Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror
Check out the new SourceForge HTML5 internet speed test! No Flash necessary and runs on all devices. ×

Comment Re:Short-term numbers versus long-term (Score 2) 127

I'm not up on state of the art on computer image/object recognition but the experience I have from about 10 years ago leads me to believe that...

Others have already responded to your other points, I just want to point out that experience from 10 years ago tells you basically nothing about the state of the art today. Deep learning methods have enabled dramatic progress on exactly the class of pattern matching problems that includes computer vision.

Personally, I still think that LIDAR is inherently superior to video cameras for this task, but Tesla's numbers are impressive, and prove that while their system may not be all that it should be, it's already better than a typical human driver -- at least than the typical Tesla buyer (note that I have no reason to believe that Tesla buyers would be worse than average drivers, but the possibility shouldn't be ignored).

Comment Re:RaspberryPi still has no competitors... (Score 2) 83

I won't call Arduino a competitor Raspberry pi - different animals. An Arduino is an embedded system while Raspberry is a minicomputer. For example you would not want to use a Raspberry pi in something that will have its power pulled abruptly or on and off - unless you go though a lot of hoops to create a read only file system and even then it is risky to use it for something in the field or embedded in a another system.

Comment Re:Whitespace takes the most space (Score 1) 175

But what is the value of an algorithm that you can't actually execute?

In the practical world, language efficiency actually matters and is a reasonable thing to discuss.

Sure, that's true. But it has no bearing on the question of whether a language can accurately be called Turing Complete -- and Turing Completeness also matters, because it defines the class of algorithms that can be implemented in the language. What's the value of an algorithm that you can't implement because the language lacks the necessary expressive power? Except in very limited circumstances, Turing Completeness is a prerequisite. Without it, there's no point in discussing efficiency.

Comment Re:You need to do a bit of research. (Score 1) 140

Star Trek Continues also violates those same guidelines (high-quality props/sets/uniforms instead of toy-store quality items, professional acting/directing/scriptwriting

Have you seen Star Trek Continues? Cheesy plots, lousy acting, terrible effects and you can't tell me their props, uniforms and sets don't look like toys.

It's like a low-budget 1960s vision of space travel.

Comment Re:Whitespace takes the most space (Score 1) 175

To be considered Turing-complete, a language must be able to simulate a Turing machine - and that's actually impossible, since it can never meet the "infinite tape" requirement.

Languages are not machines. Languages have no memory limitations, and therefore have no trouble simulating a Turing machine.

The fact that we run code written in those languages on finite machines does not change the Turing-complete nature of the languages.

Comment Re:He's missing the point. (Score 4, Insightful) 132

It would be nice if people could learn to think in terms of threats that fell somewhere between "safe to ignore" and "extinction level event". Or could distinguish between "extreme and expensive" responses and "effective" ones.

9/11 could have been prevented by simple, conservative and inexpensive countermeasures. After 9/11 politicians droned on about how "9/11 changed everything," but the cold sober fact was that it in fact changed nothing. It just showed that some of the things sensible people had already been telling us to do (like reinforcing cockpit doors or getting agencies to work together despite institutional rivalries) really did need to be done. Instead "9/11 changed everything" became the rallying cry for every pet scheme that had heretofore been correctly dismissed as too expensive, hare-brained, or just plain dumb.

Which doesn't change the fact that something needed to be done. Here's the lesson I think we should take into this infrastructure debate: we should take sensible and conservative steps to secure infrastructure against terrorism now, before events put foolish ones on the table.

Comment Re:Good but... (Score 1) 112

Or... what if anytime anyone called a residential number, a nickel was transferred from the caller's account to the callee's account.

That wouldn't stop anyone from making a call where an actual person is likely to be involved; the labor costs for a three minute conversation would swamp that. But it would discourage people from robocalling a hundred thousand people in order to turn up a handful of suckers.

And the public wouldn't have to pay a regulator to try to track down these boiler room operations.

Comment Re:Agrument in favor of modularity (Score 1) 86

I don't have to do anything. Even stored under ideal circumstances li-ion batteries lose capacity.

What matter is capacity relative to demand. In a phone like the Droid Maxx from a few years ago with plenty of surplus battery the phone will still be usable four years later. But something like a Samsung Galaxy S6 barely has enough battery to make it through the day when brand new and is pretty much unusable two years later even under ideal conditions.

Comment Re:There's a lot more iron much closer... (Score 4, Informative) 287

And there's some twenty million tons of gold dissolved in the Earth's oceans. Jules Verne made it the source of Captain Nemo's incredible wealth.

To put twenty million tons of gold in perspective, all the gold that has ever been mined by humans totals up to about 180 thousand tons. To put in another perspective: sure, it's gold, but at a concentration of thirteen billionths of a gram per liter of seawater it's worthless unless you have unlimited time and energy to extract it.

That's the problem with asteroid mining in general. Until the cost of changing an object's momentum goes down drastically it's not worth doing. If Pysche were a 1000 kg block of pure, refined platinum (market price: $34 million) you'd be hard-pressed to retrieve it and return it to Earth at a profit. Which is not to say asteroid mining is a bad idea; but first things first: you've got to reduce the price of interplanetary propulsion by a couple orders of magnitudes. One thing that never happens in a sci-fi asteroid mining scenario is the hero worrying about running out of gas. Propulsion in stories is always practically limitless and free of charge. Real propulsion will never be that good, but it could get good enough.

Slashdot Top Deals

One good suit is worth a thousand resumes.

Working...