Most laser printers do contain processors that are far beyond the capability allowed to pass through the embargo. Desperate people become very resourceful.
That reminds me of an old story about one of the early computer science research labs. Management didn't trust these upstarts with a large budget, so their actual computers were old and slow, but were happy to pay for the top-of-the-range printers, since they could understand those and the accountants etc. wanted the print speed. One of the researchers realised that these printers contained much faster processors than the computers, so rather than using the slow machines, everyone rewrote their programs in PostScript and ran them on the printers.
And what the fuck is "Letter" sized paper supposed to mean? Would you know what I was talking about if I said "Foolscap"? No self-respecting geek would use the arcane US sizing system over the international standard one.
Nothing was stopping anyone from forking XFree86, either, and they did.
But sadly it took them messing up the licensing to do that. Sun's not quite that stupid, so I can easily see OOo staggering on forever, sucking up 10x the resources that it would take to make a competitor (KOffice?) or fork better.
This is a problem for all open source projects. Once any project gets above a certain size, it becomes difficult for casual developers to make contributions. This is why open source and UNIX grew so well together - the UNIX philosophy was to have simple tools doing one thing well.
It's not just that, it's also about writing good, clean code. KDE is huge and quite tightly integrated - but it's a joy to work on, even as a new developer, because everything is clearly written and well documented. OOo inherited a huge lump of gnarly code that had never been intended to be open, and so it doesn't respond well to it. I'd argue the same is true of Mozilla, which despite receiving, as far as I can judge from the news stories, more money than any other open source project, still can't get a hold on its memory use. Wheras koffice and khtml manage to be much cleaner and more efficient, and while features may be currently missing they tend to get added at an astonishing rate when a developer turns themselves to it; I'd love to see what those projects could do with half the support and attention.
But yet who in their right mind would (all other things being equal) set up a new system with 2.4 instead of some kernel released this year?
People who care about stability. Honestly, linux seems to have gone backwards rather than forwards with 2.6 - I've been forced to "upgrade" some of my systems because of what the distributions support, but as far as I can see it's less stable and has broken a number of drivers I used to use.
3. Better video playback software. Even though mplayer and vlc are ported to Mac OS and MS Windows, they work best on linux e.g. smoother playback and response to controls, better OS integration, current feature set closer to developer feature set, etc.
But windows still has better players - better user interfaces, better format support, and far better OS integration.
4. Best videogame console emulators. Since many emulators are open source (with notable exceptions) they are primarily developed as linux apps with windows ports that lag behind in features.
Simply false; despite e.g. mupen64 being an open source project, you'll find that it's the linux port that lags in features.
Of course windows and office *default* to beginner-mode - they should, because beginners are the ones who don't change defaults. The most important thing is that the option is there.
No. Theora is not state-of-the-art, but in terms of what the format specifies it's marginally better than mpeg-4 ASP aka DivX/XviD, which is still the most common video format in use today.
I think the real question is have you tried any other browsers?
Yes I have; all the major ones (unless you count safari as major these days) and several that aren't. IE remains perhaps the 3rd-best of them.
Above all else, IE's lack of standards support stands out like a sore thumb. Every time I develop a site for the damn thing, I can't help but notice that simple features like DOM2 and CSS2 still aren't supported even though they are over a DECADE old. That means that sites have to be gimped for IE's substandard support.
Shrug. Be that as it may, as a user I don't really care. I've run across more than enough sites that seem to be developed for firefox-only (as in, they don't render properly in IE, Opera, Konqueror, or sometimes even Seamonkey), and those annoy me a lot more, because there's no defense of ignorance in that case - the developer knew full well that there are multiple browsers in existence, yet still chose to develop for only one of them)
And it shows. Modern websites perform poorly in Internet Explorer, with interfaces that are regularly slower or degraded. (I remember one application that had a column sort. Near instantaneous in Firefox, about 20 seconds in IE. I had to algorithmically optimize the living hell out of it just because IE was sucking so badly. )
Nonsense, frankly, and overgeneralisation. Some javascript is much slower in IE, some is much faster; if you look at actual benchmarks, it's much of a muchness overall.
And don't even get me started at how grating IE7 is on the eyes. I mean, the interface looks like an amateur artist took a dump all over the screen. Ungainly tabs, toolbars in odd places, over the top graphics, etc.
a) All this is subjective b) You're wrong.
Don't even get me started on Microsoft's outright unwillingness to support upcoming standards like HTML 5.
Lies; they've been a major contributor to the standard.
Funny, it never seemed to be a problem 10 years ago when their browser was the only one that implemented the standards.
And we saw where that got them. Maybe, just maybe, it's better to wait for the standards to be finalized and then implement them, rather than leaving sites having to be coded for 400 different partial implementations.
Have you tried a recent IE? It's actually a good browser. Not good enough that I'd switch from opera, but much better than firefox - lower memory use, better looks, more sensibly designed menus...
The drivers for Windows ext2/3 support cause bluescreens under various conditions, so yea those are alpha too.
The open source drivers are alpha quality and unmaintained. That's probably because of the perfectly good working freeware drivers available, which are rock solid.
