I'd REALLY REALLY REALLY like to see Boeing, Airbus et al. installing avionics and comms systems that can't be disrupted by ubiquitous and nearly free techno-gadgets.
I'll second that.
Having read the article, there are some interesting points. First off they aren't alleging that using your cellphone will make the plane crash, but rather that it might cause some sort of distracting noise in the crews headsets and at worst could conceivably cause a false alarm on one of their warning lights. Sounds a little iffy to me, but ok, better safe than sorry on a plane... then I read on.
Turns out it doesn't matter on new jets - only ones certified pre-'89. So why don't they let people use their mobiles on the newer planes where it's not an issue? Back to the old 'conspiracy theories' on that one. Controllers like control. And the high prices on the sky phones can't hurt either.
Plus, as you kind of hinted at, if a cellphone can really cause even minor systems disruption on a pre-'89 jet, just imagine what someone that was seriously trying to cause a problem could do. It's absurd. If those jets really do have systems that can be so easily disrupted, they should be grounded until they're fixed. So either way, something doesn't add up here, either they're lying (or maybe just stretching the truth very far and very consciously) or they're not even trying to do their job, take your pick.
In an age when we know there are people trying to bring jetliners down, it's absolutely absurd to be flying jetliners that are so poorly insulated against EM interference that a mobile phone is a threat to them. Period.
Hrmm looks like a step in the right direction, know when they put that up?
They have had a really rotten attitude about this for years, refusing to take any responsibility, and even sending out massive spam openly themselves at least once.
As I said, I don't want to rehash the same old arguments here so I'm simply going to ignore most of what you said, other than to state that it's a matter of personal preference.
But one of your comments is way off in a different direction, and I'll address that one.
[Slack]
doesn't try to keep track of dependencies - which after fighting with Redhat for awhile seems like a really good choice
Which statement only makes it obvious that you've never used Debian. As someone who went from SLS to Slackware to RH to Debian, perhaps I can explain. Dependencies are:
While I appreciate your humour (and a kernel of seriousness underneath) I explicitly chose to say that RH in particular made the choice of slack in this respect look good. I'm aware that Debian handles the dependency problems much better, that's why I phrased it like I did.
BTW, taste issues aside, your characterisation of slack and dependencies is quite inaccurate for 99.9% of cases at least. You can generally download anything from reputable sources and just have it run, if there's some particular dependency that will be noted. I have had the experience occasionally of installing something that did have an unsatisfied dependency, it's not difficult at all to figure out what they are. You start the program, it says fatal error could not find blah_blah.so you search for blah_blah.so and there you go. Yes, in comparison to debian magically going off and grabbing it for you that might be seen as a slight inconvenience, but it's no big deal. The nice thing is that the system doesn't care if blah_blah.so is there from a package or if you compiled it or how it got there, it just works. What does Debian do? I know Redhat is completely unable to cope in any sane way. RH fans will doubtless jump on me and point out --nodeps and --force, but then when you complain about it shitting it's own database they say it's probably because you used --nodeps and --force and you should never do that. I think the truth is more like you should never compile from source on RH, which sort of ruins the point of running a Free system in my mind. Or you should compile and jump through all the hoops to make an rpm and then install that instead, what a mess that is. On slack you can do the same thing, a lot easier, if you want, or you can just use make install and make uninstall and the package manager will play nice either way. How does Debian cope with source installs? I have used it a little but never did that on a debian system...
Oh, and one more btw, cat isn't an editor, but ed sure is.;)
They have a lot of nerve to complain about Spam. Yeah, I know they claim to have nothing to do with the spammers selling their products, but if they don't why do they refuse to do anything about it, to the point of refusing complaints? Other companies in that spot take action, Symantec ignores it at best.
Oh yeah, also, 'inappropriate Spam?' Is there any other type? Jeez.
I pretty much agree, but let me point out, contrary to popular belief, Slackware does have a binary packages system, has for ages, and it works real well. Those tarballs have install scripts, you use the package manager and it runs them and keeps a database so they can be cleanly uninstalled. The main difference between Slack and Debian/RH in terms of package management is that it doesn't try to keep track of dependencies - which after fighting with Redhat for awhile seems like a really good choice. I know the argument over whether it's better to do it one way or the other can go on forever, and I'd rather not rehash it yet again, but the point is it's simply incorrect to claim Slack doesn't have package management. It does, it was the first with it I believe, it's just that it follows the slack philosophy of simple elegance and doesn't try to do everything.
