So what ? There are several products ( IE just about any commercial development tool ) that don't allow code forking that aren't dead. If what you were implying made any sense, the world would dump commercial development tools and move to free ones. This hasn't happened. If it did happen, GPL'd development tools would also die ( because they are NOT free. )
Are you trying to argue that being able to fork the source tree will lead to wider adoption ? How many projects have become *more* succesful *because* someone has succesfully forked the tree ? There are some, but very few. And it's only happened when the maintainers have essentially fallen asleep on the job.
The QPL doesn't let me ship modified versions of the QT source tarball. How is that "more free"?
You are still allowed to ship the original tree, plus the diff of your tree against theirs. Not too bad IMO.
It is more free because you *can* do whatever you like with derived code. This is ion practice impossible with a number of GPL'd works. For example, the linux kernel has so many contributors that hunting them all down and negotiating with them is impossible. The GPL was *designed to prevent developers from making derived proprietary works, and it does an excellent job of doing just that. Maybe you can buy out the GPL authors, but then, you are asking them to license you their source code under something more free than the GPL.
Qt's restrictive license, while technically "Open Source", has very much hampered its growth
Both assertions are clearly false. QT's license is less restrictive than the GPL which forbids using the code to make non-GPL derivative works. And I am not sure *what* has hampered the growth of QT, which is looking very much unhampered.
BTW, the QPL does *not* rule out code forking. The only issue is that your source must be released in the form of diffs against the original tree.
I have no idea where you get your prediction that Gtk will leave QT "In a niche market" from. Perhaps you are conjecturing that C is the language of the future and C++ is a thing ofthe past ? Or that Gtk is more stable, better documented and easier for developers to use ? The mind boggles.
The trouble is that the QT library is so nice to use. We've got to either develop a GPL replacement, or abandon the whole thing. Either would allow the division between KDE and Gnome to end. KDE may have done a lot for usability, but it has dragged the community into a position they never wanted to be in.
Why do we need everything to be GPL'd ? The problem is that the GPL wants everything to be GPL'd. The fact is that the QT license is *MORE FREE* than the GPL. The GPL places some very firm restrictions on what you can do with derived code, the QPL says that you can do whatever you like with derived code as long as you're prepared to pay, and if you're not prepared to pay, the license behaves more or less like the GPL.
I think it's foolish, and dogmatic to blame QT, Troll and the QPL for this.
BTW, I agree with you about QT being really nice to use (-; But I also think that troll have done an excellent job of releasing something that is licensed in a way that is palatable to free software developers. It is only some of the GPL die-hards in the linux corner that are still whining about it "not being free"
QT is not the problem. The problem is that the GPL does not want to "play" with other licenses. The LGPL and BSD licenses are preferable IMO and do not have these issues.
Firstly, some things are *more* open than "GPL" , perl comes to mind. this is because the authors of this software believe that the GPL is too restrictive.
Secondly, realise that this software is usually *not* written exclusively for linux. It usually runs on many platforms that are not so GPL-centric ( such as the proprietary system V's like HPUX AIX Solaris and SCO or the Free-er than free BSDs including (Free|Open|Net)BSD )
On the other hand, some software is more restrictive. The additional restrictions are usually necessary. For example, the QPL is necessary because Troll Tech sell commercial licenses ( ie for developing propietary software ). in some sense, the QPL is less restrictive than the GPL ( because you can use it to write proprietary software ). Some software needs a not-fully-open license for other reasons: maybe the vendor wants to control the code ( ie doesn't want anyt forks ), maybe the software includes other code that is licensed in a more restrictive manner. So yes, there are some, indeed many good reasons why not everything is GPL'd.
Re:One thing to consider...
on
Which BSD?
·
· Score: 1
I'd like to ask everyone out there, since I really don't know much about Linux or *BSD, where is a good place to find out what hardware has Linux and/or BSD drivers?
If you're going for a new machine, just get linux preinstalled -- go to http://www.tcu-inc.com or somewhere like that. This will make your life *MUCH* easier. Otherwise, try Redhat's hardware compatibility list ( under the support section of their site )
If you want an office suite, Applixware is probbly the best choice, though it's not free.
The other stuff, including C/C++ compilers will come with any distribution. You also might want to play with the development tools ( like Gtk / QT depending on whether you prefer C or C++ )
Re:OpenBSD audits all its code...
on
Which BSD?
·
· Score: 1
OpenBSD's security is wonderful, but correct me if i'm wrong, it's no *remote* root exploits? I remember a bunch of FreeBSD security advisories lately, and the fixes/workarounds were proposed by Mr. De Raadt(sp?) himself. I don't know if that means they applied to OpenBSD though.
No *known* local root exploits either. Of course, the fact that we don't know about them doesn't mean there aren't any -- it would be foolish to consider this as a gaurantee that no exploits will be found. There *have* been local denial of service issues, but next to nothing in terms of root exploits.
