Why is there only one form of not kicking people in the face? I thought not kicking people in the face was about *freedom*, freedom of choice. But it seems that "not kicking people in the face" doesn't allow other violence-level paradigms.
Seriously though, OSS does give freedom of choice. People like Sun were freely choosing GNOME over KDE, which must have been at least partly because of the legal issues with GPL+QPL. Troll, it seems, have chosen to keep themselves competitive, by going with what the market wants.
How long before KDE/(QT 2.2) will be out?
on
Qt Going GPL
·
· Score: 1
When Troll announced that QT v 2.0 would be released under the QPL, there was some lag time before v2.0 was released. I guess we might be up to the next Debian release before a "completely un-QPL-dependent" KDE is released - anybody know more about timescales?
End of OTT accusations?
on
Qt Going GPL
·
· Score: 3
This is good news all round, and should hail the end of a lot of divisive mudslingling. Those who claimed that Troll's aim was to split the free software community are proven wrong; their aim was to make profit and when the GNOME foundation shifted the playing field their actions changed correspondingly. It should also show that the FSF and Debian were anti-QPL and not anti-Troll/KDE:
"I am very pleased to see that Qt is now available under the GPL," said Richard
Stallman, president of the Free Software Foundation. "This is a big win for free
software and a great gift from Trolltech to the community.
"Debian is excited to see Trolltech take this step," said Wichert Akkerman,
leader of the Debian project. "This will encourage the acceptance of Qt as a
building block for free software.
QT IS NOT PROPRIETARY [...] The QPL is an open source license. QT2 is under the QPL.
Quite right.
There, all solved.
Nononono. It's not that simple. The QPL is seriously flawed. Here's an example for you: nick bits of QT and stick them in, say, Zend. You'd now be breaking the law if you distributed that modified app, under clause 3b, which states that you have to give both "initial developers" the right to create proprietory derivatives of each other's work[*] - a right you don't have the power to grant[**].
In other words, if the QPL takes off then the free software world will fragment into hundreds of legally incompatible apps and code re-use between these apps will die an ugly death.
It might be OK to stick Just Another Unimportant QPLed App into a distribution. But making the whole GUI heavily dependent on something with this broken license is a Bad Idea.
[*]They are required to produce QPLed versions too, incidentally.
[**]The problem here is that the two apps have different "initial developers" and the QPL grants extra rights to initial developers.
Personal quibble, chmod go-rwx/home/username? What UNIX god doesn't use the octal numbers, saving themselves 3 keystrokes (chmod 700/home/username) in the process?
There is a subtle difference between the two types of command: the latter is like the former with an implicit "chmod u+rwx/home/username". In this case they're probably the same, but it's no bad thing to be in the habit of using the former cos sometimes it is different.
Acronyms are abbreviations that can be said as a word.
Hmmm, personally I do say "immhoe" and "eye-anal" - what does everyone else do? The funny thing about communicating via text is that pronunciation becomes very non-standard. Do people say "ay-pee-tee" or "apt"? Interesting.
The only difference between [...] Oracle and Windoze is that one is an O/S and one is a database.
Precisely. I couldn't have put it better myself. A dominant database doesn't have as much ability to strangle competition as a dominant OS. Business practices which would not harm the database market could certainly allow one company to exercise market power in the OS market.
Open source just doesn't count for SFA in real-world use.
Maybe not for you, a writer, who can switch word-processor without too much difficulty. If you are a business, building applications on top of somebody's web browser, and that web browser disappears off the market[*], then you might have to change browser and rewrite everything, which is a huge hassle. It's a fairly safe bet that, if an open source program is popular now, then there'll be someone maintaining that program for a long time to come. So you've got more chance of being saved from this hassle. There'll be IE for a long time to come, but in a form that's useful to MS, not necessarily a form that's useful to you. (Remember NT on Alpha?)
[*]: remember that it's sometimes neccessary to abandon your current software for reasons beyond your control. E.g., security flaws, licensing issues ("You may only use this program on a certain computer"), etc..