I've gotta wonder what you think the alternative here is.
* Do you want to recompile each package every time you want to update it, or do you want to do rpm -i?
Like I said, what are you comparing to? Debian has a package management system that beats RPM to hell. Slackware uses a simple package manager that works like a champ, and most other distros use a system based on Debian or Redhat... so what are you comparing to?
* Do you want to have to watch every mailing list for possible security problems on your software, or do you just want to look in the errata section of the RedHat web site?
Every distro has their own equivelant, many are easier to keep track of, and I can't think of one that has had as many problems to post. So what was your point again?
While I agree with this sentiment mostly, you have to realize that to the people that are paid to make these decisions (The Boss, the CIO, whatever) customizable == bad, at least as far as operating system decisions are concerned.
Then you might want to not emphasise that all flavours of free *nix are 100% customisable. But seriously, if you're looking for stability, just switch to Debian or Slack. They're both incredibly stable, don't require a hell of a lot of customisation, and will do whatever you want with the minimum of fuss. I understand that goes for FreeBSD too, but I haven't messed with it enough to say personally. But there's certainly no way I'd willingly suffer through Redhats way of doing things when there are so many perfectly usable alternatives.
If you're writing proper web-pages they'll be usable without image-loading at all. So IE users will still be able to use the page, it just won't look as pretty. Which gives you a great opportunity to educate them. There are a couple of solutions - they can get a different browser, they can get QT (most will probably already have it) and set it as the plugin to display pngs in IE (I've been told this works but haven't tested it since I don't use IE except on mac, where it works correctly to begin with,) or they can deal with the ugly page.
MS doesn't care about standards until the customers make them. And the customers can't put any pressure on them until they're aware there is a problem. So if you want MS to fix their browser, the first step is just to write proper pages and use the cases where IE doesn't handle them correctly to educate the customer. Otherwise you lose by default...
Well that's a good reason to encourage people to get a real browser.
And, in actual fact, the complaint is overblown. IE will display pngs fine as long as they don't use alpha-channel. And in fact IE will display even that subset of pngs fine if QT is installed and set as the default plugin for.pngs, which should be the case.
Most.pngs don't use alpha-channel to begin with, and if you need to use alpha-channel you must use png, gif doesn't support it anyway, so your only choice will be to use png and include a note for IE users telling them how to get QT and set it to be used as the default png plugin, and/or how to get a proper browser. Using gif instead here just isn't a choice - it won't do the same thing.
If only GIF support were this bad in major browsers! We could get rid of that accursed animated GIF altogether. Animated GIFs have no purpose in browsers except banner ads. Anything legitimate could be done better with mpeg.
Much as I feel the same, I don't think that's quite true. Surely there are some cases where it's appropriate to use a light-weight animation and something like.mpg would be overkill.
Sadly, when I think about the cases where I've seen them used, it's as you say. Of course, the same could be said for most everything besides simple html, and even that gets misused at times.
I did finally see a flash page the other day that wasn't trash though, this is a real first! The BBC Dr. Who animations... hopefully this will be a trend, and these technologies will finally start to be used in positive ways instead of always stupid ones.
There's tons of reasons for having the same code. Your suggestion of the same developers having worked on both projects at different times is one of the minor ones. BOTH Linux and SCO share code with BSD (which contains plenty of AT&T UNIX code that's perfectly legal to use under the BSD license, so that's all legit) and also, as long as we're talking about comments, tons of comments in all *nix codebases are going to be word-for-word from POSIX anyway.
So it doesn't surprise anyone with a clue that there are code and comments in common between the two codebases, and it doesn't prove anything. SCO is trying to put on this show for the media pretending this means something but it doesn't. For them to prove any wrong doing they need a lot more than 80 lines that match, they need to be able to prove that those 80 lines are actually evidence of illegal behaviour.
