Is biotech feasible, or just wank?

[wolveraptor]

I've been wondering: are those stereotypical biotech aliens, with the living tanks, and animals for spaceships, feasible? It seems to me that nature has put certain limits on what creatures can do. Geometric scaling, for example, prevents elephant-size ants from lifting fifty times their body weight. Could there really be a race that has complete biotech (like the Vong) and still beats down machinery?

How would they deal with space? Alien atmospheres? Alien gravities? Conditions inhospitable to life itself?

[Shroom Man 777]

...no. Sure, living ships might be possible. But they'd suck ass. With a vacuum pump. Hard.

[Nephtys]

Any conceivable biotech of the sci-fi scale would probably have to be based off of totally different mechanisms than what we're familiar with. The traditional 'wet sack carbon-based doohickey' biotech would be frought with issues.

A living tank isn't particularly unfeasable. Just damn unlikely to evolve, unless it's environment regularly fires explosive shells at it or something.

[Miles Teg]

I think the only area we are likely to see "bio-tech" in is modeling technology after biological products (and enhancing it). For example, spider silk is really strong, and now that we know *why* it is strong, we can expand on that knowledge and create something that is similar, but better (see: kevlar).

The most promise for this, in my oppinion, is with computers. The biological brain is still light-years ahead of any computer we can make. Computer Scientists have already been tapping the design potential of a biological brain in the form of neural nets, and other concepts based off our knowledge of the brain.

Miles Teg

[Molyneux]

Well, you could theoretically get biotech with a program of eugenics...it'd be more along the lines of a better attack dog or that kind of thing, but it's theoretically possible.

Yeah, I know I'm BSing, but I just want my Zerg swarm...

[GrandMasterTerwynn]

I've been wondering: are those stereotypical biotech aliens, with the living tanks, and animals for spaceships, feasible? It seems to me that nature has put certain limits on what creatures can do. Geometric scaling, for example, prevents elephant-size ants from lifting fifty times their body weight. Could there really be a race that has complete biotech (like the Vong) and still beats down machinery?

How would they deal with space? Alien atmospheres? Alien gravities? Conditions inhospitable to life itself?

There's a smiley on this board that describes the likelyhood of such sci-fi biotech . . .



Due to the constraints of biological materials, you're not going to build a creature that will secrete high-strength steel alloys, or anything sturdy enough to be used as a spaceship. The closest you'd get to biotech in space are applications where engineering something biological would be more practical than a mechanical solution . . . maybe some sophisticated genetically engineered biological oxygen generator that takes in your sewage and stale air and excretes breathable oxygen.

[Admiral Valdemar]

You realise by saying "biotech" and "wank", you're essentially calling an entire field useless because some sci-fi authors haven't the slightest idea about how science works. That's like calling engineering wank because some sci-fi races can make moon sized battlestations that blast planets apart.

[wolveraptor]

The only term I could think of was biotech. What else do you call a bunch of spine-shooting buggers that secret acid?

Yeah, I was pretty sure that no civilization could be based completely off of biotech.

[Gil Hamilton]

The most promise for this, in my oppinion, is with computers. The biological brain is still light-years ahead of any computer we can make. Computer Scientists have already been tapping the design potential of a biological brain in the form of neural nets, and other concepts based off our knowledge of the brain.

Miles Teg

Actually, as computers go, biological brains aren't exactly the paragons of good design and ability, they are just different than the computers we make.

In terms of speed and design, our computers are massively better than our brains. Remember, a human brain only cycles 60 times a second. Compare that with a computer, even a really old computer.

Plus, a computer is vastly better organized. The human brain is one of the most convoluted things on the planet, so much so that no engineer would dream making a system that rediculous. 90% or so of the human brain is the neurological equivlent of wires IIRC, or "white matter". One of the reasons we have so much trouble mapping it and it's functions is not that they are necessarily complex, but because the brain is so makeshift, crosswired, and convoluted that mapping it in an organized fashion is nearly impossible (this is why neurology requires alot of trial and error, like hooking people up to CAT or PET scanners and then asking them to do multiplication problems, to see what part of the brain starts cooking). Not so with computers.

Also, there is the problem of access. Human brains just aren't reliable when it comes to memory. It forgets information very easily (I've heard it said that people immediately forget 75% of the information they receive) and can't recall it readily. Plus, the human brain has plenty of other quirks that make it even less reliable, like false memory creation (a documented phemonena where memories that never happened are created by suggestion, peer pressure, preconceived notions and time... the Discovery channel had an excellent show on it in regards to UFOs). Computers don't do that. They are organized and can retrieve much of the information on them.

What computers lack is creativity and sentience. Thus, I believe the goal is to not make a computer version of the human brain, but rather make a computer that possesses the interesting aspects of the human brain that have alluded us. It can't be said, however and in my opinion, that our brains our lightyears ahead of computers, because they really aren't, and I don't see any reason why biotechnology is superior in computer science.

[wolveraptor]

What I'd like is a cyborg brain, one with the memory storage capability and lightning fast processing of a computer, and the creativity of a human mind.

[Admiral Valdemar]

The only term I could think of was biotech. What else do you call a bunch of spine-shooting buggers that secret acid?

Yeah, I was pretty sure that no civilization could be based completely off of biotech.

You'd call it biotech, but that puts it under the same umbrella as artificially produced insulin and stem cell tissue replacement.

