Page 93

Discussion regarding the Outsider webcomic, science, technology and science fiction.

Moderator: Outsider Moderators

Absalom
Posts: 718
Joined: Sat Sep 24, 2011 4:33 am

Re: Page 93

Post by Absalom »

Rational races won't seriously consider the "von Neuman's locusts gambit", because the consequences of a failure (you lose the secret of the IFF, someone else gets the secret of the IFF, the locusts lose the secret of the IFF) are all bad enough that you want the locusts close enough that you can monitor and correct them.

Unless, of course, you're about to be unequivocally wiped out, in which case spitting in your enemy's eye wins out (dead is dead, after all).

User avatar
088
Posts: 22
Joined: Sat Apr 02, 2011 12:57 am

Re: Page 93

Post by 088 »

Absalom wrote:Rational races won't seriously consider the "von Neuman's locusts gambit", because the consequences of a failure (you lose the secret of the IFF, someone else gets the secret of the IFF, the locusts lose the secret of the IFF) are all bad enough that you want the locusts close enough that you can monitor and correct them.

Unless, of course, you're about to be unequivocally wiped out, in which case spitting in your enemy's eye wins out (dead is dead, after all).
How exactly would we lose the IFF? you would have a large number of human ships/independently mobile equipment, all with the same data i doubt you could loose a few thousand/hundred thousand copies all at once. not to mention the fact that a very important device such as that would probably have a lot of spares.
I can understand someone else figuring out the proper signals to keep the locusts off of them. (If the hypothetical IFF was something simple like a broadcasted radio signal) And that is excluding the fact that there wouldn't be many human ships around to use said IFF for them to observe. but they wouldn't give out said information even if they figured it out, and by the time you came back there would still be fewer races to deal with.
and another side benefit of the Von Neuman locusts plan is after you return, and those races who are going to get wiped out have. you can turn the things into basically space livestock. They go reproduce to a certain number then return to a location to be decommissioned and recycled into things a human population would need. (you'd just have to reprogram a few and give them the orders to pass along to other populations of locusts)
and during this period of self imposed exile we could spend our time establishing ourselves elsewhere.
building ships, increasing our population, workin on our tech and other such things. so that when we do come back it will be a lot easier to go toe to toe with anyone left standing.

Even in the event of failure this strategy leaves us better off then either slavery or death. At least barring any further unfortunate circumstances like we run away only to find ourselves in the middle of yet another alien war, or monstrous planet eating super weapons, madness inducing space anomalies ect ect ect

and who said anything about dieing? sure the humans left behind are trapped on there worlds (Best case scenario) or destroyed by the other races before the locusts become numerous enough to make a dent. (Worst case scenario) but you still have the population your leaving with in either case.
and if you didn't leave the human worlds any knowledge of the IFF device the other races couldn't get it from them thus further insuring security.

discord
Posts: 629
Joined: Thu Mar 24, 2011 7:44 am
Location: Umeå, Sweden

Re: Page 93

Post by discord »

the reason is shown quite nicely on stargate universe, and earlier with the replicators, if they don't work, they are a menace to you(and wasted resources), if they do work, they are probably a menace to EVERYONE.

TrashMan
Posts: 306
Joined: Tue Apr 05, 2011 2:01 pm

Re: Page 93

Post by TrashMan »

Worst case scenario is locusts eat everything. Including you. And other races that have nothing to do with you, or the war.

LegioCI
Posts: 58
Joined: Fri May 06, 2011 3:15 pm

Re: Page 93

Post by LegioCI »

Arioch wrote:Having telepathy is also a mixed bag. It can allow very close communication within a local group of scientists, but reliance on telepathy deters communication with more distant groups and the community at large. The Loroi are very good at solving specific problems within a discipline (as in producing the Pulse Cannon from the Historian Plasma Focus example in just a few years), but not as good at the interdisciplinary collaboration that often produces true innovation (the Loroi have come up with very few new technologies on their own without an example to follow).
It seems to me that Loroi, at least in an R&D sense, are excellent up-and-down thinkers, (How do we derive a better weapon from this weapon technology?) but may have trouble with lateral thinking. (How do we use this seemingly unrelated information to give ourselves an advantage?) Do the Loroi have problems with lateral thinking and intuitive leaps in general, or is it just a problem within their scientific community?