Uh, no. To do that you have to do one of two things:
1) Pull out of the Berne Convention, which states that all works have intrinsic copyright unless otherwise stated.
DING I'll take door number one Bob.
Seriously. The original Constitutional notion of Copyright I could live with. The Berne Convention and all the nastiness it's spawned deserve to die.
You're right, if there was code-filching going on it wouldn't make sense for it to be so obvious.
On the other hand, there are plenty of known reasons for such similarities to be there, without any filching involved. BSD code is Free, and shares large sections line-by-line with it's close cousin SysV. And I believe Linux has borrowed sizable portions from BSD. There's no reason not to, it's perfectly legal to do so under the BSD license.
Also, it sounds like the non-programmer SCO is using is describing matching comments, although he doesn't use that word. A large amount of comments from Linux BSD and virtually any other Unix-like OS are cut and pasted from the POSIX definitions...
So, of course there are lots of shared lines. Everyone knows that, it's no secret. For SCO to prove anything untoward has occured, they need to prove a lot more than that. And they need to show the code to someone that actually knows what the issues involved are, rather than a non-programmer who they can easily impress.
Open source drivers will prevent cheating (you can't hide cheats in there when people can see the source,) and open specs will mean that good open source drivers can be written, which can then be used for all applications, including benchmarking.
I'd be very careful doing this. If I understand correctly, shorting is essentially a bet on the price going down within a particular timeframe. I'd be quite sure SCO stock will be worthless when this is over, but it could take quite some time, and it could easily go up before it goes down...
Quick quiz: Who was it that said "If they kill me, promise me you'll carry my body over the border to Wyoming, cause I wouldn't want to be caught dead in Utah"?
Since the lawsuit is directed at IBM, why doesn't IBM just do a diff AIXSOURCE LINUXSOURCE and publish it? It seems to me that is all anyone has to do to bury this once and for all.
That's a really freakin good idea. It would be a little more complicated than that, of course, but but not all that difficult really. Surely they've already done that internally... well I hope they have. It seems like the most obvious thing in the world now, but I'll admit I hadn't thought of it until I read the post I'm quoting.
Although we all know SCO doesn't have a case, most investors aren't that savvy. Plus, even without a case, you always have a chance in the US - you can pull a crazy judge after all. So the investors are figuring, maybe they win, maybe IBM settles or buys them out... if either of those things happened SCO stock could suddenly be worth a lot more. There's always a good amount of money in the stock market looking for the high risk gamble-moves that could bring a good payoff. Most of those investors, of course, are sensible enough to keep the bulk of their holdings in safe stuff, but they still budget a certain percentage for risky buys... anyway I'd figure those are the buyers. The sellers are probably the SCO executives and their pals, who presumably have a more realistic understanding of their chances.
I tried 3 but it absolutely refuses to run on my system. Never figured out why. Installed it, installation went fine, but it throws an exception 13 every time you try to start it. Deleted it, downloaded the installer fresh, installed again, same thing. Deleted it and never tried again. shrug
The impression I get, and it's nothing more than that, he's not been too specific, but the impression I get is that they brought him in with lots of verbal promises that he'd still have the freedom he wanted. They'd just take care of the boring stuff, and of course take the profits as well as the new owners, but he'd still run things like always. That's what the new buyers always seem to say...
But of course they have their lawyers make sure that they get the legal right to void all those verbal assurances whenever they feel they have to. So... gnutella, they put their foot down. There are discussions, a new modus vivendi is worked out... he thinks he has almost the same old freedom still... but then aimster comes along, and his wings get clipped a little more, now waste... each time they put their foot down and he gets less slack.
I guess the last round of discussions it became clear there wasn't an intersection between what they felt they had to have and what he felt he had to have, so sayonara.
I would also guess the comment about vesting isn't far from the mark, and that made the decision a lot easier for him.
Absolutely, do that, I am no expert on canadian law but from what I've heard your chances are probably good.
I would never sign such a contract, enforceable or not. I was asked to for one job, I struck out the offensive lines, initialed them and pointed out what I had done. They didn't have a problem with it. Another job they insisted, and I walked. Was a shame, it was a lot more money at a time I really needed the job, but I just couldn't do it. I'd rather work at mcdonalds, at least then my free time would still be my own...
