Say what you like, but Itanium was a nice architecture. The compiler is the proper place for the optimisations, the processor should be left to do the actual processing. It's still the most efficient way I know to do raytracing or anything multimediay, and I predict there will still be a market for them for some time.
On second thought, maybe they'll start appearing cheaply on ebay. That'd be nice.
It matters for precise physics. The reason it should be cleaned up is simply to stop it becoming 5E-6% smaller every year. People who don't need precision can still use artefacts, but some of us do need it.
The only factor that makes sense to use is a factor of "10". Why? So you can take zeroes off and add them to convert. It doesn't matter whether this "10" is actually 0xC or 8 or whatever, but the conversion should be a factor of 10. Until people are willing to junk the arabic numerals and go for hex or sexagesimal or something sensible like that, metric is the only system that makes sense.
I think it's getting there. As a product of this schooling, I'm mentally fluent in both systems of weight, and that's only because of cooking with my mother. Ask me how long something is and I'll tell you in metres. To me 70 degrees is the temperature of a cup of tea. It will take time, sure. But the switch is happening. Half of my class didn't know how many pounds there were to a stone.
Your pint of beer is rarely close to a full pint, it's just sold that way. A litre of milk is right for a couple. A kilo of sugar or flour is exactly the right size, and about all that gets sold. Seconds are the same in either. A metre is far closer to human-sized than a foot, and about the right height for some pieces of furniture. 100m is the length of my road. I know when I've cycled a kilometre. I think you're just seeing your units as more human because you're used to them. If you used the SI units, you'd find a lot of things matched up to them.
In any case, what you're referring to is utterly trivial and/or irrelevant when it comes to the real work of defining the units. Any definition of the gram suffices to define the kilogram, and vice-versa.
It does cause real problems though. I've got more than one physics question wrong because I assumed a mass of 6 miligrams meant I could put 6x10^-3 into my equations. To avoid confusion the new base unit should have a different name, let's say (since I can't think of any scientists right now) the Taco. It would be approximately equal to a kilogram, with current grams becoming militacos and the metric Tonne being a kilotaco.
Because it's overused and really horrible to program in. At least for me.
The right tool for the right job
Yes, and java has its place, like haskell (which I like, but would never try to do everything in). But its shoehorned into everything, when it's too heavily OO to be a good general-purpose language.
Java has to be doing something right
Yes. Marketing. Buzzwords. Making PHBs think they need to be using it for everything.
Developing for java is actually a pleasure since amazing IDE's like Eclipse and NetBeans have been released.
It's never been a pleasure for me, and believe me I've tried. I want to like java, really I do. It would make finding a good programming job so much easier. But every time I try to use it it's an utterly horrible experience.
cell phone emulators and the works
And you'd have thought cellphones with their resource limitations were were native optimisations were needed the most. But no, they have to have java on them. And the same for the OS.
With today's VMs I often find java is faster or on par with C++
I get shouted down as a troll whenever I say that java GUI apps are incredibly slow, but it's true. I have never seen a graphical java program that will start anywhere near native speed. In the time it takes to start jedit I can have emacs, kwrite and gvim all up and running. Same for yaggui compared to starting apollon, giftui and giftcurs, and yaggui isn't even particularly featureful. I'll grant that Azureus is a pretty nice bittorrent client, but that doesn't excuse a 20s startup time when the official client starts in 5. And when they are running, there is a perceptible delay between clicking a menu and having it open up - the only other place I see this is windows programs running under wine. Most people presumably have faster machines than the ones I'm using (650mhz celeron w/ 64mb ram and 800mhz duron/256mb ram), and so this is not a problem for them. But it is a genuine problem for me, and I simply can't afford better machines. I'll grant that java console apps don't seem to be that much slower than anything else once they've actually started up.
It really is a nice language
That hasn't been my experience at all, and I've tried to learn it and like it on three separate occasions. It seems to combine the simplicity and programming ease of C with the performance of python or lisp. The stupidly verbose syntax, the need to OO-ify everything, all the places where it's identical to C++ except for one critical difference, the confusing gui library, the horribly bloated installation... it's just been horrible.
I'll restate what I said earlier, but perhaps more in context. Unless you're one of about ten people, the compiler probably knows the language better than you do. So don't bother trying to outguess it. If you really need it super-hand-optimised, inline assembly. But don't try to write optimized C, other than at the algorithm level. Every possible C implementation of a given algorithm is likely to compile to more or less the same.
