Sure, but why should you be forced to install a library if none of the apps you're installing actually use it? I can't speak for the original poster, but that's certainly why I don't have Qt installed (although I have had in the past, and may again in the future). It does seem a bit odd to have SuSE's Gnome depend on Qt, if that's what's actually going on.
I beg to differ. Including reasonable advice with a troll does not mean it's not a troll.
I suggested he buy an MP3 player that supports Ogg Vorbis
You told him not to buy an iPod (sound advice, which I certainly will follow) and said, "the rest of the world uses MP3" (a bit of an exaggeration, but not over the top). But then you told him he "shouldn't have picked some smalltime format." If it's so smalltime, why do so many vendors already support it? Then you called Vorbis a "little 'eleet' fad format" that "Linux geeks whine about this week", and went on to your most serious troll remark: "Next week, it'll be 'GNU KewlAudio' or something." Does any of this ring a bell? The only thing you left out was "Netcraft confirms it, Vorbis is dead".:)
But the whole point here is that its still unpopular enough that the iPod doesn't support it.
If that was your point, why didn't you say that instead of spewing crap about Linux geeks and "'eleet' fad" and "KewlAudio"? Then, guess what, nobody would have accused you of trolling!
Excellent troll! (Who modded this crap as "insightful"?) Yeah, the fact that the Free Software community has been rallying behind Ogg Vorbis for the last several years definitely suggests that they'll suddenly switch to supporting something else next week. Very logical, very well reasoned.
I don't know if you've noticed, but Vorbis is popular enough that it actually is supported by some consumer-grade hardware vendors. That's hardly a "flavor of the week". That, in fact, puts it in an exclusive category with only three other formats that I know of. It may be the most obscure of the three, but this is 2004, not 1998, and Ogg is not that obscure anymore. Wake up and smell the coffee.
(Now me, I convert my.shn's to flac, but that's another story....:)
MS has shown no reluctance in the past to retaliate against vendors that refuse to toe the party line ("all Windows all the time everywhere") by changing pricing contracts. Remember, MS's OS customers are mostly OEMs like IBM, not end users. The consent decree forbad some of this behavior (which is partly, no doubt, why IBM was finally willing to publically back Linux), but MS has shown little reluctance to play fast and loose with the terms of their consent decrees, and if I were IBM, I would be treading very carefully. IBM sells a lot of systems for MS, and a small increase in what they pay for those MS systems could have a big impact on their bottom line.
IBM is not going to stop selling thier customers MS Windows hardware or services that want them.
Exactly. That's why they don't want to piss MS off any more than they absolutely have to. There's already little love lost between the two companies. Public bragging about their plans for their internal systems could well be equivalent to pouring kerosene on a blaze. If they keep it quiet, then it's purely an internal matter, and MS will probably ignore it. If they try to turn it into a publicity stunt, though, then MS will probably feel compelled to respond somehow.
He didn't say money wasn't a factor, he said money wasn't everything. So your question should really be phrased, "what else is it about?" And in this case, one of the things it seems to be about is Hebrew language support. There may be (probably are) other issues as well. (Like trust, like promoting a local software industry, like not getting locked into one-sided long-term contracts, like control of your own destiny, like freedom, etc.)
> Microsoft, you need to make cheaper software
Oh, that definitely clears things up!
Yes, if you chop out the part where he says, "You also need..." and then lists stuff, then maybe you can (pretend to?) completely misunderstand what he had to say. With journalistic skills like that, I bet you could get a job as a slashdot editor.:)
I, on the other hand, have about 5% GTK/Gnome (probably the same apps), and exactly 0% KDE. And I'm just hoping this thing doesn't mean that I'm going to have to end up installing both! It's bad enough that I have dozens of gnomish libraries that I almost never use. It'll be even worse if I have to end up dragging in all the KDE libs too, just to support the same handful of apps.
Commercial vendors have already overwhelmingly opted to use Qt instead of GTK.
Commercial vendors have overwhelmingly opted to use Windows, not Qt or GTK.:p;)
Qt is a much nicer toolkit.
