Krystallnacht is not a good comparison. That was deliberate directed attack by the Nazi's. What response there is tomorrow will be undirected and uncordinated, and unfollowed up.
Did Churchill talking about Checkslovakia stop war? Even short-term (pre-Poland), was the conquest of a soverign country by a racist imperialist dictator a good thing?
But few living pilots are going to deliberately run into the World Trade Center when they can crash it into the bay. A replica handgun won't be good enough, especially if one of the passengers gets the idea you aren't {willing|going} to kill anyone with the guns.
How long should Britian have waited before getting involved in WWI? Should they have let Germany conquer Russia, and murder every Jew and Slav west of France?
When does it become our buisness? Apparently, the mere massacre of innocents does not concern us. Apparently neither does the murder of American citizens (the Lockerbie trail), nor attacks on American embassies, nor bombs on American soil (the earlier World Trade Center bombing.)
When the US should get involved is a good question, but the answer is _not_ never. At the very least, the US has the obligation to protect American citizens, and respond to threats to them, especially on American soil.
> If the US government won't even talk about the possibility that Israel is a racist state, how can they be surprised when those affected by racism strike back?
What the US government won't talk about is condeming Israel for being a racist state, and not talking about most of the rest of the middle east or any other racist countries in the world.
> Don't attack them for their cultural differences. Just because you see some of them celebrating for the camera, doesn't mean they're bad people.
You celebrate the death of thousands of innocents, then you're bad. I don't care about your culture; it's wrong. Part of morality (as dictated by Betham, Kant, Rand, Jesus, Budda) is understanding that we're all human and that murder is wrong.
> This is because it will eliminate the need to transmit "which character set this is in".
On one hand, UTF-8 alone can do that, without this Latin-1 hack. On the other hand, this in no way solves that, as you still have to deal with ISO-8859-(2-16), SJIS and a dozen other character sets that aren't just going to go away.
> At no point does a program have to think about whether a document is ISO-8859-1 or UTF-8, it just assummes UTF-8 at all times.
What makes ISO-8559-1 special? ISO-8859-1 only may as well be ASCII only for most of the world.
> Only for characters with the high bit set. This is not a security problem.
Yes, it is. It's a little more elaborate, but you can still get into cases where one program checks the data, say forbiding access to directory e', but misses the access to e' because it's encoded differently.
> programs are forced to examine the text before displaying it
I prefer my programs to get it right, rather than do it quick.
> because there are a huge number of documents out there with these characters
Plain text documents? Not so much. The number of plain text documents is probably much smaller than by the number of KOI8-R plain text documents.
We have a solution - it's called iconv. iconv will handle all sorts of character sets, not just ISO-8859-1 and CP1252. (Actually, you don't completely handle Latin-1 either, since there are valid Latin-1 files that are valid UTF-8. But we can just handwave that away until someone gets bit by it.)
> the standards organizations were stupid for saying these are control characters, if they had assigned them
They did assign them. Single shift 1, single shift 2, etc. If the characters were needed, that's what they should have been used for, Microsoft be damned.
> UTF-8 and ISO-8859-1 can be supported at the same time for all real documents.
And we can forget about the rest of the world that doesn't use ISO-8859-1 or UTF-8? This will hopelessly munge any documents in a different charset; whereas a UTF-8 interpreter could realize it wasn't a UTF-8 document (and probably load in up in a locale charset.)
> This also makes programming easier because there are no "errors" to worry about.
But you've added another case to everything that has to handle UTF-8, and ignored real errors.
>It is also necessary to treat UTF-8 sequences that code a character in more bytes than necessary as illegal, this is vital for security reasons.
Why? You've just introduced alternate sequences for characters (the whole security problem) in another form.
>I am also in favor of adding MicroSoft's assignments to 0x80-0x9F to the standard Unicode.
Huh? All the characters are in Unicode. To mess with preexisting characters (that date back to Unicode 1.0) to add yet more redundant characters isn't going to happen, and shouldn't; you shouldn't break backward compatibilty unless there's at least a benefit.
