And haven't found any that are terribly impressive in their abilities. They'll catch certain kinds of problems, but tend to lose their way pretty fast in more complicated code. Anyway, this list might help start you out in the right direction:
There's no reason MS couldn't have taken the route Google has with branding phones (eg. the Nexus 4, actually made by LG or Asus or I don't remember). I don't think buying Nokia is going to look like a good decision down the road.
Overall, MS's continuous doubling down on mobile has succeeded only in poisoning their other products.
With some work and tweaking, you can make a reasonable interface in Windows 8 - but I can't think of anything that's a real positive.
Meanwhile, Windows 7 fixed my concerns with Vista and generally just stayed out of my way. It performed well and consistently, feeling familiar but better than their previous OS offerings.
Give us back Windows 7.
And, like any random source, you can use it for an unbreakable one time pad. That's cool.
So I guess the question is "are there problems with current hardware random number generators?", and probably "what are the failure states for this new method, how do they arise, and how hard are they to detect?"
Regardless of those answers, there's still going to be limited utility for something like this. I don't think a lot of gamers are worried about game randomness not being random enough (which is a ridiculous application suggested in the video).
Sorry, so even if an automated car was safer than most human drivers, you wouldn't want to allow them until what, there's zero possibility of them killing someone?
1. I don't see any reason for a long timeout - they should be in continuous communication. They should be able to rotate command 1000 times a second if they wanted to. Computers and communication are fast.
2. This doesn't need to be some complicated algorithm or something. They're all sharing information, so they should all be suited for command - just have the next bot in the sequence do it.
3. I think, at this point in communication theory, we could probably design a protocol whereby we don't need, like, extensive re-handshaking or something here.
It seems to me that "being in continuous communication with each other" is going to be a requirement (or a large benefit) for most tasks anyway. If these bots are going to do anything together (other than fly and not bump into each other), they're going to require co-ordination and data sharing. So why not use those links to fly and not bump into each other.
I've seen people doing flocking demonstrations for years, and it seems like something robot tinkerers spend significant time on. And it usually involves this:
Crucially, the flock does not rely on any centralized control for its behavior.
Why? Why is that crucial? Why not let the robots communicate with a central control? I understand that's not how animals do it, but animals don't have, like, RF glands. To be clear, there's no reason the central control couldn't be in one of the robots (and there's no reason the "central" robot needs to be statically defined, they could instantly elect a new one if the old one dropped out or something). It's only a difficult problem in practice, but there's not really a practical reason to impose this restriction.
Where's the big downside of a central control? The upside is the practical problem is way easier. And it is really just a practical problem - the theoretical flocking problem is much easier and can be thought about much simpler in simulation.
In general, robot tinkerers seem to spend a ton of time making up odd, practical problems that don't need practical solutions. Like the dudes a few SlashDot stories ago that were inventing a way for robots to communicate facts to each other without sharing any kind of pre-defined language. The communication thing is an interesting, useful problem - but it has nothing to do with robots, and doing it with actual robots just adds a bunch of extraneous hassles. It'd be like building counting robots to move abacuses so you could to math theory.
I mean, if you're actually building robots that need to communicate, you can just have them able to communicate in a non-ridiculous way because we know how to have computers communicate at a distance. Just like you don't need a robot to be able to physically manipulate an abacus (at least not in order to help it count).
There was money to be made at certain points, sure - and there may be more money to be made in the future. I'm sure some people have done quite well. But that doesn't mean any significant involvement with BitCoin going forward is a good idea.
Trusting "BitCoin" isn't exactly what's important. To invest in or use BitCoins significantly, you'll end up trusting other people - and how do you know to trust those people, especially as the stakes get higher and higher? Banking and securities trading have a web of trust and regulation that's been built out over centuries. There's failure states and scandals, sure, but you have reasonable tools to decide who to trust and how much.
What I see in people's experience with BitCoin is often a long string of red flags - difficulties doing withdrawals and transfers, huge fluctuations in value, varying exchange rates that nobody is able to arbitrage - all met with too few questions and far too much exuberance.
People make mistakes, intentionally or otherwise
Yes, people do make mistakes. Often while driving. The test shouldn't be whether automated cars make mistakes, but rather whether they do better than an average driver. Can they deal with icy roads as well as an average driver? That bar's pretty low, even here in Edmonton.
Once they've reached that average competence and start being deployed, they'll also improve rapidly over time; computers have the potential to be much safer drivers than humans. They'd know where other cars are and where they're going, they'd be able to apply brakes to wheels independently with lightning reactions, and would not be subject to health conditions, intoxication, aging, or inexperience.
I'm not sure how far off we are, but it's definitely coming.
It's kind of moot now that rental stores are pretty rare - but this actually isn't true. Under first sale doctrine in the US, you're allowed to rent out a DVD you own. If this wasn't true, rental places may never have taken off, as the studios would have preferred only to sell. They tried various license garbage to hinder renting, but it never held.
That's not at all what I said.
Yes, it is. You said:
as they won't have the cost of running their own dealers
If something is a cost, then it's something that loses money. I understand why you want to have not said it now (and we both know it's wrong) but you did say it, and denying that is really silly when it's right bloody there.
And pretending Apple is more like a car dealer than a car manufacturer is ridiculous, and I think you know that.. so I'm not going to bother continuing.
But, uh, there is?
There's lots of non-dealership places that I can go to fix my car, and those would exist no matter who owned the dealership I bought my car at (and might be more prevalent, even, if more car selling was direct from manufacturer).
Similarly, there's already manufacturer original parts and parts made by other companies. This has even less to do with who owns the dealership. It's not like the independent dealerships are making all the parts they use.
Unfortunately, the reverse of this actually happens, and is actually a problem. There's lots of little places where there's a Ford and a Honda and a Toyota dealer, and they're all owned by the same guy. It's lots easier to maintain this kind of local monopoly than it would be to sustain collusion between manufacturers.
And if the different manufacturers actually wanted to collude, they still obviously, obviously could as they still set the wholesale prices.
Wow what a bizarre, moronic comment.
as they won't have the cost of running their own dealers
So dealerships are actually money losers, eh? And the people who own them are just so committed to their LOVE of community that they run them anyway.
the current method is far more efficient
So having independent dealerships would just save Tesla rafts of money, but they don't want to do it? And so we need laws, to what, protect Tesla from making such a bad decision?
Why don't we let Tesla decide how Tesla wants to sell cars? I'm sure lots of other companies were worried when Apple starting setting up their own stores. But the right response isn't "NOOO, we can't let Apple do that because, uh... they'll lose money having to operate all those stores - we're really saving Apple from themselves". And it's certainly not "oh, well, those Best Buy salesmen knew their community so much better, how could Apple people ever understand unique needs?".
If you think you're right, you should have no objection to letting Tesla try this, find out they're wrong and then switching.
But you know you're not right. Obviously.
Look for Intel NUC boxes - you can get a reasonably powerful computer in a very small box. They're expensive, and they need a high speed fan to keep them running, but otherwise are pretty cool.