[Dahak]

What computers lack is creativity and sentience. Thus, I believe the goal is to not make a computer version of the human brain, but rather make a computer that possesses the interesting aspects of the human brain that have alluded us. It can't be said, however and in my opinion, that our brains our lightyears ahead of computers, because they really aren't, and I don't see any reason why biotechnology is superior in computer science.

What human brains can do vastly better than any computer today, and will do so for the forseeable future, is pattern recognition, especially image analysis and speech recognition.

[Admiral Valdemar]

What human brains can do vastly better than any computer today, and will do so for the forseeable future, is pattern recognition, especially image analysis and speech recognition.

Yes, the ability for wetware to operate without software and using amazingly powerful parallel processing abilities is what stands it out. A pocket calculator can outdo most humans serially (unless you're one who can do any sum mentally instantly), but we have nothing that can rival biological neural nets.

[Gil Hamilton]

What human brains can do vastly better than any computer today, and will do so for the forseeable future, is pattern recognition, especially image analysis and speech recognition.

I think that's because pattern recognition is something our brains evolved to be able to do well. In fact, the human brain does it too well on occasion, as the brain looks for patterns even when known exist and often overinteprets things (which is why so many optical illusions work, because it's easy to spoofe, below is a demonstration of this*).

My point is just that the human brain isn't actually better or more advanced than our computers, just different.


* http://bobstuffs.no-ip.com/gilhamilton/synvilla.jpg

[Gil Hamilton]

Yes, the ability for wetware to operate without software and using amazingly powerful parallel processing abilities is what stands it out. A pocket calculator can outdo most humans serially (unless you're one who can do any sum mentally instantly), but we have nothing that can rival biological neural nets.

I'd think that the ability to use software would be an advantage over a completely hardwired system.

[Dahak]

My point is just that the human brain isn't actually better or more advanced than our computers, just different.

But it is better in the fields it is "designed" to be good at, pattern recognition for one. (Or walking... The effort to have a robot walk on two legs is quite immense. 3 legs or one is much easier to do...)
You won't find a software that has the same performance in pattern recognition than our brain by far. It does this better than any computer currently existing.
The fact that the brain is running on a completely different principle doesn't change that...

[Admiral Valdemar]

I'd think that the ability to use software would be an advantage over a completely hardwired system.

That depends on what purpose you're looking at. As it goes, hardwired systems like our brains learn and far better than anything with software. The bottom-up approach to AI is still very basic, but it still beat the top-down approach of coding a response for every possible scenario (which is a little impossible). They've looked into using similar artificial neural nets and organic ones for use in machines.

[Miles Teg]

The most promise for this, in my oppinion, is with computers. The biological brain is still light-years ahead of any computer we can make. Computer Scientists have already been tapping the design potential of a biological brain in the form of neural nets, and other concepts based off our knowledge of the brain.

Miles Teg

Actually, as computers go, biological brains aren't exactly the paragons of good design and ability, they are just different than the computers we make.

You miss my point. My point is that we can learn from the human brain, and then apply (and improve upon) what we learn to computers to improve on them. I am not suggesting that we attempt to replicate the brain with technology, but rather learn what we can from the organics and adapt that knowledge. This is why I pulled out the spider silk/Kevlar example. Kevlar was inspired by spider silk, but isn’t simply a mechanical “copy” of spider silk.

In terms of speed and design, our computers are massively better than our brains.

Note true. Sure, a computer can crunch numbers a whole lot faster than a brain, but that's ALL a computer can do, crunch numbers. The brain is a much less specialized "computer". Put another way, what makes the brain light-years ahead of computer is sentience and the ability to think.

Remember, a human brain only cycles 60 times a second. Compare that with a computer, even a really old computer.

Wow, I never thought I would hear the Mhz myth in this context. You just made my day =)

I've never heard of any "clock speed" measurements of the human brain, and frankly, it's not a valid comparison. The human brain is not a discrete state machine that functions on a synchronous clock like a computer. It's a massively parallel, asynchronous neural net. Most computers today can only do one thing at a time (unless they have multiple processors). I don't know of any definite limitation to the parallelism of the human brain.

Plus, a computer is vastly better organized. The human brain is one of the most convoluted things on the planet...Not so with computers.

Again, I am not suggesting we try to build a “mechanical brain” but rather learn from the brain to improve on the capabilities of our computers. On a side note, better organization (at least how you describe it) is not always better. For instance, the human brain can reroute damaged pathways on the fly, partially due to its complexity and “hodge-podge design”. A Von-Neumann computer cannot do this. The best we can do with computers is to have redundant hardware, which is not the same thing.

Also, there is the problem of access… They [computers] are organized and can retrieve much of the information on them.

Again, you are forgetting that a computer is just a really fast number cruncher. Calling a computer vastly superior to a brain for this reason is akin to calling someone with a photographic memory more intelligent than Einstein.

What computers lack is creativity and sentience. Thus, I believe the goal is to not make a computer version of the human brain, but rather make a computer that possesses the interesting aspects of the human brain that have alluded us.

This is exactly my point. We will take what is useful from biological "designs" such as the brain and apply them to our technology. In the case of the brain, hopefully we can figure out what makes true thought possible, and learn from that to create true AI. Perhaps this will mean building a computer that mimics a human brain to some extent. Perhaps it doesn’t.

Miles Teg