Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
DEAL: For $25 - Add A Second Phone Number To Your Smartphone for life! Use promo code SLASHDOT25. Also, Slashdot's Facebook page has a chat bot now. Message it for stories and more. Check out the new SourceForge HTML5 Internet speed test! ×

Comment Re:Robots are good (Score 2) 236

Unfortunately, it's looking like they may be able to replace many/most jobs within a handful of, and that's not "a long time" in political terms. Especially not when we're talking about requiring major changes in a centuries-old social legend ("doing for yourself") embedded in most aspects of our social system.

Comment Re: "Destroyed" is such a harsh term... (Score 1) 88

In my experience "bricked" refers fairly exclusively to a non-recoverable state - at least through "normal" means. E.g. you've borked the firmware badly enough that you can no longer install the updates that would repair it. Hence things like "unbrickable" motherboards that have a second back-up BIOS in case something goes wrong when updating the primary one.

Granted, often times there's internal diagnostic pins that can be accessed by sufficiently knowledgeable individuals with the right equipment in order to get things working again - but it's not something your average firmware-updating geek is going to be prepared for.

Comment Re:Open Source Books (Score 1) 123

Only if it's a collaborative project and one of the collaborators objects. I'm free to distribute only the binaries to my own program under the GPL - as the copyright holder *I'm* not bound by the license, only everyone else. It'd be a jerk move, but legally fine. In which case the GPL would pretty much just grant you rights to resource modification, decompiling, etc.

For a book, where the only "source code" could well be a typewritten or even hand-drawn manuscript, there could be considerable room for argument.

Still, I'd really hope they use something readily editable, kind of kills the long term benefit otherwise.

Comment Re:Open Source Books (Score 2) 123

That would be rather stupid, as I'd bet good money that some student would get it scanned within days and start distributing copies (which would be completely okay by the terms of the GPL) and the original printer would lose a ton of goodwill.

Moreover, "open source" typically implies "readily editable by those with the right tools" so that it can be rapidly enhanced through collaboration. Which for textbooks probably means LaTeX, though something more like a .doc file might also get used.

Comment Re: Two months ago "Couldn't keep up with demand" (Score 1) 120

I don't see why. As a rule, engineers don't build things, they *design* them. Once the design is fully complete and you have a production model fully built and tested, I suspect you need far fewer engineers to build the next hundred. If you have lots of orders for existing models, but few orders for new or heavily customized designs, then you need lots of assembly technicians and may well be hitting the limits of your production facilities, even while you have a bunch of much more expensive engineers sitting around with nothing much to do.

Now, if you expect custom orders to return in the near term, maybe even the mid term, then it probably makes sense to set those engineers working on "side projects" that don't directly effect the bottom line, just to retain their expertise - long term technology overhauls like speculative next-gen plane designs and other "busy work" that might prove valuable eventually, but whose primary purpose is just to retain talent. At some point though it becomes more cost effective to just fire them and hire new engineers when business picks up again. Especially if you have a list of poor performers, "problem individuals", and those whose retirement package is about to vest. Layoffs can be an excellent way to cull such especially costly individuals without the same level of legal scrutiny risked by individual firings.

Comment Re:They could have done better with the data (Score 1) 343

Actually, not true. Even just talking on a phone is still distracting, considerably more so than talking to a person in the car with you, who will tend to pace their conversation to the complexity of surrounding traffic and may even spot impending problems before you do. As I recall, the average person talking on a hands-free phone while driving is roughly as dangerous as if they had a couple drinks under their belt.

It's all about where your attention is focused. Attention is a severely limited resource and you're much less dangerous when it's focused on the car and surrounding area. The more you focus on someone who's not physically there, the less you have to pay attention to your surroundings. You don't suddenly grow more attention just because you need it.

Comment Re:Nah (Score 2) 104

Nice goal-post moving. Define "intelligence". Keep in mind it may have nothing whatsoever to do with conscious thought.

Perhaps the biggest thing we've learned from AI research is how many "intelligent" skills and behaviors can actually be performed by machines without any apparent shred of human-like intelligence - like say being able to completely trounce Grandmaster Go players at one of the most subtle and complicated games ever created, one that is essentially impervious to the sort of brute-force analysis that makes it so easy to write an unbeatable chess-playing program.

Comment Re: Nah (Score 1) 104

An awful lot of profitable technology employs that "dead end".

Just because there's no evidence that we're actually any closer to true artificial consciousness, doesn't mean AI research hasn't developed some truly astounding applications of limited domain "intelligence". And we're still only scratching the surface of "neural" networks specifically - they failed to do much of interest in the 60s in large part because they are *very* badly suited to emulation in software and require enormous processing power to simulate even relatively simple networks. Tody we have dedicated hardware designed specifically to implement neural networks in hardware - still only small ones of a few million neurons, not even on par with a mouse brain, even before you consider how much more sophisticated real neurons are, but they're enabling some really interesting things nonetheless.

Comment Re: Question for the Physicists. (Score 3, Interesting) 79

I was referring, badly, to motors using permanent magnets in the stator or armature, as opposed to those based entirely on electromagnets.

Actually though, you can make temporarily "perpetual motion" magnetic motors that draw power entirely from draining the energy stored in the magnets - for example, picture the situation where magnets on the rim of a wheel are attracted to a stationary magnet nearby. Then, roughly at the point of closest approach, the rim magnets pass behind some form of shielding so that there is no symmetrical force required for them to be moved away again. Unlike gravity (so far as we know) magnetic fields can be blocked, and so the rim magnets will perpetually "fall" towards the stationary one until the magnetic field is drained. You can do something similar with electrostatics as well, and many people have convinced themselves they've managed to create a perpetual motion machine that way.

Slashdot Top Deals

Work continues in this area. -- DEC's SPR-Answering-Automaton

Working...