I thought it was used as an example of algorithm improvements - see, now you have the quick sort and it runs 5000 tiems as fast. If people are actually teaching that as an algorithm to be used they should be shot, of course.
A discreet logo. Something equivalent to those W3C validated buttons people use. Something that says you're using SSL and someone halfway competent has taken a look at it and it seems set up ok. I'm not sure if such a thing exists, but if it doesn't it should do. That's the sort of thing I'd want.
I don't see it as rude, on the contrary I feel they're trying to show they're part of the community. I get distrustful when a company proclaims it's releasing something as open source, but it's under some weird-ass license I've never seen before and only available from their site. Putting it on sourceforge shows it's the real deal.
This particular check can be beaten pretty easily. But the opinion of the wine devs is that that would start an "arms race" they can't win. They already know of one way of checking that would be impossible to fool without rewriting large parts of wine, and if they can think of it then MS with its much greater resources has almost certainly found it.
WTF? If they wanted to concentrate on office then they absolutely would not do this, because it just makes it harder to use office under linux. This smacks of desperation, or at least panic. They know they're slowly crumbling on the OS front, and their reaction is to take away any support they had for other platforms to try and stop people moving.
It's not windows that needs updating, but the applications. You don't need updates for windows itself, but if you're running Office under wine you'll need upgrades for Office, and so on.
That's odd, I've found muse to be far less stable than rosegarden when I'm using it. Plus it's completely jack-based, which has never been stable on my system.
Why can't it? USB2 has significantly more bandwidth than the older form of FireWire, which was seen as good enough before the new version existed. The whole point of USB is that it is Universal. You don't need separate plugs for your keyboard, mouse, printer, modem - everything can connect to USB. Why not digital video? I transfer the fairly low quality videos I make with my digital camera over old USB, with no problems. Why is it so impossible to use USB for video. You seem to imply that USB is less "reliable" in your post: how, and why?
Slashdot Mirror
On second thought, maybe they'll start appearing cheaply on ebay. That'd be nice.
It matters for precise physics. The reason it should be cleaned up is simply to stop it becoming 5E-6% smaller every year. People who don't need precision can still use artefacts, but some of us do need it.
The only factor that makes sense to use is a factor of "10". Why? So you can take zeroes off and add them to convert. It doesn't matter whether this "10" is actually 0xC or 8 or whatever, but the conversion should be a factor of 10. Until people are willing to junk the arabic numerals and go for hex or sexagesimal or something sensible like that, metric is the only system that makes sense.
I think it's getting there. As a product of this schooling, I'm mentally fluent in both systems of weight, and that's only because of cooking with my mother. Ask me how long something is and I'll tell you in metres. To me 70 degrees is the temperature of a cup of tea. It will take time, sure. But the switch is happening. Half of my class didn't know how many pounds there were to a stone.
Your pint of beer is rarely close to a full pint, it's just sold that way. A litre of milk is right for a couple. A kilo of sugar or flour is exactly the right size, and about all that gets sold. Seconds are the same in either. A metre is far closer to human-sized than a foot, and about the right height for some pieces of furniture. 100m is the length of my road. I know when I've cycled a kilometre. I think you're just seeing your units as more human because you're used to them. If you used the SI units, you'd find a lot of things matched up to them.
It does cause real problems though. I've got more than one physics question wrong because I assumed a mass of 6 miligrams meant I could put 6x10^-3 into my equations. To avoid confusion the new base unit should have a different name, let's say (since I can't think of any scientists right now) the Taco. It would be approximately equal to a kilogram, with current grams becoming militacos and the metric Tonne being a kilotaco.
If that were the case how could these treaties e.g. Kyoto be binding on countries which haven't ratified them (US)?
Because it's overused and really horrible to program in. At least for me.
The right tool for the right job
Yes, and java has its place, like haskell (which I like, but would never try to do everything in). But its shoehorned into everything, when it's too heavily OO to be a good general-purpose language.
Java has to be doing something right
Yes. Marketing. Buzzwords. Making PHBs think they need to be using it for everything.
Developing for java is actually a pleasure since amazing IDE's like Eclipse and NetBeans have been released.
It's never been a pleasure for me, and believe me I've tried. I want to like java, really I do. It would make finding a good programming job so much easier. But every time I try to use it it's an utterly horrible experience.
cell phone emulators and the works
And you'd have thought cellphones with their resource limitations were were native optimisations were needed the most. But no, they have to have java on them. And the same for the OS.