Who here thinks there is any chance that MSFT will [...] quit development on a product [which is] the dominant player in its market
I think there is a fair chance that Microsoft will choose to do something with their browser which buggers some customers. E.g. they may choose to change the APIs, then drop support for the ones you use in future versions. Then bring out web development tools which assume the latest version of their browser. This is just one scenario. They have been known to leave customers high and dry in the past. For instance, anyone who was relying on them maintaining NT on Alpha is now in a mess. I'm not saying it's impossible for AOL to try something similar, but if there's many people who are dissatisfied then somebody else will fork development. For instance, many people hate XULed components and consider it a waste of processing power. So there is a project around to release Gecko on GTK. If enough businesses wanted Gecko/GTK then they could get it released pretty fast. It's all about alternatives. With IE you're stuck relying on a single company who may not provide alternatives. With Mozilla, if your business's needs are shared by other companies then something can be developed to address the need.
It's not "pi", it's "the circumference of a circle divided by the diameter".
Not necessarily true in non-Euclidean space
Well, ok, I was taking "circle" to mean "circle in the complex plane" and defining length with respect to the Lebesgue measure. Here the complex numbers are defined as an analytically complete algebraically closed field of characteristic zero. Your definition is more elementary, though probably less intuitive.
Go forth and learn before you run your mouth again.
Sorry, we apparently have different definitions of "documented". I wouldn't count it as documented if portions are missing here and there. The Windows API is another good example of this. See all the API calls which the WINE people haven't managed to reverse-engineer. It's no good if people can only make half-baked imitations of IE's rendering. The point about w3c HTML/CSS is that you can implement the spec *precisely*.
Another kid who wants (but does not need or understand) the code.
You correctly deduced that I'm not a mozilla developer. But, as Bob Young says, would you buy a car with the hood welded shut? And how much do you know about car engines?
The point isn't that *I'm* going to save the day. The point is that if AOL were to yank mozilla then *somebody* would continue the development. Moreover, any big organisation could *pay* somebody to continue development. Wheras if IE or Opera get yanked / modified / "upgraded" to something you don't want, then nobody else can do anything.
The source code to mozilla is useful, even to someone who doesn't speak a word of C. Its very *existence* prevents you from being tied to the whim of a single company. Get the big picture here, please.
[It will be] hard for [mozilla to succeed on portable devices] with its processor speed requirements
It's only processor intensive if you have the browser being rendered in XUL. For a portable version, fast native widgets would be much better. The core rendering engine fits on a floppy, so rumour has it.
Statmarket reports that 86% of users are now using IE.
I really hope that's not the case across the entire Internet (and let's face it, sampling client browsers is unreliable, much worse than checking server OS type, for example). Otherwise we can soon kiss goodbye to HTML and hello to the undocumented MS-HTML standard, and stop expecting it to be possible to write a competing browser that will display web pages.
Slashdot Mirror
If you mean QT, it wasn't free(speech) until version 2.0 last[?] year. (The source was available)
Sorry, the quotes were from the end of the article, I should have made that more clear.
Why is there only one form of not kicking people in the face? I thought not kicking people in the face was about *freedom*, freedom of choice. But it seems that "not kicking people in the face" doesn't allow other violence-level paradigms.
Seriously though, OSS does give freedom of choice. People like Sun were freely choosing GNOME over KDE, which must have been at least partly because of the legal issues with GPL+QPL. Troll, it seems, have chosen to keep themselves competitive, by going with what the market wants.
When Troll announced that QT v 2.0 would be released under the QPL, there was some lag time before v2.0 was released. I guess we might be up to the next Debian release before a "completely un-QPL-dependent" KDE is released - anybody know more about timescales?
There's a vital freedom missing from the QPL - the freedom to combine two QPLed apps. See here.
See here.
Quite right.
Nononono. It's not that simple. The QPL is seriously flawed. Here's an example for you: nick bits of QT and stick them in, say, Zend. You'd now be breaking the law if you distributed that modified app, under clause 3b, which states that you have to give both "initial developers" the right to create proprietory derivatives of each other's work[*] - a right you don't have the power to grant[**].
In other words, if the QPL takes off then the free software world will fragment into hundreds of legally incompatible apps and code re-use between these apps will die an ugly death.
It might be OK to stick Just Another Unimportant QPLed App into a distribution. But making the whole GUI heavily dependent on something with this broken license is a Bad Idea.
[*]They are required to produce QPLed versions too, incidentally.
[**]The problem here is that the two apps have different "initial developers" and the QPL grants extra rights to initial developers.