Illustrator is fairly specific - there's no Corel Illustrator, or OpenIllustrator. There's a StarOffice, OpenOffice, and WordPerfect Office. Microsoft can't claim Office as a trademark, once everyone's used it on their product.
I once sat down and thought about what I was missing in Linux. BeOS had almost all of it.
BeOS has great font support, and excellent Unicode support. It's very fast, with the main browser (NetPositive) being much faster than Netscape. It had a nice GUI and a 64-bit journaling filesystem years before Linux did. BeOS advocates always went on about how you can play 200 videos at once smoothly. It also has fairly decent POSIX support and includes BASH as the default shell.
It was a very nice system, handicapped by a lack of applications, lack of hardware support and the other stuff that comes with being 4th in the OS market.
Thank you, Jean Dixon. Of course it's up for debate! 18 months is a long time, and a lot of things could happen in that time. Any future prediction on this level is an educated guess, at best.
According to the FBI, the vast majority of child pornography claims are actually perfectly legal pictures with no sexual activity whatsoever. So if you shut down every site someone claims "child porn", you'll be shutting down a lot of legal sites.
How long do you think it would take to shutdown a child porn site, anyway? If you had evidence, I'd be surprised if it took more than 24 hours to get a search warrant and take the server, except _maybe_ over holidays.
I wasn't trying to imply details of the history of GCC, just that GCC and EGCS had merged. In second thought, I'm not sure that EGCS wasn't a GNU compiler - they had the same requirements, including copyright assignment - as GCC. It just wasn't an _official_ GNU compiler.
> You're somewhat correct in the difficulty in maintaining these, but you overestimate the role of the FSF here. Perhaps you should re-read Ulrich Drepper's rant if you want a better idea of reality.
If I, as a student of OSU, publish a paper that means something, it will get noted as coming from OSU, even if I'm leading riots against the adminstration. If you've accepted the title of GNU maintainer and release packages in a GNU manner, you are acting for the FSF and your actions will get correctly attributed to them.
I think you underestimate the FSF. If it wasn't for the FSF, Linux would probably still be using Linux-specific versions of libc and gcc. EGCS possibly wouldn't have happened, and wouldn't have been nearly as successful, if all the threads of GCC development it tried to follow weren't all based off of GNU's GCC 2.7.2.
As for Ulrich Drepper . . . I've read his posts to the gcc list. He gives commands without explanations, often only saying "you don't understand the problems you're getting into", to people with PH.Ds who have been working on GCC for 5-10 years.
Ulrich Drepper complains about losing total control of glibc, and changing a reference in a GNU program to be conformant to GNU standards. He says "NEVER voluntarily put a project you work on under the GNU umbrella since this means in Stallman's opinion that he has the right to make
decisions for the project.". Duh. GNU is basically a trademark. If you want to use a trademark, you've got to pay the price they want. If you feel that's unreasonable, then you can chose not to use the GNU trademark, including forking glibc, if you want. If all RMS asked was that power no longer be centered in one autocrat, and that glibc uses LGPL 2.1 (and that's all Drepper specifically mentioned), then I don't see RMS as being unreasonable here.
If I recall the copyright assignment, anyone who donates the copyright to code to a GNU project to the FSF has a contract say that the FSF cannot release the code under a non-free license (I don't remember the exact wording, but that's the guist of it.) So there's some guarentee.
> You are implying that nobody ever chooses to work a low-income job. This it total nonsense
There exist people who choose to work a low-income job. There also exist people who choose to beaten with pipes. That does not negate the fact that a lot of people don't choose either. I would guess he's talking about them.
Life's not all choice. I was lucky enough to grow up in a (upper?) middle class family with a family who could and would feed my reading and computer using habits. If I had grown up to a poor family in a slum, I possibly wouldn't be at college, and almost certainly wouldn't be typing this message on an expensive computer in my room.
> The gcc used comes from egcs (as does modern gcc), which was forked a long time ago after Stallman installed a bonehead maintainer. It is true GNU software, but most of the code is from non-GNU sources.
