3.0 was supposed to introduce a stateless API, but didn't. Now 4.0 apparently hasn't either. Have they decided that it's a bad idea, or that it's too difficult, or what?
Having the API retain state is a fundamentally bad idea. As one overview points out, "Nearly all of OpenGL state may be queried". (emphasis added)
It would be much better if there were OpenGL context objects that encapsulated the state, and were explicitly passed into API calls. I was completely dumbfounded when I first looked at API and saw that it didn't work that way.
I doubt that you could do it even then. The drive hardware most likely isn't even *capable* of recording raw ones and zeros on the media, since that's not how your data is stored. Aside from the interleaved Reed-Solomon error correction, the drive uses PRML (Partial Response Maximum Likelihood) to record multiple bits simultaneously as an analog waveform. There isn't a one-to-one correspondence between a data bit and a flux transition on the medium. This is all done by the drive hardware, so even rewriting the firmware isn't going to circumvent it.
I can't RTFA because it seems to be slashdotted, but what on earth led Mr. Huang to think that Kingston made their own chips? There are only a few companies that make NAND flash chips, Sandisk and Toshiba among them, and ALL of the other vendors of flash memory cards have to buy from those few companies. The same is true of DRAM; Kingston DIMMs use other vendors' DDR memory chips.
The fact that Kingston was using chips from Sandisk and Toshiba would normally make me MORE inclined to buy Kingston cards, as usually the quality of Sandisk and Toshiba chips is quite good, though it doesn't explain why he's having trouble with them.
because the signals that go between a phone and a SIM don't have ANYTHING to do with the signals needed to pass data between cellular and anything else. The connection to the SIM is a very low bandwidth connection, only a few kilobits per second, because all it does is store the subscriber idenity (IMSI), PIN, and phone book. When the phone is doing EDGE, 3G, or other high bandwidth data services, none of that data normally goes anywhere near the SIM.
It might be possible to engineer something like this that would have higher bandwidth to the phone, but only by also specially engineering the phone. It's not something that could work with existing phones that are designed for normal SIM cards.
As cool as it would be if this were real, I don't see how it can be. Seems like a marketing idea, not an engineered product.
Which ones are those? I haven't found any; I've only seen sexually frustrating websites. It's less satisfying than going to a strip club or even just "reading" Playboy.
There may eventually be private launches for NASA manned missions to ISS, but it looks pretty unlikely that there will be any NASA manned missions to anywhere else.
...of having NASA do unmanned stuff and private industry do manned. Manned is far more challenging, and less likely to be profitable, so I would have expected it to make sense for NASA to do manned and private industry to do unmanned.
That's just an observation. It's not intended to be criticism of the plan. I have plenty of criticism of the old plan, but I don't yet know enough about the new one.
OK, the people in TFA used 176 FPGAs rather than 1856 ASICs, which is presumably slightly cheaper.
FPGAs of any respectable size are going to be more expensive than the ASICs. The point of using FPGAs isn't cost, but rather that if next week you want to brute-force a different cipher, or do protein folding, or some other massively parallel task, the FPGAs can do it, while the ASICs can't.
I'm sure there's some of that, but I personally know a number of parents that are home schooling because they don't believe that their children will get a good education in the public schools, and not out of any desire to keep their children out of classes that might conflict with religious dogma. Among other things, California is trying to disallow home schooling unless a parent has teaching credentials. I can understand why someone who hasn't throught it through might think that is a good idea, but in practice a teaching credential has nothing whatsoever to do with whether a person is capable of properly teaching. I think it's far better to simply require that the home-schooled students, along with the state- and privately-schooled students, take standardized tests to determine their educational progress. Even that isn't a perfect solution; you can't ban home-schooling because home-schooled Johnny scored below a threshold in some subject any more than you can disqualify a public school teacher because his or her student scored below the threshold.
My biggest concern about home schooling when it is done properly (appropriate curriculum, etc.) is the effect of the student not getting as much socialization with peers.
One of my friends who is an aerospace engineer for a major defense contractor was home schooled, and as far as I can tell his K-12 education was better than mine. He doesn't seem to have suffered from any socialization deficit.
But will the US offer asylum to parents from California that want to home-school their kids?
While my question is obviously somewhat facetious, California is making it extremely difficult if not impossible to legally home-school children.
The retail price is primarily determined by what the market will pay for the unit. It's just that simple.
Actually it ISN'T that simple, and Economics 101 will teach you that. The retail price isn't determined by what the market will pay. It is determined by the INTERSECTION of the supply and demand curves, and the cost to make a product is usually a major factor in the supply curve. If the demand curve doesn't change, but the cost of making the product changes, the supply curve will change, thus the intersection of the supply and demand curves will change, thus the retail price will change.
The next step in refining the analysis would be to consider elasticity of demand. The PC market is fairly elastic, but not so close to perfectly elastic to render the cost of product completely irrelevant.
Though not for the same reason. You get a complete PC for less than a thin client because complete PCs are made in insanely high volumes compared to thin clients, which are a niche item.
You're suggesting that COFEE is able to distinguish criminal activity from non-criminal, and only allow law enforcement to use it to get forensics for criminal activity? That's some pretty f*&#ing amazing artificial intelligence in there, since the law enforcement officers themselves often have a very hard time making that distinction and it ends up having to be settled by courts.
I am confused as to how google and amazon got rights to all these books. Doesn't this violate copyright laws. If I create a book will google and amazon get it without my knowledge.