They are allegeric to paying license fees for software. Even though it's just a per-seat developer license, they're unwilling to spend a few thousand dollars for every developer who might have to edit the program in the future. Nevermind that 2 weeks of programmer labour nearly exceeds the cost of a license. Partly, this is due to capital expenditures coming from a different part of the budget than payroll. (If a manager has a programmer under him, he can't stop paying him without finding a reason to fire, or admitting that his department is overstaffed. But dollars to equip that guy can be easily diverted to other needs.)
Hahah SNAFU.
Typical. Hrmmm gives me an idea though.
Maybe they should offer Trolltech a barter - they could pay for the license in programmer-hours.;)
Slashdot Mirror
I'll second that.
Having read the article, there are some interesting points. First off they aren't alleging that using your cellphone will make the plane crash, but rather that it might cause some sort of distracting noise in the crews headsets and at worst could conceivably cause a false alarm on one of their warning lights. Sounds a little iffy to me, but ok, better safe than sorry on a plane... then I read on.
Turns out it doesn't matter on new jets - only ones certified pre-'89. So why don't they let people use their mobiles on the newer planes where it's not an issue? Back to the old 'conspiracy theories' on that one. Controllers like control. And the high prices on the sky phones can't hurt either.
Plus, as you kind of hinted at, if a cellphone can really cause even minor systems disruption on a pre-'89 jet, just imagine what someone that was seriously trying to cause a problem could do. It's absurd. If those jets really do have systems that can be so easily disrupted, they should be grounded until they're fixed. So either way, something doesn't add up here, either they're lying (or maybe just stretching the truth very far and very consciously) or they're not even trying to do their job, take your pick.
In an age when we know there are people trying to bring jetliners down, it's absolutely absurd to be flying jetliners that are so poorly insulated against EM interference that a mobile phone is a threat to them. Period.
Hrmm looks like a step in the right direction, know when they put that up?
They have had a really rotten attitude about this for years, refusing to take any responsibility, and even sending out massive spam openly themselves at least once.
As I said, I don't want to rehash the same old arguments here so I'm simply going to ignore most of what you said, other than to state that it's a matter of personal preference.
But one of your comments is way off in a different direction, and I'll address that one.
While I appreciate your humour (and a kernel of seriousness underneath) I explicitly chose to say that RH in particular made the choice of slack in this respect look good. I'm aware that Debian handles the dependency problems much better, that's why I phrased it like I did.
BTW, taste issues aside, your characterisation of slack and dependencies is quite inaccurate for 99.9% of cases at least. You can generally download anything from reputable sources and just have it run, if there's some particular dependency that will be noted. I have had the experience occasionally of installing something that did have an unsatisfied dependency, it's not difficult at all to figure out what they are. You start the program, it says fatal error could not find blah_blah.so you search for blah_blah.so and there you go. Yes, in comparison to debian magically going off and grabbing it for you that might be seen as a slight inconvenience, but it's no big deal. The nice thing is that the system doesn't care if blah_blah.so is there from a package or if you compiled it or how it got there, it just works. What does Debian do? I know Redhat is completely unable to cope in any sane way. RH fans will doubtless jump on me and point out --nodeps and --force, but then when you complain about it shitting it's own database they say it's probably because you used --nodeps and --force and you should never do that. I think the truth is more like you should never compile from source on RH, which sort of ruins the point of running a Free system in my mind. Or you should compile and jump through all the hoops to make an rpm and then install that instead, what a mess that is. On slack you can do the same thing, a lot easier, if you want, or you can just use make install and make uninstall and the package manager will play nice either way. How does Debian cope with source installs? I have used it a little but never did that on a debian system...
Oh, and one more btw, cat isn't an editor, but ed sure is. ;)
They have a lot of nerve to complain about Spam. Yeah, I know they claim to have nothing to do with the spammers selling their products, but if they don't why do they refuse to do anything about it, to the point of refusing complaints? Other companies in that spot take action, Symantec ignores it at best.
Oh yeah, also, 'inappropriate Spam?' Is there any other type? Jeez.
