

...means semantics are unimportant?

"I used the term R&D becuase I think there will be far less humans involved in the entire process than an equivlent system today.

So I envision the human involvement to be pushed closer to the design end."


Fewer humans being directly involved is possible, but that doesn't meant they're doing R&D. It means that it takes fewer people to do the task because (presumably) they are working at a more abstracted level. But it's still taking a settled hardware specification and writing the code to make it do work. No research involved -- just analysis and design.

Anonymous said...

Tony,

But it's still taking a settled hardware specification and writing the code to make it do work. No research involved -- just analysis and design

=============

Nah the computer will write that code.

It will come up with it based on the computer model which was produced from the computer aided design of whatever the hardware was..

Which was put together by an engineer ..

Before you ever even build the first example of said hardware.

The programer will come in ahead of that process, setting up the CAD to Computer Model to Code ability.

I was calling that computer R&D, since it involved setting up the computer systems to do that -- but realistically its also pre-R&D from the Engineer's point of view.

(SA Phil)

Tony said...

SA Phil:

"Nah the computer will write that code.

It will come up with it based on the computer model which was produced from the computer aided design of whatever the hardware was..

Which was put together by an engineer ..

Before you ever even build the first example of said hardware.

The programer will come in ahead of that process, setting up the CAD to Computer Model to Code ability."


You can't write code for something that has no shape or performance specification. Even with some list of common functions dealt with by code generators working off of templates, the programmer still ahs to analyze the machines functions and design a system to schedule the execution of all of those functions, respond to events, and resolve conflicts.

There's no way to write software for the arbitrary situation. The hardware has to exist in least specification form -- which means you're past R&D and working on physical design. Then for the software builder it's analysis of the specification and design of the software system.

"I was calling that computer R&D, since it involved setting up the computer systems to do that -- but realistically its also pre-R&D from the Engineer's point of view."

There may be some amount of software R&D going on along with hardware R&D, but the software design cant be finalized and implemented without at least a semi-hard hardware spec.

Anonymous said...

Tony,

There's no way to write software for the arbitrary situation. The hardware has to exist in least specification form

===============
No way you or I are aware of perhaps.

However it is impossible to design anything that can not be described by mathmatics

Anything that can be described by mathmatics requiring software, can potentially be coded.

If you have a sophisticated enough system it will be able to make that transition.

(SA Phil)

Tony said...

SA Phil:

"No way you or I are aware of perhaps.

However it is impossible to design anything that can not be described by mathmatics

Anything that can be described by mathmatics requiring software, can potentially be coded.

If you have a sophisticated enough system it will be able to make that transition."


You're talking about a formal system in which all possible algorithms are rigidly defined, in all of their possible details, and the bounds of all possible domains are perfectly known, for all possible circumstances. That's the only way a non-creative entity -- IOW a computer -- could possibly automatically write software as a consequences of physical design.

To put it succinctly: no way, bra.

Anonymous said...

Tony,

To put it succinctly: no way, bra

-------
Not only do I think it will happen..

It will be routine, common place and boring.

And people wont even realize how good they have it.

No more than our kids think about how easy it is to press a button on a remote control and change the channel without having to get up, turn a dial, stand on one foot and adjust the clothes hanger at the back of the set.

Tony said...

SA Phil:

"Not only do I think it will happen..

It will be routine, common place and boring."


Repeat these statements out loud:

All possible algorithms will be known.

All possible algorithm parameters of all possible algorithms will be known.

All possible domains will be known.

All possible shpaes of all possible domains will be known.

An automaton will be programmable so that it can choose from all possible lagorithms for all possible domains and assemble a software system based on automated analysis of a physical design.

Now, do you really believe that?

Anonymous said...

Tony,

(Typical straw-man pigeonhole you should be famous for)

=================

Save it. Seriously.

That is not what would be required.

(SA Phil)

Anonymous said...

I have come to the conclusion that no one knows what the capabilities of computers will be 20 minutes from now, much less 20 years from now, and 200 years from now I suspect they will be unrecognizable, but for reasons we can't imagine.