EGCS was a short fork that was remerged over a year ago. The GNU project may not have written a lot of the code (any more than the Apache group wrote all of Apache, or Linus wrote all of Linux), but they did the merging and releasing, awesome projects for something as large and complex as GCC. If the FSF had not been there, there would be ten thousand GCCs, with poorly maintained forks for every system out there.
Also, it's a huge project to get any open-source project started. Whatever anyone else may have done to GCC or glibc or gdb, the structure and framework were built by RMS (for GCC and GDB) or by someone working for the FSF (as in getting paid) (for glibc), and that structure was obviously valuable enough to build on.
there is not a single piece of GNU software that is completely essential to Linux - it would be a pain to replace it all (especially the compilers and thier ilk), but it *could* be done.
Replacing the kernel in a Linux system is not a drop-in solution, but changing the kernel to a BSD kernel with Linux emulation, and pretty much only the kernel, could be done in probably under a day - definetly under a week.
How long would it take to replace GCC? Even the BSD's still use GCC! To replace the GNU C compiler for ix86 would probably take years of effort; to replace the GCC (with C, C++, Fortran, Java, and Objective C frontends) for every platform Linux runs on - heck, the GCJ frontend alone took several years by a bunch of full-time programmers. Why do you think so many companies use GCC instead of making thier own compiler? Because it's trivial to duplicate?
Linux is Linux, GNU is a set of mediocre Unix utility ripoffs.
Linux is just a mediocre Unix kernel ripoff, by that measure - there's many "ridiculous" Linux-isms, for example. I've heard many people say the first thing they do on a new Unix (no *) system is install all the GNU utilities, which many people find vastly improved over the Unix utilities. (I've met a couple compiler crashes, but I never seen any sign of instability in the GNU utilities.
Why do you say he dissed goto's? The paper of his I rememeber on the subject, said that they were frequently not the best way to do things, then spent the rest of the paper talking about places where they were the most efficent way to do things.
But a hand-made index is better at finding a subject than searching the text; it doesn't skip synonmys and stop at words in the wrong context. Any decent technical book will hopefully have a handmade index that beats using grep.
I believe that Mozilla includes a built-in demoroniser, by assuming that stuff claiming to be Latin-1 but having characters in the C1 range used by CP 1252, is really CP 1252.
> If you don't know what post-modern rationalism is, do philosophers assume you are dumb?
Some of them will rip you to shreads for not having a basic understanding of philosophy, yes. People in every field don't understand how the general public can be so stupid as not to understand the basics of their subject.
> If you don't know Russian, do Russian professors mock you
Maybe not just Russian, but if you revealed you knew but one language, probably. You've never read people trashing Americans for being monolingugal before?
> Computer nerds are the last bastion of
> unadulterated bigotry
Yeah, whatever. Any discipline are going to have people who think themselves superior for knowing that discipline. English majors don't understand how people can be so poorly read; geography majors can't understand how people don't know where countries are. I complain about many of the customers at Homeland, as do most of my co-workers. Computer nerds are just human, doing what humans do.
American tourists have been arrested in Germany for distributing Nazi propoganda over the internet. In the case I'm thinking of, they actually tried to extradite him first, but the US government refused (but didn't notify him.) I don't remember reading about any governmental objections; appearing to support a Nazi is bad political fodder.
The FSF doesn't give you the right to pull back the copyright. They give you the right to use the code you donated under any license you want, given a month's notice. (They claim the month's notice is merely some sort of legal CYA thing.)
What was so bad about the '99 task? 'cat' was an acceptable solution! Even I was able to toss together a solution, and didn't do half bad. (A couple friends and I are the Icon implementation.)
Now, the '00 task? That was nasty. It looked like you had to have a background in computer graphics to even put together an entry. I couldn't manage to assemble anything for that one.
> Ada83 was not OO in the sense we normally associate with the term -- no inheritance, for example.
Ada83 wasn't OO, but Ada95 is. In any case, Ada83 did have inheritance - type foo is new bar with record... end record; has been in the language since the first. What it did not have was dynamic dispatching.