Highly subjective. If you're working in Lisp or Python or Eiffel or Ada or Scheme or Smalltalk or Ruby or Haskell, there's not a lot of difference either way. If you're working in C, you're screwed no matter what you do. And if you're working in C++ -- you have my sympathy.:)
Your other three points are reasonably valid, however, so I'm not gonna give you too much flack. But none of them is strong enough to make me want to install any KDE libs at present.
They (claim to) have 6000 source licensees, companies that have paid to use their source code, mostly companies like IBM and Sun that make their own flavors of UNIX(tm)[1]. They have many more (possibly the millions you remember) users of Unixware and their other systems who don't have (and certainly don't want at this point) any source license. So there is no contradiction (although that doesn't necessarily mean that they're telling the truth about anything).
[1] UNIX is a trademark of the Open Group, not of SCO/Caldera/Novell or anyone else. Despite SCOldera's claims to "own UNIX."
the one thing that the FSF said made the original BSD license non-free was...
They have never said that the original BSD license was non-free. If you look here, you'll see that it is listed in the section titled, "The following licenses are free software licenses, but are not compatible with the GNU GPL."
The advertising clause of the original BSD license was just a bad idea. I think ultimately the Regents were persuaded to drop it, not because of pressure from the FSF and its supporters, but because of pressure from businesses (especially small businesses and consultants) who found the advertising clause unduly onerous. Giving credit where credit is due is one thing. Being required to list 50 names in a small classified ad is another.
Finally, the GPL does require that credit be given. Just not in advertising material. "You may copy and distribute...provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice..."
OS/2 may have been a failure in the home/desktop market, but it was a pretty big success in the business/embedded market. It's use in bank ATMs alone may well qualify it as the 2nd most successful OS to date.
How about Taligent?
Better, although it might be disqualified on a technicality: does something have to exist before you can really call it a flop?:)
What about the Disney Sound Doohicky
I dunno, never heard of it. Are you sure it isn't just ordinary crap? To be a flop, there has to be an expectation of success, and to be a huge flop, there has to be an expectation of huge success. So things can be amazingly crappy without ever being a flop. In fact, when it comes to high-tech, crap is almost the rule, rather than the exception. And everyone knows this, which is why expectations are usually low, which in turn is why huge flops are kinda rare, despite all the utter crap that's out there.:)
When those books are first piled on the desk, the very first thing that happens is 3/4 of them are tossed in trash without ever getting past the cover page.
The cover page can reveal a lot. Like, whether you have the faintest clue about the publisher's guidelines, whether you can compose anything resembling a coherent sentence, whether you've had the common sense to talk to an agent, etc. A lot of publishers don't even accept unsolicited manuscripts. So, if you send your "precious first novel" to one of these, you have no one but yourself to blame for the result. Investigate the publishers before trying to get published. Or better yet, try to get an agent, since they know what various publishers want -- it's their job to know. (Though most agents will want to see a portfolio, so write a few things before attempting this step).
Being published and having written a good book are niether mutually exclusive nor mutually inclusive and really enjoy little corelation to one another.
I agree that it's neither exclusive nor inclusive. I disagree that there's little correlation. I think there's a strong correlation: 90% of all writers are unpublished; 90% of all writers are bad. Give or take (those numbers are probably underestimating the situation). I've never read slush-piles myself, but because I know a lot of writers (and a few other people in the industry), I've been inflicted with a fair number of unpublished manuscripts by desperate wanna-be writers. And I have yet to see one that was even halfway decent.
There's a myth around that writing is easy -- it's not, not good writing. The dream of taking a year off and writing "the Great American Novel" is crap. Other artists take the time to learn their craft, and know that their early works are unlikely to be of interest to anyone. Only wanna-be writers (and singers who queue up for American Idol) assume that their work is the sin qa non from day one. (And the singers are more likely to be right -- singing is much easier than writing.) Too many clueless wanna-be's assume that writing one single novel is enough, and their job is done at that point.
Writing is a complex skill. There's much more involved than just the ability to form coherent sentences. There's plotting, pacing, character development, seamless exposition that doesn't pull the reader out of the story, and so on, and so on. Natural talent at all those elements is so rare as to be almost non-existent. So even someone who has an innate skill at writing (and there are such people) is unlikely to produce a good work on the very first try.