Perhaps if humans take on a larger military role within the Loroi-Umiak war, human eggheads would start requesting all sorts "useless" information about the Umiak and using this information in new ways. ("Human, why did your species paint your fleet green with purple stripes?" "The information you gave us about Empire's pre-industrial biology included the Tkkik'ti, one of the few natural predators that preyed on the Umiak into historical times. We think that by painting our ships to look like the Tkkik'ti we may provoke a useful psychological response in their hivemind.")
"But notice how the Human thinks. 'Interesting... how can I use this as a weapon?'" - Arioch

User avatar
bunnyboy
Posts: 543
Joined: Sat Mar 05, 2011 9:21 pm
Location: Finland

Re: Page 93

Post by bunnyboy »

LegioCI wrote:"Human, why did your species paint your fleet green with purple stripes?" "The information you gave us about Empire's pre-industrial biology included the Tkkik'ti, one of the few natural predators that preyed on the Umiak into historical times. We think that by painting our ships to look like the Tkkik'ti we may provoke a useful psychological response in their hivemind."
And on this point, the loroi ambassador did know, that they should make peace with umiaks and exterminate the humans.
Supporter of forum RPG

User avatar
icekatze
Posts: 1399
Joined: Sat Mar 05, 2011 8:35 pm
Location: Middle of Nowhere
Contact:

Re: Page 93

Post by icekatze »

hi hi

I didn't read much of Saberhagen's Berzerker series, didn't much care for them, but he did bring up some good points about the flaws of self replicating war machines.

First off is quality control. Can you expect them to make a 100% perfect copy every time? If they make 100% exact copies every time, they cannot adapt to a learning enemy. If they don't make 100% exact copies and take steps to counteract enemy advancements, then you cannot ensure they will not eventually turn on you, or even that they won't turn on each other.

Its hard to say whether or not etiquette requires a vassal state to avoid showing up their parent state. I guess it depends on the ruler. :)

User avatar
Mjolnir
Posts: 452
Joined: Sat Mar 05, 2011 1:24 pm

Re: Page 93

Post by Mjolnir »

icekatze wrote:First off is quality control. Can you expect them to make a 100% perfect copy every time? If they make 100% exact copies every time, they cannot adapt to a learning enemy. If they don't make 100% exact copies and take steps to counteract enemy advancements, then you cannot ensure they will not eventually turn on you, or even that they won't turn on each other.
For the software, you actually can expect a 100% perfect copy. With forward error correction codes, it's rather straightforward to get the mean time between uncorrected errors to be longer than the lifetime of any stars you're interested in. And manufacturing variations will almost certainly not be propagated to offspring...a slightly longer bolt isn't going to lead to offspring with the same over-length bolt, and a misplaced electrical component is most likely going to make some subsystem just not function. The manufacturing process is equivalent to the developmental process from a fertilized egg...there will be differences between genetically identical individuals, but those differences won't be passed to the offspring.

As for adaptation, you won't want to rely on random manufacturing variations for this in the first place. As above, even a useful variation is unlikely to be of a nature that can be passed to offspring. You'd want the design to evolve, which can be constrained (and tested in-silico prior to manufacture) in a much more sensible fashion to give results that are viable (not having bolt threads of the wrong pitch, for example, or engines pointed at the main CPU, or cables too short to reach their targets...), more likely to make a difference in adaptation to the environment, yet still responding properly to commands.

There's another problem, though. Without some other population control, they'll expand in an area until they exhaust some critical resource. You end up with a system full of drones, but lacking in exploitable deposits of rare elements (which may only be used in minuscule amounts in individual drones, making recovery from that source difficult) or in various consumable resources that the drones have used up (asteroids with accessible volatiles, fissile isotopes, etc). You'll definitely want to place constraints on population growth, and to be damn sure those constraints are working before you leave them unsupervised for any period of time.

LegioCI
Posts: 58
Joined: Fri May 06, 2011 3:15 pm

Re: Page 93

Post by LegioCI »

Mjolnir wrote:There's another problem, though. Without some other population control, they'll expand in an area until they exhaust some critical resource. You end up with a system full of drones, but lacking in exploitable deposits of rare elements (which may only be used in minuscule amounts in individual drones, making recovery from that source difficult) or in various consumable resources that the drones have used up (asteroids with accessible volatiles, fissile isotopes, etc). You'll definitely want to place constraints on population growth, and to be damn sure those constraints are working before you leave them unsupervised for any period of time.
Actually, if you do it right, you have an entire system worth of exploitable resources pre-harvested for you. Just create a command where after they've eaten everything in a system, they begin to cannibalize themselves back into chunks of their constituent elements, along with some manner of beacon that would allow friendly harvesters to pick up the remains at a later date. For volatiles and fissiles, you might just have to accept that a certain amount of those elements are going to be lost as a cost for both clearing a system of hostiles and harvesting it's resources.

Also, I'd include some manner of protection for "earth-like" planets, since planets with useful, stable ecosystems are likely far more valuable than the sum of their resources.
"But notice how the Human thinks. 'Interesting... how can I use this as a weapon?'" - Arioch

Post Reply