Why can't the Federation build androids?

[Phantasee]

If Riker can be cloned by a transporter accident, implying the transporters can recreate a person completely, and Data has been transported several times (at least), why can't they just make new Datas that way? Or maybe isolate part of his brain and make that again? IIRC, his head was separated from his body in ancient San Fransisco and they beamed that up at one point.

[Havok]

Then you are getting into the morality of cloning. Once that new Data is created, he becomes his own sentient being, protected by the same ruling that saved Data's ass in the first place.

GR: It never stuck me as the body being the the issue. Hell La Forge could probably engineer the body by himself. The issue has always been the magic of the positronic brain. So far, only Soong figured out how to do it successfully. Even Data, who obviously has a vast knowledge into his own workings, couldn't replicate it successfully. I am also going to doubt that Starfleet hasn't kept working on it, and they just haven't had that breakthrough yet. Soong did spend a good portion of his life working on it.

[Ghost Rider]

Then you are getting into the morality of cloning. Once that new Data is created, he becomes his own sentient being, protected by the same ruling that saved Data's ass in the first place.

GR: It never stuck me as the body being the the issue. Hell La Forge could probably engineer the body by himself. The issue has always been the magic of the positronic brain. So far, only Soong figured out how to do it successfully. Even Data, who obviously has a vast knowledge into his own workings, couldn't replicate it successfully. I am also going to doubt that Starfleet hasn't kept working on it, and they just haven't had that breakthrough yet. Soong did spend a good portion of his life working on it.

But that's the thing. Soong only has access to so much resources. It's something I really hate Trek kept perpetrating about how engineering and discoveries working. Soong coming upon it, with other having near completed ones...fine. Soong being the only one? It's literally ridiculous. May as well claim Ford developed the only working car.

They have entire detail scans of the positronic brain. The only mystery is playing with it, which again makes no sense given they found Data and put him back together the first time, let alone the few other times. Hell, they never make the implication that they haven't made prototypes, and such not.

The only issue is the writers have told us "Starfleet just can't!". Which is pretty much saying Starfleet is run by exceptional monkeys, which is nothing more then bad writing. Especially in light of how much they focus upon Data.

[Broomstick]

By the way - what did happen to the EMH Doctor after Voyager? Wouldn't call him an android but he is an artificial intelligence.

[Ghost Rider]

By the way - what did happen to the EMH Doctor after Voyager? Wouldn't call him an android but he is an artificial intelligence.

Pfft...like they'd care. He's probably slaving away with his brothers in the mines of Saturn or some shit.

No, really...I think literature has said something that he's still around, possibly on Voyager, there is no mention of any of them other then Janeway becoming an admiral. Worse still? That slaving away in mines is what they did with his sentient brothers, because the creator wasn't satisfied.

Makes you wonder maybe Lore wasn't completely off his rocker about "Fuck the Federation" kick.

[Junghalli]

I was thinking cheapness and ubiquitousness of components was definitely what led to the adoption of the holograms like the EMH, and especially for dangerous situations like mining work or, since a Hologram can do everything a robot can do, with two special extras. One being that if there is a cave in, the Hologram doesn't care unless the emitter is hit, and if the emitter is hit, the Hologram is still unharmed, and emitters can be connected by very lengthy wires to the mainframe.

To a certain degree the same should also be true of androids; the fact that you can't fit a self-sufficient AI into one doesn't mean they're useless. You could have the main intelligence contained in some sessile computer system and have it manipulate android remotes, ideally through some kind of wireless link. For that matter just an android surrogate that could be controlled by a human through telepresence would be very useful for all sorts of hostile environments and hazardous missions. Given that they seem to have no problem making communicators the size of name tags that can talk with ships in orbit it's rather surprising that many dangerous away team missions aren't handled by such surrogates. I imagine would have saved an awful lot of Redshirts.

Why make artificial intelligences at all when it is so easy and cheap to make/replace natural intelligences?

Organic brains have a serious inherent limitation in that they seem to be "programmed" largely by physically changing the shape of the hardware and they were never designed to have their contents easily read. This makes copying and modification much harder than it would be in a system where you could easily read the software and rewrite it without having to change the hardware. Being able to copy and modify minds easily would have a couple of dramatic consequences:

1) It would allow mass production of minds. For example, instead of having to train thousands of Chief Engineers for its starships Starfleet could simply train or create one Chief Engineer, copy him thousands of times, and distribute one copy of him to each starship. It would do for skilled minds what factories did for physical products; instead of spending months or years creating each and every specialist you could produce new ones the same way you'd produce copies of Microsoft Word. Theoretically you could do this with organic brains, but it would require observing the physical brain at high resolution and then creating a high resolution physical copy. Copying software is much faster and easier.

2) It would allow minds to be readily modified and upgraded. You could, for example, teach somebody a skill (like flying an airplane) in the same way you would upload new software into a computer. Doing the same for an organic brain would probably require complex microsurgery.

So AIs could have significant advantages over meat people even if the hardware was more expensive to produce. Yes, it costs money to make an AI to do the job of Geordie whereas all it took to make him were his mom and dad having sex, but the question is does it cost more than his training, factoring in not only its monetary costs but the time it took?

[Ted C]

Also on the note of can or do? Starfleet has said it cannot construct an android like Data. Not that it doesn't, they cannot. Data was so far the only one to come close to any notable success and even then it was regarded a failure. Hell, Starfleet has demonstrated they can do AI well beyond Data, but the question is why no android. And the answer is what Ted C pointed out, compacting the AI. Whatever Soong did they CANNOT repeat.

To be fair, I'm far from the first to have made that observation. In fact, I think it's in the Canon Database.

What does an AI bring to the table that a natural intelligence doesn't? So far the only thing mentioned is a more durable humanoid form. That's it. But your average Federation mope doesn't need that. Of course, the Federation puts silly light-up touch panels on things that don't need them, either, so I suppose they don't look beyond pretty lights and shiny objects.

An android combines some human-like ingenuity and self-determination with substantially superior ability to store and recall detailed information. Data's superior physical attributes are a bonus.

[Ted C]

They have entire detail scans of the positronic brain. The only mystery is playing with it, which again makes no sense given they found Data and put him back together the first time, let alone the few other times. Hell, they never make the implication that they haven't made prototypes, and such not.

It seems as though the software is also a major part of the problem. Data and Lore were the first androids Soong made that didn't crash after a while, presumably in much the same way Lal did. Even though Starfleet can apparently make positronic brains (the guy in "Measure of a Man" was working with one), they can't seem to keep them running.

[Ghost Rider]

They have entire detail scans of the positronic brain. The only mystery is playing with it, which again makes no sense given they found Data and put him back together the first time, let alone the few other times. Hell, they never make the implication that they haven't made prototypes, and such not.

It seems as though the software is also a major part of the problem. Data and Lore were the first androids Soong made that didn't crash after a while, presumably in much the same way Lal did. Even though Starfleet can apparently make positronic brains (the guy in "Measure of a Man" was working with one), they can't seem to keep them running.

That's a logical and worthwhile thought. The software being the real problem rather the hardware at least makes Starfleet look competent and I could see how Soong tweaking the OS, if you will, and Starfleet not knowing what line or what difference he used. It fits better how one man could come up with the solution.

It says more of Starfleet not being a pack of monkeys and just running into a logical stopgap.