You don't need any LLVM code in the compiler. It might be easier to have it there, but it's not a requirement - you can just output LLVM assembly and call external tools on that.
My point was more that if you might just be better off in the long run by starting from scratch, instead of taking on the maintenance nightmare that is gcc.
Then, when we realized it was a problem, we went totally arse over teakettle: banned the stuff completely and pressured other countries to do the same, rather than realizing that it was the irresponsible use that was really to blame, and that there were parts of the world where any rational cost/benefit analysis still called for it. No, not quite. You're falling for FUD spread by chemical companies.
Now, I did not read the actual emails yet beyond the summary, but that contains a hint most people are missing. The goal of this isn't to entrap people for downloading material. That idea is dumb, everybody knows it, and these people aren't so dumb they think that would work.
No, what I see hints of is that their client would contain code to disrupt OTHER P2P networks. Their efforts to disrupt traffic are easily thwarted by blocking their IP ranges. What they might be going for is creating a botnet of sorts, so they can attack from entirely random IPs.
And that's the thing they don't want getting out at any cost.
Yeah, considering the utter disregard for decent interface design on any level that the GIMP team has shown in the past, I'm not really holding out much hope for this one. Perhaps we'll get a new coat of paint on top of the old interface, but the whole thing will still be a horrid programmer-interface mess.
Or perhaps they will really create a competent design team and let them dictate every detail of the interface. But with the usual open source ego contests, that seems a tad unlikely.
"largely settled matters"... in 1404, a flat Earth was a "largely settled matter" Yes, it was largely a settled matter that the Earth was not flat, but round. This was known since antiquity.
That's simply untrue. There is far, far less work involved in making a binary that runs on multiple versions of Windows than making one that runs on multiple Linux distros. Programs run on multiple Windows versions pretty much by default. The only thing you need to watch out for is using features which were only introduced in new versions. Not so on Linux, where you have a million headaches to deal with before a binary will run across different systems.
Windows is one of the best OSes when it comes to backwards compatibility. It's not perfect, but comparing it to the insane mess that is Linux is just idiotic. Linux can't even manage binary compatibility between current versions, never mind older versions.
You can't go in a straight line - unless you have a thruster that gets you half a gee, in which case, just aim and go. Gravity would cause but a minor correction.
Nowadays, it isn't too hard anymore to create binaries that will work on most recent mainstream distros. Funny how "nowadays" a problem that no other OS but Linux has ever had is not so bad anymore. Still there, mind you, just not as bad.
Maybe the ability to keep his damned mouth shut when he doesn't have anything worthwhile to say? Thaaat would be a solid "no", I'd think. He is well known to be an opinionated asshole.
No, what he has is an urge to actually make something worthwhile, instead of just an ideological imperative. People aren't going to contribute code to an ideology, they are going to contribute to something where their efforts will actually be of use.
Right, "by 2020" we can power a handful of these plants, worldwide. We can't really afford to wait that long. I'd love to see most of our power being solar and wind based, but you've gotta be pragmatic: It ain't happening any decade soon. In the meantime, we still need to get rid of the coal. And that means nuclear.
A nice thought, but:
We list every IP which is known to run a tor server and allow their clients to connect to one of the following ports: 25, 194, 465, 587, 994, 6657, 6660-6670, 6697, 7000-7005, 7070, 8000-8004, 9000, 9001, 9998, 9999 So it seems to be somewhat less effective for web sites, as 80 is not on the list. Of course, maybe it includes most Tor exit nodes anyway, I don't know how many would allow 80 but none of the above.
Well, that's what you end up doing, but that takes a lot of effort and time, and you have to let the attacker keep attacking to find out what the exit nodes are.
Slashdot Mirror
You don't need any LLVM code in the compiler. It might be easier to have it there, but it's not a requirement - you can just output LLVM assembly and call external tools on that.
My point was more that if you might just be better off in the long run by starting from scratch, instead of taking on the maintenance nightmare that is gcc.
With a llvm backend for pcc you could have a good compromise, though.
No. You are falling for anti-environmentalist FUD spread by chemical companies.
http://timlambert.org/2005/10/crime-of-the-century/
http://scienceblogs.com/deltoid/2007/05/this_week_in_the_unending_war.php
And so on.
Big, maybe. Here's an idea I'm just going to throw out here: Miivi wasn't a honeypot. It was an attempt to create a botnet.
http://slashdot.org/comments.pl?sid=299011&cid=20620947
My guess is: No, not entrapment. Botnet.
http://slashdot.org/comments.pl?sid=299011&cid=20620947
Now, I did not read the actual emails yet beyond the summary, but that contains a hint most people are missing. The goal of this isn't to entrap people for downloading material. That idea is dumb, everybody knows it, and these people aren't so dumb they think that would work.
No, what I see hints of is that their client would contain code to disrupt OTHER P2P networks. Their efforts to disrupt traffic are easily thwarted by blocking their IP ranges. What they might be going for is creating a botnet of sorts, so they can attack from entirely random IPs.
And that's the thing they don't want getting out at any cost.
Yeah, considering the utter disregard for decent interface design on any level that the GIMP team has shown in the past, I'm not really holding out much hope for this one. Perhaps we'll get a new coat of paint on top of the old interface, but the whole thing will still be a horrid programmer-interface mess.
Or perhaps they will really create a competent design team and let them dictate every detail of the interface. But with the usual open source ego contests, that seems a tad unlikely.
That's simply untrue. There is far, far less work involved in making a binary that runs on multiple versions of Windows than making one that runs on multiple Linux distros. Programs run on multiple Windows versions pretty much by default. The only thing you need to watch out for is using features which were only introduced in new versions. Not so on Linux, where you have a million headaches to deal with before a binary will run across different systems.
So because you can kludge your way around the problem, it does not exist?
My point was that none of that work is needed on other platforms.
Windows is one of the best OSes when it comes to backwards compatibility. It's not perfect, but comparing it to the insane mess that is Linux is just idiotic. Linux can't even manage binary compatibility between current versions, never mind older versions.
You can't go in a straight line - unless you have a thruster that gets you half a gee, in which case, just aim and go. Gravity would cause but a minor correction.
Tanegashima is "tiny" now? The island's 50 km long. It's got a city on it!
Let me guess - you were saying it was going to fail.
Perhaps he is a sane person, and never looked under "Multilanguage" for his font settings.
No, what he has is an urge to actually make something worthwhile, instead of just an ideological imperative. People aren't going to contribute code to an ideology, they are going to contribute to something where their efforts will actually be of use.
Right, "by 2020" we can power a handful of these plants, worldwide. We can't really afford to wait that long. I'd love to see most of our power being solar and wind based, but you've gotta be pragmatic: It ain't happening any decade soon. In the meantime, we still need to get rid of the coal. And that means nuclear.
25, 194, 465, 587, 994, 6657, 6660-6670, 6697, 7000-7005, 7070, 8000-8004, 9000, 9001, 9998, 9999 So it seems to be somewhat less effective for web sites, as 80 is not on the list. Of course, maybe it includes most Tor exit nodes anyway, I don't know how many would allow 80 but none of the above.
Well, that's what you end up doing, but that takes a lot of effort and time, and you have to let the attacker keep attacking to find out what the exit nodes are.