Free Tools!

[tyggerjai]

It's been years since I knew where my tech gun was, let alone used it in anger.

Ad boy, have they come a long way. I'm now the proud owner of a Black and Decker 14.4V FireStorm Cordless Drill. I'm not entirely sure about calling a tech gun "FireStorm", nor am I sure red is an appropriate color. It's not black, nor is it Power Tool Green. IT is, however, a tech gun, and hopefully not a bad one. And I also have an angle grinder - good, solid, Power Tool Green B&D, with no wanky name. Ph34r!

And Ben and I disassembled meese on the weekend in preparation for building a couple of these. Under $25 for the parts for each, which is nifty.

Question probably for lederhosen and maybe Thorfy, arising from discussions on "How I'd re-engineer the human body if I were God." on saturday. Me, I'd start with knees. If they're only meant to bend one way, then the actual joint should only bend one way - none of this ridiculous "joint bending any way you like with arbitrary and inadequate constraints retrofitted in the form of vulnerable ligaments." Feh. It's ridiculous. But I digress. It was a somewhat beery discussion, but the main point I recall is someone mentioning the human retina, because it' somehow "backwards". The point was made that it's still perfectly functional, but things that are backwards and *still* functional are guaranteed to annoy engineers. So the question I'd had too much beer to bother asking on Saturday is: In what sense is the human retina backwards?

And can I fix it with a tech gun and an angle grinder? ;)

(update: GIYF. For those similarly in the dark (har har), apparently the photoreceptors in human (and other vertebrates?) eyes point *away* from the light source, and presumably rely on reflection to a large extent. Which is indeed backwards, but apparently works OK. So the question now becomes, are there real advantages to that structure?)

sol.
.

Comments

[lederhosen.livejournal.com]

Me, I'd start with knees. If they're only meant to bend one way, then the actual joint should only bend one way - none of this ridiculous "joint bending any way you like with arbitrary and inadequate constraints retrofitted in the form of vulnerable ligaments."

Engineers like joints to have either complete or zero freedom in any given mode, because it simplifies the theory, but that's not necessarily the best solution in a body. Adding a little bit of rotational freedom probably reduces the vulnerability to other kinds of stress, while increasing vulnerability to others; it's a trade-off.

Feh. It's ridiculous. But I digress. It was a somewhat beery discussion, but the main point I recall is someone mentioning the human retina, because it' somehow "backwards". The point was made that it's still perfectly functional, but things that are backwards and *still* functional are guaranteed to annoy engineers. So the question I'd had too much beer to bother asking on Saturday is: In what sense is the human retina backwards? ... For those similarly in the dark (har har), apparently the photoreceptors in human (and other vertebrates?) eyes point *away* from the light source, and presumably rely on reflection to a large extent. Which is indeed backwards, but apparently works OK. So the question now becomes, are there real advantages to that structure?

Hmm. Hadn't heard about them pointing away from the light source per se - my understanding is that the photoreceptors work in either reception. (Although it's entirely possible that I've missed hearing about something, since my area of specialty is the other end of the eye). Reflection is useful for night vision, because it means light that misses the photoreceptors the first time through gets a second chance, but the first time still counts too. If it was just that the receptors were pointed in the wrong direction, it probably wouldn't be too hard for evolution to turn them around.

No, the 'backwards' bit is that the nerves that relay signals from said receptors sit in FRONT of the receptors. Light has to get through those nerves before it has a chance to trip the receptors. This results in some loss of light (not a great deal AFAIK, but undesirable nevertheless). It also results in a blind spot where the nerves pass through to the back, and it makes the retina somewhat fragile - it might be more stable if it was anchored to the back of the eye by the nerves, instead of having them sitting in front.

(Squid, by contrast, have their eyes wired the other way around. Yay them.)

This point is sometimes used as an argument against creationism: it's not too hard to understand that such a feature might not have been a problem in a primitive eye, and that we're now locked into this - you can't get to a better configuration without going through a worse intermediate, and evolution is a step-by-step process. IIRC, Dawkins is quite fond of this example.

Creationists, by contrast, argue that this 'backwards' configuration is actually advantageous for us. I'm not really qualified to assess the minutiae of the argument; the eye is a complex enough structure that I doubt even a good scientist could be certain that changing the configuration would improve it, but if I had to put money on one side or the other I'd pick the evolutionists on track record.

Some creation/evolution arguments here, and here.

[tyggerjai.livejournal.com]

Adding a little bit of rotational freedom probably reduces the vulnerability to other kinds of stress, while increasing vulnerability to others; it's a trade-off.
Absolutely - if they had no lateral give, we'd snap actual bone every time they got any serious sideways pressure. Which is probably worse than just doing ligaments, especially in a primitive society. But, dammit, that's not the point - as a man with very dodgy anterior cruciate ligaments, I feel sure there must be a better way :)

My sources were the second link you posted, and also this,which has a little diagram showing both the "nerves in front" and the photoreceptors pointing the wrong way, as I understand it.

I conflated the "nerves in front"and the "pointing the wrong way" bit because it actually makes more sense to have the nerve fibre at the opposite end from the receptor, at least to me, so the one follows from the other.

But then it does make more sense to have the photo-receptor pointing at the light. And yet, we are not yet ruled by cephalopod overlords, so we must be doing something right.

Hm. Interesting, innit?

sol.
.

[thorfinn]

If we're talking human redesign, then I want more regenerative ability, as priority number one. Actually what I want is full conscious access to physical configuration-fu, a-la Iain M. Banks' Culture humans.

As for joint redesign, I'd want to fix elbows before I fix knees... Elbows have this stupid spot where the major arm nerve is not covered by anything except skin (hooray for the funny bone), and it's pretty clear that that was due to the muscle shifting that turned them from legs to arms.

Knees could just do with a bit more beefing up in the ligaments department, otherwise, I don't think there's a lot wrong with them. Eyes are screwy, though, yes. Cephalopod eyes (and also I think hunting bird eyes) are definitely better quality, and human eyes also have really really crap focus. We only have a tiny visual region that's in focus at any moment, which is actually quite an interesting tradeoff (it lets us spend detail focus attention only on bits that are marked as "interesting"), but I'd still like more sharp focus to be available if I want it.

Err, and the neck is a very screwy object. Not sure what could be done to improve the situation, but it's definitely screwy.

[tyggerjai.livejournal.com]

If we're talking human redesign, then I want more regenerative ability
Nah - we were talking elements of the current design that need rethinking. Otherwise, hell, sign me up for wings and infra/ultra vision.

So it's really just better implementations of existing "features". You don't get to change the Func. Spec., just the implementation.

jai.
.

[thorfinn]

We do have regen (after all, bones reknit, skin reforms, muscle regrows)... The problem is that regen doesn't work on an actually snapped ligament, and is also really difficult on an actually severed nerve, and other specialised organs won't re-grow either. As far as implementations go, then, I want the bloody vagus nerve ('... the longest of the cranial nerves. Its name is derived from Latin meaning "wandering". True to its name the vagus nerve wanders from the brain stem through organs in the neck, thorax and abdomen.') sorted out. I mean, reuse is all good, in theory, but there's no bloody redundancy, and cross-signalling leakage causes all sorts of ickiness.

[tyggerjai.livejournal.com]

Oh, and ...

The first link you posted has this:

Not only does our eye have the wires coming out the front; Fig. 2 shows that the light-sensitive parts of the visual receptor cells (the rods and cones) face backwards, away from the incoming light, where they butt against the opaque pigment layer.

so, yeah - completely backwards :)

sol.
.