Bottom line: "how do I get my first novel published, should I self-publish?" is almost always a sign of someone whose work is, frankly, unpublishable. And, given that attitude, may always be. The question about your first novel should really be: "how can I find people who are willing to suffer through my rookie mistakes and offer constructive criticism so I can learn how to be a better writer and maybe get good enough to have a chance of being published someday?"
Yes, and when it says, "the core team was no longer representative of the active, experienced and skilled XFree86 developers," it actually sounds like they might be opening up the project a little more, rather than disbanding it. Given some of the negative comments I've heard in the past about the rigidity and bureaucracy of core team, this could well be a very good thing for XFree86 overall.
I've seen this several times in comments so far, the idea that publishers are magical and if they don't accept a book it must suck.
While it's not necessarily true, the fact is that publishers (yes, all of them) have huge piles of unsolicited manuscripts, and yes, most of them DO suck! An occasional good work may get lost in this pile. But they (or at least most) do actually have people to read these "slush piles" (the most feared job in the industry), and they occasionally find good things in there. But the plain fact of the matter is that far more people think they can write than actually can.
Do you realize that Stephen King couldn't get published for YEARS?
This is true of most authors. And, having read a couple of early King works, published later after he was famous (and probably after heavy editing), my opinion is that those early attempts were rejected for damn good reason. They didn't deserve to be published!
If you're serious about writing, the main trick is to keep writing! NOT to keep trying to get your first attempt published. If it gets rejected several times, stick it in a drawer, and come back to it after a while and see if you can figure out how or if it can be salvaged. In the mean time, write more, and keep trying. This is the path that the vast majority of successful (and semi-successful) authors have taken.
Stephen King and Piers Anthony didn't keep trying to get those early works published. They kept writing! And that's the secret to their success.
Also, don't expect to get rich or famous. Most authors, even those with dozens of published books, barely make a living. So, if you don't enjoy the process of writing, try another line of work. And if you do, keep writing; eventually you'll probably learn enough to write something that sells.
And don't forget the short-story market. It's a lot easier to write a decent short story than to put together an entire novel. Plus it takes less work, and is quicker.
(No, I'm not a writer, but I grew up around writers, and have been good friends with dozens, and acquainted with many more.)
Minor quibble: I don't think memtest86 is going to do a whole lot of good on the systems that were being tested in this particular case, since the systems that were being tested were PPC, which doesn't run x86 software very well. But yes, it's a nice program to keep around for your x86 hardware. I got my copy from the BBC bootable linux business card CD.
Re:source code escrow not very useful
on
Source Code Escrow
·
· Score: 1
If the developer goes out of business, getting the source code by itself is almost always useless
No. If the developer goes out of business, you're screwed. If you have the code in escrow, you may be less screwed. Not necessarily, but possibly.
almost no single customer will have the resources to maintain and extend it.
Escrow is usually expensive, and is only done by companies that can afford it. Companies that are big enough to be able to do something with the code. True, in most cases, the company is not going to want to take over long-term maintenance and will certainly not want to bother extending it. But having the code (and hiring consultants) may allow them to fix urgent problems while they work on a transition plan to switch to something supported. And may make it easier/cheaper to extract precious data from proprietary formats, so it can be uploaded into the new, supported systems. These two factors alone could potentially save a company millions.
Source code is only cost effective if...
You seem to be saying that code escrow isn't as good as open source (from a user's perspective). That's obviously true, but that doesn't mean that escrow is useless. Escrow is a viable compromise when the vendor isn't willing to open their code, and you're not willing to trust their long-term viability, and can afford to hedge your bets. It's not a perfect solution, but it certainly gives you options you wouldn't have otherwise, and could potentially save a lot of money. It's a trade-off, and should be evaluated as such.
Saying it's not worth considering because it's not a perfect solution is silly. Few things in life are perfect. Escrow is an option with advantages and disadvantages, and should be evaluated as such. But the fact that it's been a common practice for decades should say something.
(The slashdot article does kind of give the impression that this is something new, and if you were under that impression, then I can understand your reaction. But in fact, code escrow is nothing new, and there is no news here, and the article should never have been posted in the first place. As "news", it's about as exciting as hearing that Indian developers are using language compilers to avoid the need to program directly in machine language.)