When your files are more than 16 exabytes, you should probably consider splitting them a bit.
Splitting a file which is logically a single piece of data is never a good thing. Yes, for anything that today would take up 16 exabytes, you probably want to split it up - it'd be a big database or similar - but it seems likely that we'll sooner or later get to a point when you have a 16 EB file that represents a single logical unit. Whether or not it happens with video (I think you were being very conservative in your resolution and especially framerate estimates, but accept the point that there's a lot of room), what about 3D formats? Or the next new technology we can't even think of yet.
Even today, very few files are over 4GB (mostly DV footage and DVD images), and we've had support for those for around a decade now.
And they're still causing trouble; I've lost count of the number of times I've found a useful little AVI utility only available in binary form, and broken it on my larger files. The problem here is mitigated by the fact that one can (in C, at least, where FILE is opaque) just recompile with the correct options. In Java that won't be available to us - the language is fundamentally limited to 64 bits, and unlike C it doesn't have any notion that different implementations may have different integer sizes. It's not an immediate problem, but it's one that bears considering.
But laptops are also designed to suspend when you close a lid - and I hate to break it to you but there are plenty of times I have shifted a laptop around substantially while burning a disc. Usually that worked out just fine.
Well, good for you; not my experience. Twice I've accidentally knocked over a laptop while playing a DVD on it - and both times it scratched the disc to the point where the player gets stuck at that point. Makes me glad I rip everything immediately on purchase.
While this is pretty cool I can't really see the point in it. The facility won't be easy to fit cooling, power and connectivity too and because it's underground there is a significant and on going risk of flooding.
You're forgetting that these things may well have already been dealt with. It was designed to withstand a nuclear war - and to provide a working military command center during one. It will certainly have a serious power supply and an even more serious ventilation system; connectivity may well be a problem, but it's likely to have at least a few lines in and out, which should mean tunnels have already been dug; with luck it may be as simple as threading fibre optics through them.
I would have thought a purpose built above ground facility with soild 5m razor wire topped walls and lots of hungry dogs would have been better.
Sure it would be. It'd also be a lot more expensive - those sorts of facilities are useful to all sorts of organizations. No-one would be considering housing a data center in an underground bunker like this if they had to build it themselves - but they don't. The military has many that it can't really use, and they'll do well to get any money selling them, because really there's very little such a facility can be used for. So the price is low, and that makes it worthwhile.
My university has a plan of potentially buying an old nuclear bunker and making it the backup data center. There're a few concerns about being able to get enough cabling in though.
Everyone is wining on Apple to write drives for every thinkable gadget out there when it should be pretty obvious to ask the manufacturer of that gadget to do just that. Is this so hard?! It's not Apple's fault nor responsibility that MP3 player X doesn't integrate with iTunes, or cell phone Y with iSync, or video card Z
And yet people use those same arguments as the reason Linux isn't ready...
Slashdot Mirror
I've found it's the only way to get a browser that can show Flash without crashing. I wish I was joking.
That reminds me of an old story about one of the early computer science research labs. Management didn't trust these upstarts with a large budget, so their actual computers were old and slow, but were happy to pay for the top-of-the-range printers, since they could understand those and the accountants etc. wanted the print speed. One of the researchers realised that these printers contained much faster processors than the computers, so rather than using the slow machines, everyone rewrote their programs in PostScript and ran them on the printers.
And what the fuck is "Letter" sized paper supposed to mean? Would you know what I was talking about if I said "Foolscap"? No self-respecting geek would use the arcane US sizing system over the international standard one.
Huh? Hp does effective, nonintrusive drivers that sit there quietly and work; seriously, they're one of the best on this.
But sadly it took them messing up the licensing to do that. Sun's not quite that stupid, so I can easily see OOo staggering on forever, sucking up 10x the resources that it would take to make a competitor (KOffice?) or fork better.
It's not just that, it's also about writing good, clean code. KDE is huge and quite tightly integrated - but it's a joy to work on, even as a new developer, because everything is clearly written and well documented. OOo inherited a huge lump of gnarly code that had never been intended to be open, and so it doesn't respond well to it. I'd argue the same is true of Mozilla, which despite receiving, as far as I can judge from the news stories, more money than any other open source project, still can't get a hold on its memory use. Wheras koffice and khtml manage to be much cleaner and more efficient, and while features may be currently missing they tend to get added at an astonishing rate when a developer turns themselves to it; I'd love to see what those projects could do with half the support and attention.
But yet who in their right mind would (all other things being equal) set up a new system with 2.4 instead of some kernel released this year?
People who care about stability. Honestly, linux seems to have gone backwards rather than forwards with 2.6 - I've been forced to "upgrade" some of my systems because of what the distributions support, but as far as I can see it's less stable and has broken a number of drivers I used to use.
Or perhaps that there is a moral distinction to be made between secured and unsecured loans?
But windows still has better players - better user interfaces, better format support, and far better OS integration.
4. Best videogame console emulators. Since many emulators are open source (with notable exceptions) they are primarily developed as linux apps with windows ports that lag behind in features.
Simply false; despite e.g. mupen64 being an open source project, you'll find that it's the linux port that lags in features.