Slashdot Mirror
Krystallnacht is not a good comparison. That was deliberate directed attack by the Nazi's. What response there is tomorrow will be undirected and uncordinated, and unfollowed up.
Did Churchill talking about Checkslovakia stop war? Even short-term (pre-Poland), was the conquest of a soverign country by a racist imperialist dictator a good thing?
But few living pilots are going to deliberately run into the World Trade Center when they can crash it into the bay. A replica handgun won't be good enough, especially if one of the passengers gets the idea you aren't {willing|going} to kill anyone with the guns.
How long should Britian have waited before getting involved in WWI? Should they have let Germany conquer Russia, and murder every Jew and Slav west of France?
When does it become our buisness? Apparently, the mere massacre of innocents does not concern us. Apparently neither does the murder of American citizens (the Lockerbie trail), nor attacks on American embassies, nor bombs on American soil (the earlier World Trade Center bombing.)
When the US should get involved is a good question, but the answer is _not_ never. At the very least, the US has the obligation to protect American citizens, and respond to threats to them, especially on American soil.
> If the US government won't even talk about the possibility that Israel is a racist state, how can they be surprised when those affected by racism strike back?
What the US government won't talk about is condeming Israel for being a racist state, and not talking about most of the rest of the middle east or any other racist countries in the world.
> Don't attack them for their cultural differences. Just because you see some of them celebrating for the camera, doesn't mean they're bad people.
You celebrate the death of thousands of innocents, then you're bad. I don't care about your culture; it's wrong. Part of morality (as dictated by Betham, Kant, Rand, Jesus, Budda) is understanding that we're all human and that murder is wrong.
> This is because it will eliminate the need to transmit "which character set this is in".
On one hand, UTF-8 alone can do that, without this Latin-1 hack. On the other hand, this in no way solves that, as you still have to deal with ISO-8859-(2-16), SJIS and a dozen other character sets that aren't just going to go away.
> At no point does a program have to think about whether a document is ISO-8859-1 or UTF-8, it just assummes UTF-8 at all times.
What makes ISO-8559-1 special? ISO-8859-1 only may as well be ASCII only for most of the world.
> Only for characters with the high bit set. This is not a security problem.
Yes, it is. It's a little more elaborate, but you can still get into cases where one program checks the data, say forbiding access to directory e', but misses the access to e' because it's encoded differently.
> programs are forced to examine the text before displaying it
I prefer my programs to get it right, rather than do it quick.
> because there are a huge number of documents out there with these characters
Plain text documents? Not so much. The number of plain text documents is probably much smaller than by the number of KOI8-R plain text documents.
We have a solution - it's called iconv. iconv will handle all sorts of character sets, not just ISO-8859-1 and CP1252. (Actually, you don't completely handle Latin-1 either, since there are valid Latin-1 files that are valid UTF-8. But we can just handwave that away until someone gets bit by it.)
> the standards organizations were stupid for saying these are control characters, if they had assigned them
They did assign them. Single shift 1, single shift 2, etc. If the characters were needed, that's what they should have been used for, Microsoft be damned.
> UTF-8 and ISO-8859-1 can be supported at the same time for all real documents.
And we can forget about the rest of the world that doesn't use ISO-8859-1 or UTF-8? This will hopelessly munge any documents in a different charset; whereas a UTF-8 interpreter could realize it wasn't a UTF-8 document (and probably load in up in a locale charset.)
> This also makes programming easier because there are no "errors" to worry about.
But you've added another case to everything that has to handle UTF-8, and ignored real errors.
>It is also necessary to treat UTF-8 sequences that code a character in more bytes than necessary as illegal, this is vital for security reasons.
Why? You've just introduced alternate sequences for characters (the whole security problem) in another form.
>I am also in favor of adding MicroSoft's assignments to 0x80-0x9F to the standard Unicode.
Huh? All the characters are in Unicode. To mess with preexisting characters (that date back to Unicode 1.0) to add yet more redundant characters isn't going to happen, and shouldn't; you shouldn't break backward compatibilty unless there's at least a benefit.