RIght, and this is why people who keep asking "is SCO going to stop shipping Samba, etc." are missing the point. SCO is arguing that the limitations of the GPL are unconstitutional. They have never denied that people have the right to place works in the public domain (PD). They simply think it has to be all or nothing, and if you try to go "half-way" towards PD (as they seem to view it), then legally you have indicated your intention to go all the way, and your work should be treated as PD.
Of course, companies that rely on the GPL to protect their own software, and who actually earn license fees for non-GPL-compatible use (like MySQL and Trolltech) will fight this interpretation tooth and nail.
Apple II, Atari 800, C-64 drives
on
First Computers
·
· Score: 1
The Commodore 1541 floppy drive was painfully slow as well.
I developed for all three machines (Apple II, Atari 800, C-64) back in the day, and, while I don't remember the exact details, I do remember that our tests showed that the Apple II's drives were ten times faster than the Atari's, and the Atari's were ten times faster than the C-64's. Which made the C-64's drives a whopping one hundred times slower than the Apple's.
Of course, the Apple drives also used up all the CPU while they were running, so it was a good thing they were so fast.:)
Most tools for measuring LOC are smart enough that this doesn't really matter.
Strip the comments.
Most tools for measuring LOC ignore comments.
Format conditionals, blocks, and function calls consistently
Most tools for measuring LOC don't care how they're formatted.
Running "wc -l" does not measure LOC, it measures lines of text (LOT?). If they claim to have measured LOC, then either they used a decent tool (probably, such things are readily available[1]), or they're incompetent fools (possible, but somewhat less likely IMO). Don't assume the latter without evidence.
[1] Debian includes a package called "sloccount", for example.
Problem with that is that it assumes the same "code density".
Strictly speaking, yes. Of course, while that might mean that MySQL was rated too highly (if it had a lower code density), it might also mean that MySQL's quality was underrated! (If it had a higher code density). Of course, since MySQL is free/libre/open sores, it can be examined to determine whether the former or latter is more probable.
I will point out, though, that the findings are reasonably consistent with a more emperical test, the University of Wisconsin's fuzz tests, which stress-tested standard Unix utilities on several systems, including GNU, and found that the GNU code was much more solid and robust overall.
I'm just saying that I wouldn't take any statistic that is derived using "lines of code" as a variable as a serious, hard number.
Absolutely not. On the other hand, it's somewhat suggestive. And a error rate six times higher is hard to explain purely as a matter of code density. Or much of anything else, for that matter. And it's consistent with other studies. But I agree that anything which mentions "lines of code" should be taken with a huge grain of salt.
He may not be an official member of the GNOME Foundation Board, but that hardly means that he's no longer a leader of the GNOME project itself. Let's keep some perspective here. I seriously doubt if anyone is saying, "he didn't meet the election deadline, we'd better shut off his CVS access." Or even, "we'd better stop listening to what he has to say."
It's even possible that not being on the Foundation could allow Miguel to spend more time actually working on GNOME.
And one of the advantages of that (unlike some gcc extensions) is that you can fairly easily make the code still portable:
#ifndef __GCC__ #define __attribute__(x) #endif
Of course, this way you'll still get the warnings on non-gcc compilers, but so what? Nothing's perfect.
GCC is littered with HUNDREDS of very cool extensions. Just make sure it's worth giving up portability...
Even if you do use non-portable gcc extensions, you're not giving up that much portability. Unless you're still working with 16-bit (or smaller) platforms, you're going to have a hard time finding systems that don't support gcc!
Compiler-independence is much less important when your target compiler is nearly as portable as the language it provides.:)
(Note: I generally avoid gcc extensions myself, unless there's a good reason to use them -- I'm just saying, it's not necessarily a big deal if you do use 'em.)
Owns stock in IBM is a more likely theory (IMO). Now with stock in both companies, they're covered no matter which way the case goes. (I can't claim this theory--I saw it at Groklaw--but it seems to make sense.)
Slashdot Mirror
YaST, in particular, uses Qt.
Yup, that makes sense. I thought it might be something like that. (I don't actually run SuSE -- I'm just curious about Novell's general activities.)
Thanks for the clarification.
QT is open source and a good library.