I pretty much agree, but let me point out, contrary to popular belief, Slackware does have a binary packages system, has for ages, and it works real well. Those tarballs have install scripts, you use the package manager and it runs them and keeps a database so they can be cleanly uninstalled. The main difference between Slack and Debian/RH in terms of package management is that it doesn't try to keep track of dependencies - which after fighting with Redhat for awhile seems like a really good choice. I know the argument over whether it's better to do it one way or the other can go on forever, and I'd rather not rehash it yet again, but the point is it's simply incorrect to claim Slack doesn't have package management. It does, it was the first with it I believe, it's just that it follows the slack philosophy of simple elegance and doesn't try to do everything.
I've gotta wonder what you think the alternative here is.
Like I said, what are you comparing to? Debian has a package management system that beats RPM to hell. Slackware uses a simple package manager that works like a champ, and most other distros use a system based on Debian or Redhat... so what are you comparing to?
Every distro has their own equivelant, many are easier to keep track of, and I can't think of one that has had as many problems to post. So what was your point again?
Then you might want to not emphasise that all flavours of free *nix are 100% customisable. But seriously, if you're looking for stability, just switch to Debian or Slack. They're both incredibly stable, don't require a hell of a lot of customisation, and will do whatever you want with the minimum of fuss. I understand that goes for FreeBSD too, but I haven't messed with it enough to say personally. But there's certainly no way I'd willingly suffer through Redhats way of doing things when there are so many perfectly usable alternatives.
Thank you! Yes, that's a very good example, I'll have to remember that.
If you're writing proper web-pages they'll be usable without image-loading at all. So IE users will still be able to use the page, it just won't look as pretty. Which gives you a great opportunity to educate them. There are a couple of solutions - they can get a different browser, they can get QT (most will probably already have it) and set it as the plugin to display pngs in IE (I've been told this works but haven't tested it since I don't use IE except on mac, where it works correctly to begin with,) or they can deal with the ugly page.
MS doesn't care about standards until the customers make them. And the customers can't put any pressure on them until they're aware there is a problem. So if you want MS to fix their browser, the first step is just to write proper pages and use the cases where IE doesn't handle them correctly to educate the customer. Otherwise you lose by default...
Well that's a good reason to encourage people to get a real browser.
And, in actual fact, the complaint is overblown. IE will display pngs fine as long as they don't use alpha-channel. And in fact IE will display even that subset of pngs fine if QT is installed and set as the default plugin for .pngs, which should be the case.
Most .pngs don't use alpha-channel to begin with, and if you need to use alpha-channel you must use png, gif doesn't support it anyway, so your only choice will be to use png and include a note for IE users telling them how to get QT and set it to be used as the default png plugin, and/or how to get a proper browser. Using gif instead here just isn't a choice - it won't do the same thing.
Much as I feel the same, I don't think that's quite true. Surely there are some cases where it's appropriate to use a light-weight animation and something like .mpg would be overkill.
Sadly, when I think about the cases where I've seen them used, it's as you say. Of course, the same could be said for most everything besides simple html, and even that gets misused at times.
I did finally see a flash page the other day that wasn't trash though, this is a real first! The BBC Dr. Who animations... hopefully this will be a trend, and these technologies will finally start to be used in positive ways instead of always stupid ones.
There's tons of reasons for having the same code. Your suggestion of the same developers having worked on both projects at different times is one of the minor ones. BOTH Linux and SCO share code with BSD (which contains plenty of AT&T UNIX code that's perfectly legal to use under the BSD license, so that's all legit) and also, as long as we're talking about comments, tons of comments in all *nix codebases are going to be word-for-word from POSIX anyway.
So it doesn't surprise anyone with a clue that there are code and comments in common between the two codebases, and it doesn't prove anything. SCO is trying to put on this show for the media pretending this means something but it doesn't. For them to prove any wrong doing they need a lot more than 80 lines that match, they need to be able to prove that those 80 lines are actually evidence of illegal behaviour.
Funny that Loos doesn't even mention Hargroves argument though. It seems to me by far the strongest.
DING I'll take door number one Bob.
Seriously. The original Constitutional notion of Copyright I could live with. The Berne Convention and all the nastiness it's spawned deserve to die.