Maybe the "fixes" and "workarounds" had already been applied to OpenBSD before someone worked out how to exploit them. The advantage with OpenBSD's security model is they don't wait until someone finds an exploit before they fix things. This is what they call their "proactive" approach to security.
OpenBSD - a fairly recent splinter form/Free|Net/BSD. Very significant security features, though I'm not sure how they affect usability.
Their "stability" concerns ( ie sync writes ) do slow down disk speed. Other than that, the security measures don't greatly affect usability. OpenBSD doesn't have as much software available for it as FreeBSD or NetBSD. If you're looking for something on the desktop, FreeBSD is probably the best choice. OTOH, the other two make good servers.
You know, I couldn't agree more with you. The slashdot reaction disgusts and dissappoints me. We are hearing all of these people who fancy themselves as individuals blindly echoing the "slashdot party line".
I am reminded of the "two minutes of hate" in 1984 -- where everyone gets together and screams their rage at their enemy ( in this case, the enemy is the "geek profiler" ). Thanks for adding something a little different to the debate (-;
PS: If you ever have to take a test like this, answer honestly.
Bzzzt. If you ever have to take a test like this, you had BETTER be honest, because if you're taking the test, you are an ADMINISTRATOR, not a STUDENT. Read the article.
Did you even read the article ? The students don't answer the questions. The questions are for the administrators and they pertain to the students observed behaviour
What, the software , or the ignorance of slashdot's peanut gallery ??? Wahhhh, we're gonna be "geek-profiled" !!! help !
Get a grip. It's not a test. It's software that analyses the administrator's data on the student. The student can't "cheat on it" unless they "cheat" by not behaving in a way that identifies them as an obvious discipline problem.
I didn't read the article, so I don't know who's brilliant plan this was.
Then you're just responding based on your preconceptions and you don't really understand what it's about ? Or did you just want to echo the cry of the "slashdot collective" ?
Schools are horrendously giddy these days, worried about who the next problem might stem from.
If the schools are really this bad, they will make life hell for the students, software or no software.
However, I worry what the schools will do with this information -- it wouldn't surprise me at all of they punished students for being considered "high-risk" by the program.
This would draw lawyers the way blood draws a hungry wolf. The schools are more legally conscious than you think.
Good for the newbies (or uncaring), but not very flexible. Especially not for the serious text editor or email fanatic.;)
Pine allows you to use any editor. Look at your pine config and check "enable alternate editor command" if you want to be able to switch ( with ^_ ) or "enable alternate editor implicitly" if you want it to automatically use your favourite editor ( instead of pico ) . I do this so that I can switch to vim if I'm writing a message longer than 3 lines...
As for email fanatics... real email fanatics use procmail for their filtering.
WWN is a joke. This is the publication that was running stories like "talking cat" , "dog faced girl", "the aliens are coming to take over" they are late -- they were supposed to be here over a year ago... or
maybe you're all aliens and I just didn't notice.
It's good to read if you're in need of a good laugh.
I won't pay a Microsoft tax to run the software I want to use.
If you need to use their operating system, you certainly should buy a license for it. If you have an overwhelming need not to support them, don't use their products.
For serving static webpages ? yes. If they really want to measure how well linux meets tomorrow's demands, testing static page serving isn't good enough.
Yes the test was funded by Micorosft.
Hence it wasn't independent. Hence it is misleading to portray it as independent.
If Linux advocates wanted to prove them wrong in a different scenario why not have Redhat or whomever hire them for a different test?
Because independent tests should be just that. Independent. Why should Redhat stoop to MS's level and perform an expensive marketting stunt ? Other independent pundits ( such as PC Week, C'T and others ) will conduct independent tests anyway.
Okay, Linux advocates insulted Mindcraft's proffesionalism.
You can't expect to publish statements like "NT is 3.5 times as fast as a webserver as linux" without coming under fire. This statement itself is both biased and misleading. The way they summarised their results would be appropriate for a microsoft press release or marketting document but not appropriate for an independent benchmark.
Mindcraft's results where questioned and rightly so. I don't see why their professionalism shouldn't have been questioned given the circumstances ( a Microsoft funded test at MS HQ... under the guise of an "independent test" ).
The tests proved the end result, NT wins in that scenario, Linux looses.
Their "end result" was not about "NT winning", their end result was quantitative data asserting that NT was several times faster than linux. Their quantitative data was vastly inaccurate.
BTW, Linux doesn't "loose" or even "lose"... unless you need to saturate multiple T3s with static content. Wake me up when yopu need to do that.
If nobody pushed Linux as a web server platform before it was suitable, would it ever have become suitable?
Yes. Apache development would have gone on whether or not linux existed. Linux was adopted as a webserver because it was a reliable free operating system that could run on X86 hardware.
At one point, Linux wasn't suitable for anything.