Illustrator is fairly specific - there's no Corel Illustrator, or OpenIllustrator. There's a StarOffice, OpenOffice, and WordPerfect Office. Microsoft can't claim Office as a trademark, once everyone's used it on their product.
I once sat down and thought about what I was missing in Linux. BeOS had almost all of it.
BeOS has great font support, and excellent Unicode support. It's very fast, with the main browser (NetPositive) being much faster than Netscape. It had a nice GUI and a 64-bit journaling filesystem years before Linux did. BeOS advocates always went on about how you can play 200 videos at once smoothly. It also has fairly decent POSIX support and includes BASH as the default shell.
It was a very nice system, handicapped by a lack of applications, lack of hardware support and the other stuff that comes with being 4th in the OS market.
> this isn't even up for debate anymore
Thank you, Jean Dixon. Of course it's up for debate! 18 months is a long time, and a lot of things could happen in that time. Any future prediction on this level is an educated guess, at best.
According to the FBI, the vast majority of child pornography claims are actually perfectly legal pictures with no sexual activity whatsoever. So if you shut down every site someone claims "child porn", you'll be shutting down a lot of legal sites.
How long do you think it would take to shutdown a child porn site, anyway? If you had evidence, I'd be surprised if it took more than 24 hours to get a search warrant and take the server, except _maybe_ over holidays.
I wasn't trying to imply details of the history of GCC, just that GCC and EGCS had merged. In second thought, I'm not sure that EGCS wasn't a GNU compiler - they had the same requirements, including copyright assignment - as GCC. It just wasn't an _official_ GNU compiler.
> You're somewhat correct in the difficulty in maintaining these, but you overestimate the role of the FSF here. Perhaps you should re-read Ulrich Drepper's rant if you want a better idea of reality.
If I, as a student of OSU, publish a paper that means something, it will get noted as coming from OSU, even if I'm leading riots against the adminstration. If you've accepted the title of GNU maintainer and release packages in a GNU manner, you are acting for the FSF and your actions will get correctly attributed to them.
I think you underestimate the FSF. If it wasn't for the FSF, Linux would probably still be using Linux-specific versions of libc and gcc. EGCS possibly wouldn't have happened, and wouldn't have been nearly as successful, if all the threads of GCC development it tried to follow weren't all based off of GNU's GCC 2.7.2.
As for Ulrich Drepper . . . I've read his posts to the gcc list. He gives commands without explanations, often only saying "you don't understand the problems you're getting into", to people with PH.Ds who have been working on GCC for 5-10 years.
Ulrich Drepper complains about losing total control of glibc, and changing a reference in a GNU program to be conformant to GNU standards. He says "NEVER voluntarily put a project you work on under the GNU umbrella since this means in Stallman's opinion that he has the right to make
decisions for the project.". Duh. GNU is basically a trademark. If you want to use a trademark, you've got to pay the price they want. If you feel that's unreasonable, then you can chose not to use the GNU trademark, including forking glibc, if you want. If all RMS asked was that power no longer be centered in one autocrat, and that glibc uses LGPL 2.1 (and that's all Drepper specifically mentioned), then I don't see RMS as being unreasonable here.
If I recall the copyright assignment, anyone who donates the copyright to code to a GNU project to the FSF has a contract say that the FSF cannot release the code under a non-free license (I don't remember the exact wording, but that's the guist of it.) So there's some guarentee.
> > If I "choose" to work a low-income job,
> You are implying that nobody ever chooses to work a low-income job. This it total nonsense
There exist people who choose to work a low-income job. There also exist people who choose to beaten with pipes. That does not negate the fact that a lot of people don't choose either. I would guess he's talking about them.
Life's not all choice. I was lucky enough to grow up in a (upper?) middle class family with a family who could and would feed my reading and computer using habits. If I had grown up to a poor family in a slum, I possibly wouldn't be at college, and almost certainly wouldn't be typing this message on an expensive computer in my room.