You're right, if there was code-filching going on it wouldn't make sense for it to be so obvious.
On the other hand, there are plenty of known reasons for such similarities to be there, without any filching involved. BSD code is Free, and shares large sections line-by-line with it's close cousin SysV. And I believe Linux has borrowed sizable portions from BSD. There's no reason not to, it's perfectly legal to do so under the BSD license.
Also, it sounds like the non-programmer SCO is using is describing matching comments, although he doesn't use that word. A large amount of comments from Linux BSD and virtually any other Unix-like OS are cut and pasted from the POSIX definitions...
So, of course there are lots of shared lines. Everyone knows that, it's no secret. For SCO to prove anything untoward has occured, they need to prove a lot more than that. And they need to show the code to someone that actually knows what the issues involved are, rather than a non-programmer who they can easily impress.
Au contraire.
Open source drivers will prevent cheating (you can't hide cheats in there when people can see the source,) and open specs will mean that good open source drivers can be written, which can then be used for all applications, including benchmarking.
Hello?
We don't want their stinking source code. We want the specs. Even if their binaries are encumbered, the specs aren't.
I'd be very careful doing this. If I understand correctly, shorting is essentially a bet on the price going down within a particular timeframe. I'd be quite sure SCO stock will be worthless when this is over, but it could take quite some time, and it could easily go up before it goes down...
Quick quiz: Who was it that said "If they kill me, promise me you'll carry my body over the border to Wyoming, cause I wouldn't want to be caught dead in Utah"?
That's a really freakin good idea. It would be a little more complicated than that, of course, but but not all that difficult really. Surely they've already done that internally... well I hope they have. It seems like the most obvious thing in the world now, but I'll admit I hadn't thought of it until I read the post I'm quoting.
No, it's just the gambling instinct.
Although we all know SCO doesn't have a case, most investors aren't that savvy. Plus, even without a case, you always have a chance in the US - you can pull a crazy judge after all. So the investors are figuring, maybe they win, maybe IBM settles or buys them out... if either of those things happened SCO stock could suddenly be worth a lot more. There's always a good amount of money in the stock market looking for the high risk gamble-moves that could bring a good payoff. Most of those investors, of course, are sensible enough to keep the bulk of their holdings in safe stuff, but they still budget a certain percentage for risky buys... anyway I'd figure those are the buyers. The sellers are probably the SCO executives and their pals, who presumably have a more realistic understanding of their chances.
I tried 3 but it absolutely refuses to run on my system. Never figured out why. Installed it, installation went fine, but it throws an exception 13 every time you try to start it. Deleted it, downloaded the installer fresh, installed again, same thing. Deleted it and never tried again. shrug
The impression I get, and it's nothing more than that, he's not been too specific, but the impression I get is that they brought him in with lots of verbal promises that he'd still have the freedom he wanted. They'd just take care of the boring stuff, and of course take the profits as well as the new owners, but he'd still run things like always. That's what the new buyers always seem to say...
But of course they have their lawyers make sure that they get the legal right to void all those verbal assurances whenever they feel they have to. So... gnutella, they put their foot down. There are discussions, a new modus vivendi is worked out... he thinks he has almost the same old freedom still... but then aimster comes along, and his wings get clipped a little more, now waste... each time they put their foot down and he gets less slack.
I guess the last round of discussions it became clear there wasn't an intersection between what they felt they had to have and what he felt he had to have, so sayonara.
I would also guess the comment about vesting isn't far from the mark, and that made the decision a lot easier for him.
It's all guesswork though.
Absolutely, do that, I am no expert on canadian law but from what I've heard your chances are probably good.
I would never sign such a contract, enforceable or not. I was asked to for one job, I struck out the offensive lines, initialed them and pointed out what I had done. They didn't have a problem with it. Another job they insisted, and I walked. Was a shame, it was a lot more money at a time I really needed the job, but I just couldn't do it. I'd rather work at mcdonalds, at least then my free time would still be my own...
Hahah SNAFU.
Typical. Hrmmm gives me an idea though.
Maybe they should offer Trolltech a barter - they could pay for the license in programmer-hours. ;)