Sure, but why should you be forced to install a library if none of the apps you're installing actually use it? I can't speak for the original poster, but that's certainly why I don't have Qt installed (although I have had in the past, and may again in the future). It does seem a bit odd to have SuSE's Gnome depend on Qt, if that's what's actually going on.
It's not a troll.
:)
I beg to differ. Including reasonable advice with a troll does not mean it's not a troll.
I suggested he buy an MP3 player that supports Ogg Vorbis
You told him not to buy an iPod (sound advice, which I certainly will follow) and said, "the rest of the world uses MP3" (a bit of an exaggeration, but not over the top). But then you told him he "shouldn't have picked some smalltime format." If it's so smalltime, why do so many vendors already support it? Then you called Vorbis a "little 'eleet' fad format" that "Linux geeks whine about this week", and went on to your most serious troll remark: "Next week, it'll be 'GNU KewlAudio' or something." Does any of this ring a bell? The only thing you left out was "Netcraft confirms it, Vorbis is dead".
But the whole point here is that its still unpopular enough that the iPod doesn't support it.
If that was your point, why didn't you say that instead of spewing crap about Linux geeks and "'eleet' fad" and "KewlAudio"? Then, guess what, nobody would have accused you of trolling!
Next week, it'll be...
.shn's to flac, but that's another story....:)
Excellent troll! (Who modded this crap as "insightful"?) Yeah, the fact that the Free Software community has been rallying behind Ogg Vorbis for the last several years definitely suggests that they'll suddenly switch to supporting something else next week. Very logical, very well reasoned.
I don't know if you've noticed, but Vorbis is popular enough that it actually is supported by some consumer-grade hardware vendors. That's hardly a "flavor of the week". That, in fact, puts it in an exclusive category with only three other formats that I know of. It may be the most obscure of the three, but this is 2004, not 1998, and Ogg is not that obscure anymore. Wake up and smell the coffee.
(Now me, I convert my
MS has shown no reluctance in the past to retaliate against vendors that refuse to toe the party line ("all Windows all the time everywhere") by changing pricing contracts. Remember, MS's OS customers are mostly OEMs like IBM, not end users. The consent decree forbad some of this behavior (which is partly, no doubt, why IBM was finally willing to publically back Linux), but MS has shown little reluctance to play fast and loose with the terms of their consent decrees, and if I were IBM, I would be treading very carefully. IBM sells a lot of systems for MS, and a small increase in what they pay for those MS systems could have a big impact on their bottom line.
IBM is not going to stop selling thier customers MS Windows hardware or services that want them.
Exactly. That's why they don't want to piss MS off any more than they absolutely have to. There's already little love lost between the two companies. Public bragging about their plans for their internal systems could well be equivalent to pouring kerosene on a blaze. If they keep it quiet, then it's purely an internal matter, and MS will probably ignore it. If they try to turn it into a publicity stunt, though, then MS will probably feel compelled to respond somehow.
It's 3-4% of the total money that Microsoft's Israeli branch ("MS Israel") takes in, not 3-4% of what the gov't of Israel spends.
:)
> If the number is correct it must exclude the Military and the health sector.
I don't think MS Israel has a Military or health sector.
> Money is not everything
:)
Then what is it about?
He didn't say money wasn't a factor, he said money wasn't everything. So your question should really be phrased, "what else is it about?" And in this case, one of the things it seems to be about is Hebrew language support. There may be (probably are) other issues as well. (Like trust, like promoting a local software industry, like not getting locked into one-sided long-term contracts, like control of your own destiny, like freedom, etc.)
> Microsoft, you need to make cheaper software
Oh, that definitely clears things up!
Yes, if you chop out the part where he says, "You also need..." and then lists stuff, then maybe you can (pretend to?) completely misunderstand what he had to say. With journalistic skills like that, I bet you could get a job as a slashdot editor.
I, on the other hand, have about 5% GTK/Gnome (probably the same apps), and exactly 0% KDE. And I'm just hoping this thing doesn't mean that I'm going to have to end up installing both! It's bad enough that I have dozens of gnomish libraries that I almost never use. It'll be even worse if I have to end up dragging in all the KDE libs too, just to support the same handful of apps.
:p ;)
:)
Commercial vendors have already overwhelmingly opted to use Qt instead of GTK.
