Become a fan of Slashdot on Facebook


Forgot your password?
Back for a limited time - Get 15% off sitewide on Slashdot Deals with coupon code "BLACKFRIDAY" (some exclusions apply)". ×

Comment Re:This is a breakthrough? (Score 2) 26

Yeah... I really don't understand a lot of robotics research. They seem to be forever chasing these awkward "proof-of-concept" implementations of concepts that are completely uninteresting. This is a perfect example: obviously you could make some robots that could do this, but it's really unclear what you'd learn by doing so, and the result is useless.

I mean, if they actually wanted this behavior for some purpose, and this was a reasonable way to approach that practical purpose? Sure, do it. Of course. If there was some question whether this was possible (which obviously there wasn't, at least not serious), then sure, prove it's possible. If there was some question we'd be answering (about biology, maybe?) or some challenge that would be interesting to overcome, then, uh, maybe. But it seems much more likely that such questions could be answered easier by a simulation.

As it stands, it seems like we've got a HUGE surplus of research on, to pick a random example, how to have robots all go somewhere without bumping into each other - most of it saddled with arbitrary restrictions that are also completely uninteresting. Can the swarming robots co-ordinate without any predetermined communication protocol? Yes, obviously, and we can prove that in simulation. But that's not good enough for some reason, have to actually build some crappy robots to mong into each other while we re-solve a bunch of boring practical problems with batteries and light sensors.

If you just want to have a challenge, or get some practice building robots that have to deal with the real world, why not build them to at least attempt something interesting or useful, or at least entertaining?

Comment Re:Why? why now? (Score 1) 158

If you're a developer working in a shop writing code for bethesda or valve or EA, chances are your windows site license for desktops and servers is already heavily discounted thanks to your generous interest in a visual studio license despite eclipse being right there. Chances are even better that in order to keep this generous discount your manager has started shoehorning C# into your project requirements to 'maximize the investment value' of what basically amounts to a protection racket for good customers.

This is one of the dumbest things I've read on /. for a long while.

Comment Yeah, well, naturally. (Score 2, Funny) 91

I mean, this is like the core principle of homeopathy. You make permanent holes in the water, and the holes are just the right size for the class of toxins you're dealing with. Then if you want more holes, you dilute the water to make the holes split (obviously you want to be careful with this in practice).

I thought everyone knew this? How did you guys all think homeopathy worked? Magic?

Comment Re:Fingerprints are Hashable (Score 1) 242

Sure this is a problem, but not an unsolveable one.

As long as only a small percentage of measurements fall into ambiguous bands, you can solve this in practice by simply jittering any of the measurements that were very close. IE, if you have 9 measurements that are clearly in a band, and one that's right on the edge, you can just try both the nearby values for the tenth and see if either matches the hash. If you have too many measurements that are equivocal, then your system has failed as you couldn't (and wouldn't want to) be jittering everything... but overall, this is a design challenge, not an absolute showstopper or something.

Comment Re:Premise is not necessarily correct. (Score 1) 242

Wow, that was a stupid, bizarrely aggressive post. What's extra bonkers is that it seems like you agree with me.

No I'm not an expert on fingerprints (or security or cryptography in general), but that doesn't mean I can't clarify a pretty simple misunderstanding. In this case, my point was that building a hashing system suitable for fingerprint authentication may have challenges, but is not inherently impossible (as the original article implies).

Everything you said agrees with that determination (challenging practically, but no reason to think it's actually impossible). I didn't think this would be a controversial stance at all really, and I certainly didn't expect to induce hilarious rage-posts like yours here.

Comment Re:Premise is not necessarily correct. (Score 1) 242

I'm not suggesting hashing the image, I'm suggesting hashing a stable digital digest, the contents of which might be determinations like "are there significantly more of feature X than Y" that were repeatable. I thought that was clear?

Your next point is a reasonable possible problem with practicality - if your "stable digest" is too small then it obviously won't work. But there's no reason this digest couldn't be usably large with a scan of appropriate quality; to be clear, I'm taking issue with the idea that hashing a fingerprint is impossible in principle, I'm not saying it would practically work right now in realistic scenarios.

And I would say fingerprints are secrets to some extent. If someone were able to harvest a ton of identified fingerprint images from a database breach, I think that would be a significant negative. Being able to hash is a positive for an authentication system.

Comment Re:Premise is not necessarily correct. (Score 1) 242

I don't think you read my post?

My whole point is that you don't have to compare image to image. I'm saying that you could take a fingerprint image and digest it into a set of boolean qualities that are stable for various images of that fingerprint (ie. this one has significantly more dipsees than doodads, so that can be a bit in our stable digest). Then you hash that digest and store it. For a new authentication attempt, you do the similar digest and hash, and the two hashes have to match (like a normal password check).

Finally, to repeat from my previous, I grant that this might be difficult in practice, but it is not impossible (and thus this is not some intrinsic limit to using fingerprints for authentication).

Comment Premise is not necessarily correct. (Score 3, Interesting) 242

It's more awkward to hash a fingerprint than a password, sure, but it's certainly not impossible. An image of a fingerprint is mutable and "analog" feeling, but you could, instead, base your fingerprint comparison on a more "digital" digest of information from that fingerprint (eg. you boil image data down to bits that are repeatable in the face of repeated scans, like you check whether feature X is significantly more prevalent than feature Y in this print).

It'd be tricky, sure, and potentially impractical given current scan quality - but non-hashability is not some inherent limitation of fingerprints or biometrics in general.

Comment Re:Let me get this straight: (Score 1) 428

This actually doesn't hold as well for humans as it does for mice - overall, humans tend to live longer if they're moderately overweight. This - - is kind of a fluffy article, but it's a good summary of the research.

Nothing in diet/health is simple (despite 95% of the comments in this thread saying that "it's obviously X").

Comment Re:Let me get this straight: (Score 4, Informative) 428

We're now not supposed to eat meat of any kind

If you want to misrepresent what was said, that's your prerogative. The WHO didn't recommend not eating meat, only not to eat processed/smoked meats, and to limit red meat. As usual, the concept of moderation goes *woosh* over people's heads as they furiously go about constructing their strawmen ..

Comment Re:Distance? (Score 1) 257

So you simulate it at a lower level for a while to figure out the probability/frequency of that observable macro-level behavior, then just have it happen (on an observable scale) at an appropriate set of times. The whole observable world has to be at least somewhat consistent with the lowest level behavior, but that doesn't mean you have to simulate all those particles all the time.

Comment Re:Distance? (Score 1) 257

The idea of the (reasonable) simulation argument is that not everything would be calculated out to the particle level all the time - the simulator would only figure that stuff out if there was some reason to. I mean, you can simulate the observable, macro-scale behavior of the sun without actually figuring out the position of every quark inside it.

But sometimes, like if some scientist is looking very close, you really do have to figure out exactly where every particle is in order to render that scientist's experience - and once you've done some of the particles, you have to store the value for the other ones too. Once they quit looking too closely, you just simulate the aggregate results. It's a tidy explanation of dual-slit type results and entanglement.

It's not, like, strong evidence or something, but it makes sense. And, while I can't think of a test, that doesn't mean a simulation hypothesis isn't testable. Under this view, these sorts of quantum effects are effectively simulation artifacts - and perhaps there's more dramatic artifacts that can be induced that would be more clear.

At the source of every error which is blamed on the computer you will find at least two human errors, including the error of blaming it on the computer.