> The gcc used comes from egcs (as does modern gcc), which was forked a long time ago after Stallman installed a bonehead maintainer. It is true GNU software, but most of the code is from non-GNU sources.
EGCS was a short fork that was remerged over a year ago. The GNU project may not have written a lot of the code (any more than the Apache group wrote all of Apache, or Linus wrote all of Linux), but they did the merging and releasing, awesome projects for something as large and complex as GCC. If the FSF had not been there, there would be ten thousand GCCs, with poorly maintained forks for every system out there.
Also, it's a huge project to get any open-source project started. Whatever anyone else may have done to GCC or glibc or gdb, the structure and framework were built by RMS (for GCC and GDB) or by someone working for the FSF (as in getting paid) (for glibc), and that structure was obviously valuable enough to build on.
Replacing the kernel in a Linux system is not a drop-in solution, but changing the kernel to a BSD kernel with Linux emulation, and pretty much only the kernel, could be done in probably under a day - definetly under a week.
How long would it take to replace GCC? Even the BSD's still use GCC! To replace the GNU C compiler for ix86 would probably take years of effort; to replace the GCC (with C, C++, Fortran, Java, and Objective C frontends) for every platform Linux runs on - heck, the GCJ frontend alone took several years by a bunch of full-time programmers. Why do you think so many companies use GCC instead of making thier own compiler? Because it's trivial to duplicate?
Linux is Linux, GNU is a set of mediocre Unix utility ripoffs.
Linux is just a mediocre Unix kernel ripoff, by that measure - there's many "ridiculous" Linux-isms, for example. I've heard many people say the first thing they do on a new Unix (no *) system is install all the GNU utilities, which many people find vastly improved over the Unix utilities. (I've met a couple compiler crashes, but I never seen any sign of instability in the GNU utilities.
Why do you say he dissed goto's? The paper of his I rememeber on the subject, said that they were frequently not the best way to do things, then spent the rest of the paper talking about places where they were the most efficent way to do things.
But a hand-made index is better at finding a subject than searching the text; it doesn't skip synonmys and stop at words in the wrong context. Any decent technical book will hopefully have a handmade index that beats using grep.
I believe that Mozilla includes a built-in demoroniser, by assuming that stuff claiming to be Latin-1 but having characters in the C1 range used by CP 1252, is really CP 1252.
> If you don't know what post-modern rationalism is, do philosophers assume you are dumb?
Some of them will rip you to shreads for not having a basic understanding of philosophy, yes. People in every field don't understand how the general public can be so stupid as not to understand the basics of their subject.
> If you don't know Russian, do Russian professors mock you
Maybe not just Russian, but if you revealed you knew but one language, probably. You've never read people trashing Americans for being monolingugal before?
> Computer nerds are the last bastion of
> unadulterated bigotry
Yeah, whatever. Any discipline are going to have people who think themselves superior for knowing that discipline. English majors don't understand how people can be so poorly read; geography majors can't understand how people don't know where countries are. I complain about many of the customers at Homeland, as do most of my co-workers. Computer nerds are just human, doing what humans do.
American tourists have been arrested in Germany for distributing Nazi propoganda over the internet. In the case I'm thinking of, they actually tried to extradite him first, but the US government refused (but didn't notify him.) I don't remember reading about any governmental objections; appearing to support a Nazi is bad political fodder.
The FSF doesn't give you the right to pull back the copyright. They give you the right to use the code you donated under any license you want, given a month's notice. (They claim the month's notice is merely some sort of legal CYA thing.)
What was so bad about the '99 task? 'cat' was an acceptable solution! Even I was able to toss together a solution, and didn't do half bad. (A couple friends and I are the Icon implementation.)
Now, the '00 task? That was nasty. It looked like you had to have a background in computer graphics to even put together an entry. I couldn't manage to assemble anything for that one.
> Ada83 was not OO in the sense we normally associate with the term -- no inheritance, for example.
... end record; has been in the language since the first. What it did not have was dynamic dispatching.
Ada83 wasn't OO, but Ada95 is. In any case, Ada83 did have inheritance - type foo is new bar with record