Commercial vendors have overwhelmingly opted to use Windows, not Qt or GTK.
Qt is a much nicer toolkit.
Highly subjective. If you're working in Lisp or Python or Eiffel or Ada or Scheme or Smalltalk or Ruby or Haskell, there's not a lot of difference either way. If you're working in C, you're screwed no matter what you do. And if you're working in C++ -- you have my sympathy.
Your other three points are reasonably valid, however, so I'm not gonna give you too much flack. But none of them is strong enough to make me want to install any KDE libs at present.
They (claim to) have 6000 source licensees, companies that have paid to use their source code, mostly companies like IBM and Sun that make their own flavors of UNIX(tm)[1]. They have many more (possibly the millions you remember) users of Unixware and their other systems who don't have (and certainly don't want at this point) any source license. So there is no contradiction (although that doesn't necessarily mean that they're telling the truth about anything).
[1] UNIX is a trademark of the Open Group, not of SCO/Caldera/Novell or anyone else. Despite SCOldera's claims to "own UNIX."
the one thing that the FSF said made the original BSD license non-free was...
They have never said that the original BSD license was non-free. If you look here, you'll see that it is listed in the section titled, "The following licenses are free software licenses, but are not compatible with the GNU GPL."
The advertising clause of the original BSD license was just a bad idea. I think ultimately the Regents were persuaded to drop it, not because of pressure from the FSF and its supporters, but because of pressure from businesses (especially small businesses and consultants) who found the advertising clause unduly onerous. Giving credit where credit is due is one thing. Being required to list 50 names in a small classified ad is another.
Finally, the GPL does require that credit be given. Just not in advertising material. "You may copy and distribute...provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice..."
Shit, OS/2 ain't even on the list.
:)
:)
OS/2 may have been a failure in the home/desktop market, but it was a pretty big success in the business/embedded market. It's use in bank ATMs alone may well qualify it as the 2nd most successful OS to date.
How about Taligent?
Better, although it might be disqualified on a technicality: does something have to exist before you can really call it a flop?
What about the Disney Sound Doohicky
I dunno, never heard of it. Are you sure it isn't just ordinary crap? To be a flop, there has to be an expectation of success, and to be a huge flop, there has to be an expectation of huge success. So things can be amazingly crappy without ever being a flop. In fact, when it comes to high-tech, crap is almost the rule, rather than the exception. And everyone knows this, which is why expectations are usually low, which in turn is why huge flops are kinda rare, despite all the utter crap that's out there.
When those books are first piled on the desk, the very first thing that happens is 3/4 of them are tossed in trash without ever getting past the cover page.
The cover page can reveal a lot. Like, whether you have the faintest clue about the publisher's guidelines, whether you can compose anything resembling a coherent sentence, whether you've had the common sense to talk to an agent, etc. A lot of publishers don't even accept unsolicited manuscripts. So, if you send your "precious first novel" to one of these, you have no one but yourself to blame for the result. Investigate the publishers before trying to get published. Or better yet, try to get an agent, since they know what various publishers want -- it's their job to know. (Though most agents will want to see a portfolio, so write a few things before attempting this step).
Being published and having written a good book are niether mutually exclusive nor mutually inclusive and really enjoy little corelation to one another.
I agree that it's neither exclusive nor inclusive. I disagree that there's little correlation. I think there's a strong correlation: 90% of all writers are unpublished; 90% of all writers are bad. Give or take (those numbers are probably underestimating the situation). I've never read slush-piles myself, but because I know a lot of writers (and a few other people in the industry), I've been inflicted with a fair number of unpublished manuscripts by desperate wanna-be writers. And I have yet to see one that was even halfway decent.
There's a myth around that writing is easy -- it's not, not good writing. The dream of taking a year off and writing "the Great American Novel" is crap. Other artists take the time to learn their craft, and know that their early works are unlikely to be of interest to anyone. Only wanna-be writers (and singers who queue up for American Idol) assume that their work is the sin qa non from day one. (And the singers are more likely to be right -- singing is much easier than writing.) Too many clueless wanna-be's assume that writing one single novel is enough, and their job is done at that point.