Anyway; gizmos are stand-alone devices, widgits are subsystems of thanamabobs, and whatchamacallits is a technical term for the other guy's thanamabobs. :)

So, what are different types of spacecraft going to be called, PMF (90 - 500 years out)? Will solar electric freighters be called 'butterflys', will nuc electric manned logistics ships be called 'glowies', will chemfuel rockets be called 'burners'; or will they be called something else? Maybe in referance to their crew endurance? Or delta-V? or acceleration? Or maybe a combination of all of those? I tend to think that they will have official designations and unofficial nicknames for both the type and individual ships.
(YE-102SI Deep Space Transport, i.e. "Llamma", with tail number 77 being called "Doug's Cart" by it's crew)

Ferrell

jollyreaper said...

SA Phil,

I'm not certain I completely follow your line of conjecture. I think we might be in a situation like Charles Babbage talking about difference engines. It's like ok, we can talk about how wonderful a computing device would be and it's all nice and theoretical but people are scratching their heads about how one could be built or put into practice. Fundamental discoveries that would be required to make computers a reality are outside of the 19th century Plausible Mid-Future.

*IF* we develop strong AI, and by that I mean human-equivalent, I think that will be the end of programming. Alternatively, if we do *NOT* develop strong AI but find ourselves stuck in a prolonged decelerando or just a tremendous tech plateau, there remains the possibility of solving all the problems one is likely to encounter and 99% of people will simply be reaching for an off-the-shelf solution.

Now the weird area I don't understand all that well is the neural net, self-programming robots.

http://blogsci.com/science/robot-learns-to-walk

The key here is that the robot is learning to do this – the adaptation is not programmed. Thus it’s one model of the human neural networks that govern human walking.

There's a lot I can accept in scifi so long as the time horizon is far enough out. Terminator cyborgs? Sure, a minimum of a hundred years out. 20 minutes in the future? No way. Beanstalk space elevators? A hundred years out. Five years out, no way. Resurrected woolly mammoths? 20 years, sure, why not. 2012, too soon.

Computer tech is harder to predict. I don't have a strong feeling one way or the other if strong AI is possible, not qualified to hold an opinion. But I know I'd want at least 50 years out for it to even feel plausible if pitched in a story. I know I would have put self-driving cars 20 years out and Google's already demonstrated working models.

Rick said...

My take on computers is that computers 'as we know them,' based on some form of the familiar digital architecture, will always need to be programmed. They are not intelligent; they are purely projections of human architecture.

Strong AI would be a very different matter - by definition it does not need to be programmed, any more than we need to be programmed. But at this point we haven't the faintest clue how strong AI would actually work. It is not an extrapolable technology.

I could perfectly well buy a setting in which it is 3011, and we still have no strong AI - or a setting in which it is 2031, and we do have strong AI.

(Implicitly I'm guessing that if the key breakthrough is being made as I type this, it will take 20 years to mature into implementations that would fit our sense of strong AI.)


At the moment tanks are the best way to deal with other tanks -- in any numbers, anyway. But in any other terrain besides a desert, you have to support them with infantry and artillery, and have at least conditional control of the air.

Space, in some important respects, is like a desert only more so. That is not, as such, an 'argument for laserstars,' but it points at a way of thinking about the problem.

Anonymous said...

Jollyreaper,

I'm not certain I completely follow your line of conjecture.

=========

It goes a bit like this.

Ill use a Washing Machine example.

The engineer designs the washing machine on a Cad Drawing maybe even just a sketch. Or picks from thousands of existing designs to modify.

The computer will take that drawing and make a computer model. It will add a motor to make it spin, a control system to give it various speeds and cycles ,etc.

The Engineer will look at the Model and decide he wants a bigger motor, or an extra cycle, add time to the spin, etc.

The Motor could be off the shelf or the Engineer could even tinker with those parameters. Its just wheels within wheels. All the math to make these things work is basic physics the computer will just apply it.