If it's open source, the only enforcement they'll have over things like hard drives being banned, screen size restrictions, only web apps, etc. will be control of their trademarks. If Chrome offers something sufficiently compelling that people want to run it on "noncompliant" hardware, or run non-web-apps, they will fork it.
It's more of a risk because a package can install setuid binaries, or install config files in directories such that they that are used or interpreted by processes running as another user or root. Installing a package can do a lot more than you can do as an unprivileged user.
Not a single claim of this patent would be applicable to sudo. The independent claims 1, 2, and 9 don't apply to sudo, because they describe significant behavior that is not part of sudo. By extension, none of the dependent claims can apply to sudo either.
You can still argue over whether it meets the obviousness criterion, but trying to spin this a "Microsoft patents sudo" is deliberately spreading FUD.
Slashdot Mirror
Even though I use Gnome, I really don't especially care whether 2.30 is an April Fool's Day prank or not.
In some cases, probably a lot worse than the first time you did it. horrific beetle sex
Fluorinert is not mineral oil, nor even very similar to mineral oil.
Having the API retain state is a fundamentally bad idea. As one overview points out, "Nearly all of OpenGL state may be queried". (emphasis added)
It would be much better if there were OpenGL context objects that encapsulated the state, and were explicitly passed into API calls. I was completely dumbfounded when I first looked at API and saw that it didn't work that way.
However, AFAICT the courts have never found any defendant to be not guilty as a result of the "Other Rights, Etc., Not Affected" clause.
Android has most of the standard Java classes other than AWT/Swing. I was pleasantly surprised to find that it even has the Java reflection API.
Yes. Are you actually surprised?
I doubt that you could do it even then. The drive hardware most likely isn't even *capable* of recording raw ones and zeros on the media, since that's not how your data is stored. Aside from the interleaved Reed-Solomon error correction, the drive uses PRML (Partial Response Maximum Likelihood) to record multiple bits simultaneously as an analog waveform. There isn't a one-to-one correspondence between a data bit and a flux transition on the medium. This is all done by the drive hardware, so even rewriting the firmware isn't going to circumvent it.
Plenty of nine-track tape was still in use on mainframes in the 1980s.
The fact that Kingston was using chips from Sandisk and Toshiba would normally make me MORE inclined to buy Kingston cards, as usually the quality of Sandisk and Toshiba chips is quite good, though it doesn't explain why he's having trouble with them.
It might be possible to engineer something like this that would have higher bandwidth to the phone, but only by also specially engineering the phone. It's not something that could work with existing phones that are designed for normal SIM cards.
As cool as it would be if this were real, I don't see how it can be. Seems like a marketing idea, not an engineered product.
Apple automatically rejects all applications that are NOT submitted. However, they don't notify the non-submitter when this happens.
Which ones are those? I haven't found any; I've only seen sexually frustrating websites. It's less satisfying than going to a strip club or even just "reading" Playboy.
There may eventually be private launches for NASA manned missions to ISS, but it looks pretty unlikely that there will be any NASA manned missions to anywhere else.
That's just an observation. It's not intended to be criticism of the plan. I have plenty of criticism of the old plan, but I don't yet know enough about the new one.
FPGAs of any respectable size are going to be more expensive than the ASICs. The point of using FPGAs isn't cost, but rather that if next week you want to brute-force a different cipher, or do protein folding, or some other massively parallel task, the FPGAs can do it, while the ASICs can't.
My biggest concern about home schooling when it is done properly (appropriate curriculum, etc.) is the effect of the student not getting as much socialization with peers.
One of my friends who is an aerospace engineer for a major defense contractor was home schooled, and as far as I can tell his K-12 education was better than mine. He doesn't seem to have suffered from any socialization deficit.
But will the US offer asylum to parents from California that want to home-school their kids? While my question is obviously somewhat facetious, California is making it extremely difficult if not impossible to legally home-school children.
Actually it ISN'T that simple, and Economics 101 will teach you that. The retail price isn't determined by what the market will pay. It is determined by the INTERSECTION of the supply and demand curves, and the cost to make a product is usually a major factor in the supply curve. If the demand curve doesn't change, but the cost of making the product changes, the supply curve will change, thus the intersection of the supply and demand curves will change, thus the retail price will change.
The next step in refining the analysis would be to consider elasticity of demand. The PC market is fairly elastic, but not so close to perfectly elastic to render the cost of product completely irrelevant.
Though not for the same reason. You get a complete PC for less than a thin client because complete PCs are made in insanely high volumes compared to thin clients, which are a niche item.
You're suggesting that COFEE is able to distinguish criminal activity from non-criminal, and only allow law enforcement to use it to get forensics for criminal activity? That's some pretty f*&#ing amazing artificial intelligence in there, since the law enforcement officers themselves often have a very hard time making that distinction and it ends up having to be settled by courts.
I am confused as to how google and amazon got rights to all these books. Doesn't this violate copyright laws. If I create a book will google and amazon get it without my knowledge.
All your books are belong to Google!
If it's open source, the only enforcement they'll have over things like hard drives being banned, screen size restrictions, only web apps, etc. will be control of their trademarks. If Chrome offers something sufficiently compelling that people want to run it on "noncompliant" hardware, or run non-web-apps, they will fork it.
It's more of a risk because a package can install setuid binaries, or install config files in directories such that they that are used or interpreted by processes running as another user or root. Installing a package can do a lot more than you can do as an unprivileged user.
You can still argue over whether it meets the obviousness criterion, but trying to spin this a "Microsoft patents sudo" is deliberately spreading FUD.