Writing is a complex skill. There's much more involved than just the ability to form coherent sentences. There's plotting, pacing, character development, seamless exposition that doesn't pull the reader out of the story, and so on, and so on. Natural talent at all those elements is so rare as to be almost non-existent. So even someone who has an innate skill at writing (and there are such people) is unlikely to produce a good work on the very first try.
Bottom line: "how do I get my first novel published, should I self-publish?" is almost always a sign of someone whose work is, frankly, unpublishable. And, given that attitude, may always be. The question about your first novel should really be: "how can I find people who are willing to suffer through my rookie mistakes and offer constructive criticism so I can learn how to be a better writer and maybe get good enough to have a chance of being published someday?"
Yes, and when it says, "the core team was no longer representative of the active, experienced and skilled XFree86 developers," it actually sounds like they might be opening up the project a little more, rather than disbanding it. Given some of the negative comments I've heard in the past about the rigidity and bureaucracy of core team, this could well be a very good thing for XFree86 overall.
I've seen this several times in comments so far, the idea that publishers are magical and if they don't accept a book it must suck.
While it's not necessarily true, the fact is that publishers (yes, all of them) have huge piles of unsolicited manuscripts, and yes, most of them DO suck! An occasional good work may get lost in this pile. But they (or at least most) do actually have people to read these "slush piles" (the most feared job in the industry), and they occasionally find good things in there. But the plain fact of the matter is that far more people think they can write than actually can.
Do you realize that Stephen King couldn't get published for YEARS?
This is true of most authors. And, having read a couple of early King works, published later after he was famous (and probably after heavy editing), my opinion is that those early attempts were rejected for damn good reason. They didn't deserve to be published!
If you're serious about writing, the main trick is to keep writing! NOT to keep trying to get your first attempt published. If it gets rejected several times, stick it in a drawer, and come back to it after a while and see if you can figure out how or if it can be salvaged. In the mean time, write more, and keep trying. This is the path that the vast majority of successful (and semi-successful) authors have taken.
Stephen King and Piers Anthony didn't keep trying to get those early works published. They kept writing! And that's the secret to their success.
Also, don't expect to get rich or famous. Most authors, even those with dozens of published books, barely make a living. So, if you don't enjoy the process of writing, try another line of work. And if you do, keep writing; eventually you'll probably learn enough to write something that sells.
And don't forget the short-story market. It's a lot easier to write a decent short story than to put together an entire novel. Plus it takes less work, and is quicker.
(No, I'm not a writer, but I grew up around writers, and have been good friends with dozens, and acquainted with many more.)
it would be even more valuable if the same tests were performed on a variety of operating systems
How many operating systems run on IBM's pSeries machines? AIX and...?
Minor quibble: I don't think memtest86 is going to do a whole lot of good on the systems that were being tested in this particular case, since the systems that were being tested were PPC, which doesn't run x86 software very well. But yes, it's a nice program to keep around for your x86 hardware. I got my copy from the BBC bootable linux business card CD.
If the developer goes out of business, getting the source code by itself is almost always useless
No. If the developer goes out of business, you're screwed. If you have the code in escrow, you may be less screwed. Not necessarily, but possibly.
almost no single customer will have the resources to maintain and extend it.
Escrow is usually expensive, and is only done by companies that can afford it. Companies that are big enough to be able to do something with the code. True, in most cases, the company is not going to want to take over long-term maintenance and will certainly not want to bother extending it. But having the code (and hiring consultants) may allow them to fix urgent problems while they work on a transition plan to switch to something supported. And may make it easier/cheaper to extract precious data from proprietary formats, so it can be uploaded into the new, supported systems. These two factors alone could potentially save a company millions.
Source code is only cost effective if...
You seem to be saying that code escrow isn't as good as open source (from a user's perspective). That's obviously true, but that doesn't mean that escrow is useless. Escrow is a viable compromise when the vendor isn't willing to open their code, and you're not willing to trust their long-term viability, and can afford to hedge your bets. It's not a perfect solution, but it certainly gives you options you wouldn't have otherwise, and could potentially save a lot of money. It's a trade-off, and should be evaluated as such.
Saying it's not worth considering because it's not a perfect solution is silly. Few things in life are perfect. Escrow is an option with advantages and disadvantages, and should be evaluated as such. But the fact that it's been a common practice for decades should say something.