With today's VMs I often find java is faster or on par with C++
I get shouted down as a troll whenever I say that java GUI apps are incredibly slow, but it's true. I have never seen a graphical java program that will start anywhere near native speed. In the time it takes to start jedit I can have emacs, kwrite and gvim all up and running. Same for yaggui compared to starting apollon, giftui and giftcurs, and yaggui isn't even particularly featureful. I'll grant that Azureus is a pretty nice bittorrent client, but that doesn't excuse a 20s startup time when the official client starts in 5. And when they are running, there is a perceptible delay between clicking a menu and having it open up - the only other place I see this is windows programs running under wine. Most people presumably have faster machines than the ones I'm using (650mhz celeron w/ 64mb ram and 800mhz duron/256mb ram), and so this is not a problem for them. But it is a genuine problem for me, and I simply can't afford better machines. I'll grant that java console apps don't seem to be that much slower than anything else once they've actually started up.
It really is a nice language
That hasn't been my experience at all, and I've tried to learn it and like it on three separate occasions. It seems to combine the simplicity and programming ease of C with the performance of python or lisp. The stupidly verbose syntax, the need to OO-ify everything, all the places where it's identical to C++ except for one critical difference, the confusing gui library, the horribly bloated installation... it's just been horrible.
I'll restate what I said earlier, but perhaps more in context. Unless you're one of about ten people, the compiler probably knows the language better than you do. So don't bother trying to outguess it. If you really need it super-hand-optimised, inline assembly. But don't try to write optimized C, other than at the algorithm level. Every possible C implementation of a given algorithm is likely to compile to more or less the same.
As you should do. Really, assembler is sometimes the best tool for the job, and this sounds like a case where it is.
I thought it was used as an example of algorithm improvements - see, now you have the quick sort and it runs 5000 tiems as fast. If people are actually teaching that as an algorithm to be used they should be shot, of course.
Believe it or not, I posted this in links, and I recommend it. Means you don't get any annoying flash ads
If it needs optimization so much that you can't trust the compiler to handle it, you should be coding it in assembly.
A discreet logo. Something equivalent to those W3C validated buttons people use. Something that says you're using SSL and someone halfway competent has taken a look at it and it seems set up ok. I'm not sure if such a thing exists, but if it doesn't it should do. That's the sort of thing I'd want.
You kidding? Just find the +5 insightful comments from the last SCO thread and copy them. Mods can't tell the difference anyway.
As a python guy who hates Java, does this mean I should be supporting PHP? Or are we sworn enemies as well?
I don't see it as rude, on the contrary I feel they're trying to show they're part of the community. I get distrustful when a company proclaims it's releasing something as open source, but it's under some weird-ass license I've never seen before and only available from their site. Putting it on sourceforge shows it's the real deal.
Yeah, sure, just like (L)ame (A)in't an (M)P3 (E)ncoder.
This particular check can be beaten pretty easily. But the opinion of the wine devs is that that would start an "arms race" they can't win. They already know of one way of checking that would be impossible to fool without rewriting large parts of wine, and if they can think of it then MS with its much greater resources has almost certainly found it.
WTF? If they wanted to concentrate on office then they absolutely would not do this, because it just makes it harder to use office under linux. This smacks of desperation, or at least panic. They know they're slowly crumbling on the OS front, and their reaction is to take away any support they had for other platforms to try and stop people moving.
It's not windows that needs updating, but the applications. You don't need updates for windows itself, but if you're running Office under wine you'll need upgrades for Office, and so on.
Microsoft releases updates and patches for things like Office. And you can install them fine under wine, and they work.
Are you sure it isn't? I clearly remember reading it a week or so ago. That was probably at OSnews or something though.
That's odd, I've found muse to be far less stable than rosegarden when I'm using it. Plus it's completely jack-based, which has never been stable on my system.
Why can't it? USB2 has significantly more bandwidth than the older form of FireWire, which was seen as good enough before the new version existed. The whole point of USB is that it is Universal. You don't need separate plugs for your keyboard, mouse, printer, modem - everything can connect to USB. Why not digital video? I transfer the fairly low quality videos I make with my digital camera over old USB, with no problems. Why is it so impossible to use USB for video. You seem to imply that USB is less "reliable" in your post: how, and why?