There is a subtle difference between the two types of command: the latter is like the former with an implicit "chmod u+rwx
Not neccessarily. For example, it's still an unsolved problem as to whether a program like this terminates for all positive integer values of i:
$i=$ARGV[0];
while($i != 1) {
($i % 2 == 0) ? ($i
}
Hmmm, personally I do say "immhoe" and "eye-anal" - what does everyone else do?
The funny thing about communicating via text is that pronunciation becomes very non-standard. Do people say "ay-pee-tee" or "apt"? Interesting.
Would you take a nice lampshade produced by slave labour? Cos most people usually say they wouldn't.
But not very realistic. I mean, quite well removed from a good test of real life performance.
This is partly the "Linux kernel" versus "GNU/Linux system" thing. There's a hell of a lot of ideas in something like Debian.
Isn't "-ium" a Latin ending? In which case, wouldn't the unmixed version be "Quintium" rather than "Pentium"?
Precisely. I couldn't have put it better myself. A dominant database doesn't have as much ability to strangle competition as a dominant OS. Business practices which would not harm the database market could certainly allow one company to exercise market power in the OS market.
Not many that were ever really big. Certainly, much less than the proportion of once-popular proprietory software which is now dead and unbuyable.
Maybe not for you, a writer, who can switch word-processor without too much difficulty. If you are a business, building applications on top of somebody's web browser, and that web browser disappears off the market[*], then you might have to change browser and rewrite everything, which is a huge hassle. It's a fairly safe bet that, if an open source program is popular now, then there'll be someone maintaining that program for a long time to come. So you've got more chance of being saved from this hassle. There'll be IE for a long time to come, but in a form that's useful to MS, not necessarily a form that's useful to you. (Remember NT on Alpha?)
[*]: remember that it's sometimes neccessary to abandon your current software for reasons beyond your control. E.g., security flaws, licensing issues ("You may only use this program on a certain computer"), etc..
I think there is a fair chance that Microsoft will choose to do something with their browser which buggers some customers. E.g. they may choose to change the APIs, then drop support for the ones you use in future versions. Then bring out web development tools which assume the latest version of their browser. This is just one scenario. They have been known to leave customers high and dry in the past. For instance, anyone who was relying on them maintaining NT on Alpha is now in a mess. I'm not saying it's impossible for AOL to try something similar, but if there's many people who are dissatisfied then somebody else will fork development. For instance, many people hate XULed components and consider it a waste of processing power. So there is a project around to release Gecko on GTK. If enough businesses wanted Gecko/GTK then they could get it released pretty fast. It's all about alternatives. With IE you're stuck relying on a single company who may not provide alternatives. With Mozilla, if
your business's needs are shared by other companies then something can be developed to address the need.
According to the article you cite, the balance has now tipped the other way.
Well, ok, I was taking "circle" to mean "circle in the complex plane" and defining length with respect to the Lebesgue measure. Here the complex numbers are defined as an analytically complete algebraically closed field of characteristic zero. Your definition is more elementary, though probably less intuitive.
Sorry, we apparently have different definitions of "documented". I wouldn't count it as documented if portions are missing here and there. The Windows API is another good example of this. See all the API calls which the WINE people haven't managed to reverse-engineer. It's no good if people can only make half-baked imitations of IE's rendering. The point about w3c HTML/CSS is that you can implement the spec *precisely*.
You correctly deduced that I'm not a mozilla developer. But, as Bob Young says, would you buy a car with the hood welded shut? And how much do you know about car engines?
The point isn't that *I'm* going to save the day. The point is that if AOL were to yank mozilla then *somebody* would continue the development. Moreover, any big organisation could *pay* somebody to continue development. Wheras if IE or Opera get yanked / modified / "upgraded" to something you don't want, then nobody else can do anything.
The source code to mozilla is useful, even to someone who doesn't speak a word of C. Its very *existence* prevents you from being tied to the whim of a single company. Get the big picture here, please.
It's only processor intensive if you have the browser being rendered in XUL. For a portable version, fast native widgets would be much better. The core rendering engine fits on a floppy, so rumour has it.
I really hope that's not the case across the entire Internet (and let's face it, sampling client browsers is unreliable, much worse than checking server OS type, for example). Otherwise we can soon kiss goodbye to HTML and hello to the undocumented MS-HTML standard, and stop expecting it to be possible to write a competing browser that will display web pages.
Is that greatly different from being handled by a shared library, libjpeg?
Isn't that the same as shared libraries? (which is how the image handling gets done in Mozilla)