(The slashdot article does kind of give the impression that this is something new, and if you were under that impression, then I can understand your reaction. But in fact, code escrow is nothing new, and there is no news here, and the article should never have been posted in the first place. As "news", it's about as exciting as hearing that Indian developers are using language compilers to avoid the need to program directly in machine language.)
RIght, and this is why people who keep asking "is SCO going to stop shipping Samba, etc." are missing the point. SCO is arguing that the limitations of the GPL are unconstitutional. They have never denied that people have the right to place works in the public domain (PD). They simply think it has to be all or nothing, and if you try to go "half-way" towards PD (as they seem to view it), then legally you have indicated your intention to go all the way, and your work should be treated as PD.
Of course, companies that rely on the GPL to protect their own software, and who actually earn license fees for non-GPL-compatible use (like MySQL and Trolltech) will fight this interpretation tooth and nail.
The Commodore 1541 floppy drive was painfully slow as well.
:)
I developed for all three machines (Apple II, Atari 800, C-64) back in the day, and, while I don't remember the exact details, I do remember that our tests showed that the Apple II's drives were ten times faster than the Atari's, and the Atari's were ten times faster than the C-64's. Which made the C-64's drives a whopping one hundred times slower than the Apple's.
Of course, the Apple drives also used up all the CPU while they were running, so it was a good thing they were so fast.
Formats the code consistently.
Most tools for measuring LOC are smart enough that this doesn't really matter.
Strip the comments.
Most tools for measuring LOC ignore comments.
Format conditionals, blocks, and function calls consistently
Most tools for measuring LOC don't care how they're formatted.
Running "wc -l" does not measure LOC, it measures lines of text (LOT?). If they claim to have measured LOC, then either they used a decent tool (probably, such things are readily available[1]), or they're incompetent fools (possible, but somewhat less likely IMO). Don't assume the latter without evidence.
[1] Debian includes a package called "sloccount", for example.
Problem with that is that it assumes the same "code density".
Strictly speaking, yes. Of course, while that might mean that MySQL was rated too highly (if it had a lower code density), it might also mean that MySQL's quality was underrated! (If it had a higher code density). Of course, since MySQL is free/libre/open sores, it can be examined to determine whether the former or latter is more probable.
I will point out, though, that the findings are reasonably consistent with a more emperical test, the University of Wisconsin's fuzz tests, which stress-tested standard Unix utilities on several systems, including GNU, and found that the GNU code was much more solid and robust overall.
I'm just saying that I wouldn't take any statistic that is derived using "lines of code" as a variable as a serious, hard number.
Absolutely not. On the other hand, it's somewhat suggestive. And a error rate six times higher is hard to explain purely as a matter of code density. Or much of anything else, for that matter. And it's consistent with other studies. But I agree that anything which mentions "lines of code" should be taken with a huge grain of salt.
Be careful how you speak of Linus' daughter's godfather.
He may not be an official member of the GNOME Foundation Board, but that hardly means that he's no longer a leader of the GNOME project itself. Let's keep some perspective here. I seriously doubt if anyone is saying, "he didn't meet the election deadline, we'd better shut off his CVS access." Or even, "we'd better stop listening to what he has to say."
It's even possible that not being on the Foundation could allow Miguel to spend more time actually working on GNOME.
And one of the advantages of that (unlike some gcc extensions) is that you can fairly easily make the code still portable:Of course, this way you'll still get the warnings on non-gcc compilers, but so what? Nothing's perfect.
GCC is littered with HUNDREDS of very cool extensions. Just make sure it's worth giving up portability...
Even if you do use non-portable gcc extensions, you're not giving up that much portability. Unless you're still working with 16-bit (or smaller) platforms, you're going to have a hard time finding systems that don't support gcc!
Compiler-independence is much less important when your target compiler is nearly as portable as the language it provides.
(Note: I generally avoid gcc extensions myself, unless there's a good reason to use them -- I'm just saying, it's not necessarily a big deal if you do use 'em.)
So RBC does business with Red Hat...?
Owns stock in IBM is a more likely theory (IMO). Now with stock in both companies, they're covered no matter which way the case goes. (I can't claim this theory--I saw it at Groklaw--but it seems to make sense.)