Kopete, dude. It works, simple as that. Always annoys me when people annoy these apps just because they're KDE.
Of course windows and office *default* to beginner-mode - they should, because beginners are the ones who don't change defaults. The most important thing is that the option is there.
No. Theora is not state-of-the-art, but in terms of what the format specifies it's marginally better than mpeg-4 ASP aka DivX/XviD, which is still the most common video format in use today.
Yes I have; all the major ones (unless you count safari as major these days) and several that aren't. IE remains perhaps the 3rd-best of them.
Above all else, IE's lack of standards support stands out like a sore thumb. Every time I develop a site for the damn thing, I can't help but notice that simple features like DOM2 and CSS2 still aren't supported even though they are over a DECADE old. That means that sites have to be gimped for IE's substandard support.
Shrug. Be that as it may, as a user I don't really care. I've run across more than enough sites that seem to be developed for firefox-only (as in, they don't render properly in IE, Opera, Konqueror, or sometimes even Seamonkey), and those annoy me a lot more, because there's no defense of ignorance in that case - the developer knew full well that there are multiple browsers in existence, yet still chose to develop for only one of them)
And it shows. Modern websites perform poorly in Internet Explorer, with interfaces that are regularly slower or degraded. (I remember one application that had a column sort. Near instantaneous in Firefox, about 20 seconds in IE. I had to algorithmically optimize the living hell out of it just because IE was sucking so badly. )
Nonsense, frankly, and overgeneralisation. Some javascript is much slower in IE, some is much faster; if you look at actual benchmarks, it's much of a muchness overall.
And don't even get me started at how grating IE7 is on the eyes. I mean, the interface looks like an amateur artist took a dump all over the screen. Ungainly tabs, toolbars in odd places, over the top graphics, etc.
a) All this is subjective b) You're wrong.
Don't even get me started on Microsoft's outright unwillingness to support upcoming standards like HTML 5.
Lies; they've been a major contributor to the standard.
Funny, it never seemed to be a problem 10 years ago when their browser was the only one that implemented the standards.
And we saw where that got them. Maybe, just maybe, it's better to wait for the standards to be finalized and then implement them, rather than leaving sites having to be coded for 400 different partial implementations.
Have you tried a recent IE? It's actually a good browser. Not good enough that I'd switch from opera, but much better than firefox - lower memory use, better looks, more sensibly designed menus...
The open source drivers are alpha quality and unmaintained. That's probably because of the perfectly good working freeware drivers available, which are rock solid.
It's mature and stable by now - unlike any newer MS server OS.
When your files are more than 16 exabytes, you should probably consider splitting them a bit.
Splitting a file which is logically a single piece of data is never a good thing. Yes, for anything that today would take up 16 exabytes, you probably want to split it up - it'd be a big database or similar - but it seems likely that we'll sooner or later get to a point when you have a 16 EB file that represents a single logical unit. Whether or not it happens with video (I think you were being very conservative in your resolution and especially framerate estimates, but accept the point that there's a lot of room), what about 3D formats? Or the next new technology we can't even think of yet.
Even today, very few files are over 4GB (mostly DV footage and DVD images), and we've had support for those for around a decade now.
And they're still causing trouble; I've lost count of the number of times I've found a useful little AVI utility only available in binary form, and broken it on my larger files. The problem here is mitigated by the fact that one can (in C, at least, where FILE is opaque) just recompile with the correct options. In Java that won't be available to us - the language is fundamentally limited to 64 bits, and unlike C it doesn't have any notion that different implementations may have different integer sizes. It's not an immediate problem, but it's one that bears considering.
Well, good for you; not my experience. Twice I've accidentally knocked over a laptop while playing a DVD on it - and both times it scratched the disc to the point where the player gets stuck at that point. Makes me glad I rip everything immediately on purchase.
You're forgetting that these things may well have already been dealt with. It was designed to withstand a nuclear war - and to provide a working military command center during one. It will certainly have a serious power supply and an even more serious ventilation system; connectivity may well be a problem, but it's likely to have at least a few lines in and out, which should mean tunnels have already been dug; with luck it may be as simple as threading fibre optics through them.
I would have thought a purpose built above ground facility with soild 5m razor wire topped walls and lots of hungry dogs would have been better.
Sure it would be. It'd also be a lot more expensive - those sorts of facilities are useful to all sorts of organizations. No-one would be considering housing a data center in an underground bunker like this if they had to build it themselves - but they don't. The military has many that it can't really use, and they'll do well to get any money selling them, because really there's very little such a facility can be used for. So the price is low, and that makes it worthwhile.
My university has a plan of potentially buying an old nuclear bunker and making it the backup data center. There're a few concerns about being able to get enough cabling in though.
And yet people use those same arguments as the reason Linux isn't ready...
They're following the LSB standard. Maybe you should get a distro that does.
You've been able to do applets in a much better, open language - TCL - for years. Since before you could in Java, in fact.
With today's standards, sure. But what about in the future when that file's just a high-quality movie, say?
It's not that surprising when you think about it. Which one do you think uses more undocumented MS-only APIs?