The computer will take any adjustments and then make a new drawing.

A little back and forth and the Engineer decides the Washing Machine will work.

It might have designed circuit boards, but more likely it picks and off the shelf controller. It then programs the controller to do what the agreed on Model said.

It will be able to completely display every drawing involved in the process and once a working example is built, download the control system code to the washing machine and then literally when you hook it up and plug it in and it will work just like the computer model predicted.

===================
Why do I think it will work this way?

Because the infant steps of all of these things are already starting to be done on limited scales throughout design processes.

I mentioned standardized control "autocoding" from Logic diagrams. There are also Hardware in the loop systems available that do something similar.

Sophisticated working computer models have a lot of effort being poured into them in several fields. The driving force is that it will save billions of dollars in testing and R&D. Once you have a working computer model it is conceivable the computer would know what you want to control. The inputs and outputs will be defined.

You don't design memory/logic boards logic line by logic line anymore, there is software that basically draws them up depending on what you need done. Then it flashes the whole design onto a programmable chip and it is finished. There is no reason to think other electronic circuit boards wont follow in years to come.
==================

So all I am doing is projecting what these types of tools do onto a future with billions the times of computer hardware and software complexity as ours.

Essentially as long as you can get the computer model to do what you need, it will be able to come up with the specific code you need to control the real world variant.

The software engineers/programmers etc will be the ones "teaching" / "programing" w/e the computers to do these things, and there will be constant improvement over time.

200 years is a long time into the computer future. And the number is pretty arbitrary. Maybe it will be 500. Maybe parts of my projections will only be 50 years away.

We are only in computer infancy now. When I was a kid the computer room at Naval reactors was still using punch-cards.

I think its far *less* likely that systems hundreds of years from now will be set up and programmed essentially the same way they are today.

Anonymous said...

^^^^^
(SA Phil)

the previous post was mine BTW although I am sure by my ramblings most guessed that.

Anonymous said...

=Milo=



Jollyreaper:

"*IF* we develop strong AI, and by that I mean human-equivalent, I think that will be the end of programming."

Do you count programming done by the strong AI?

Remember, the moment something has strong AI, you need to start treating it as a character rather than a tool.



Rick:

"My take on computers is that computers 'as we know them', based on some form of the familiar digital architecture, will always need to be programmed."

It's like I said before. To get a computer to do something for you, you need to be able to explain to the computer what you want. Explaining to a computer what you want is, per definition, coding.

Future programmers might make use of a tool that automatically optimizes an algorithm for a given objective, but first you need to explain in terms the computer can understand what your objective is. That's coding.


"Strong AI would be a very different matter - by definition it does not need to be programmed, any more than we need to be programmed."

Until you want to make an even smarter AI, or tweak its personality parameters.

Again, it could be strong AIs programming other strong AIs, but there is still a clear benefit to someone doing some programming (although, as the example of humans shows, mass produced intelligence can be quite intelligent even if you don't have the understanding to fine-tune the details).

Anonymous said...

Err, that last bit should read "can be quite useful". Sorry.

Thucydides said...

The actual problem with the Babbage machine and the Analytical Engine turned out to be mid Victorian machine technology wasn't advanced enough to make the precision gears needed for the thing to run smoothly. The difference engine itself works (modern replicas exist built with modern machining techniques), and Lady Ada Lovelace had laid ot much of the structure that is the basis of modern programing. If an accurate gear cutter existed, we might be talking about the IT department as the "black gang", and your introduction to working in IT would be shoveling coal as a stoker or lubricating the gears.

How computers will evolve seems to be a bit of a mugs game; current computers are essentially brute force devices and supercomputers mostly rely on harnessing even larger clusters of processors and hard drives to the point of diminishing returns. Some sort of "Cambrian revolution" may take place where one or more breakthroughs allow radically different hardware and programing architecture(s) to replace digital computers. If holographic computers or quantum devices take over from digital computers, then our entire way of thinking about computers will also have to change.

«Oldest ‹Older   1 – 200 of 231   Newer› Newest»