By the time Linus made his famous usenet announcement, Linux was free OS that ran on x86 , and had bash and gcc running on it. That in itself was somewhat exciting from the point of view of the free software movement. So yes, it was suitable even at this stage, as a toy for developers to hack at.
What you just don't seem to get is that no one needed to push it. It did just fine on its merits , without some marketeers making misleading claims about what it could do.
Slashdot Mirror
You are still allowed to ship the original tree, plus the diff of your tree against theirs. Not too bad IMO.
It is more free because you *can* do whatever you like with derived code. This is ion practice impossible with a number of GPL'd works. For example, the linux kernel has so many contributors that hunting them all down and negotiating with them is impossible. The GPL was *designed to prevent developers from making derived proprietary works, and it does an excellent job of doing just that. Maybe you can buy out the GPL authors, but then, you are asking them to license you their source code under something more free than the GPL.
Both assertions are clearly false. QT's license is less restrictive than the GPL which forbids using the code to make non-GPL derivative works. And I am not sure *what* has hampered the growth of QT, which is looking very much unhampered.
BTW, the QPL does *not* rule out code forking. The only issue is that your source must be released in the form of diffs against the original tree.
I have no idea where you get your prediction that Gtk will leave QT "In a niche market" from. Perhaps you are conjecturing that C is the language of the future and C++ is a thing ofthe past ? Or that Gtk is more stable, better documented and easier for developers to use ? The mind boggles.
Why do we need everything to be GPL'd ? The problem is that the GPL wants everything to be GPL'd. The fact is that the QT license is *MORE FREE* than the GPL. The GPL places some very firm restrictions on what you can do with derived code, the QPL says that you can do whatever you like with derived code as long as you're prepared to pay, and if you're not prepared to pay, the license behaves more or less like the GPL.
I think it's foolish, and dogmatic to blame QT, Troll and the QPL for this.
BTW, I agree with you about QT being really nice to use (-; But I also think that troll have done an excellent job of releasing something that is licensed in a way that is palatable to free software developers. It is only some of the GPL die-hards in the linux corner that are still whining about it "not being free"
Secondly, realise that this software is usually *not* written exclusively for linux. It usually runs on many platforms that are not so GPL-centric ( such as the proprietary system V's like HPUX AIX Solaris and SCO or the Free-er than free BSDs including (Free|Open|Net)BSD )
On the other hand, some software is more restrictive. The additional restrictions are usually necessary. For example, the QPL is necessary because Troll Tech sell commercial licenses ( ie for developing propietary software ). in some sense, the QPL is less restrictive than the GPL ( because you can use it to write proprietary software ). Some software needs a not-fully-open license for other reasons: maybe the vendor wants to control the code ( ie doesn't want anyt forks ), maybe the software includes other code that is licensed in a more restrictive manner. So yes, there are some, indeed many good reasons why not everything is GPL'd.
If you're going for a new machine, just get linux preinstalled -- go to http://www.tcu-inc.com or somewhere like that. This will make your life *MUCH* easier. Otherwise, try Redhat's hardware compatibility list ( under the support section of their site )
If you want an office suite, Applixware is probbly the best choice, though it's not free.
The other stuff, including C/C++ compilers will come with any distribution. You also might want to play with the development tools ( like Gtk / QT depending on whether you prefer C or C++ )
No *known* local root exploits either. Of course, the fact that we don't know about them doesn't mean there aren't any -- it would be foolish to consider this as a gaurantee that no exploits will be found. There *have* been local denial of service issues, but next to nothing in terms of root exploits.
Maybe the "fixes" and "workarounds" had already been applied to OpenBSD before someone worked out how to exploit them. The advantage with OpenBSD's security model is they don't wait until someone finds an exploit before they fix things. This is what they call their "proactive" approach to security.
Their "stability" concerns ( ie sync writes ) do slow down disk speed. Other than that, the security measures don't greatly affect usability. OpenBSD doesn't have as much software available for it as FreeBSD or NetBSD. If you're looking for something on the desktop, FreeBSD is probably the best choice. OTOH, the other two make good servers.
I am reminded of the "two minutes of hate" in 1984 -- where everyone gets together and screams their rage at their enemy ( in this case, the enemy is the "geek profiler" ). Thanks for adding something a little different to the debate (-;
Bzzzt. If you ever have to take a test like this, you had BETTER be honest, because if you're taking the test, you are an ADMINISTRATOR, not a STUDENT. Read the article.
Did you even read the article ? The students don't answer the questions. The questions are for the administrators and they pertain to the students observed behaviour
Read the article. The students don't anwer it.
What, the software , or the ignorance of slashdot's peanut gallery ??? Wahhhh, we're gonna be "geek-profiled" !!! help !
Get a grip. It's not a test. It's software that analyses the administrator's data on the student. The student can't "cheat on it" unless they "cheat" by not behaving in a way that identifies them as an obvious discipline problem.
Then you're just responding based on your preconceptions and you don't really understand what it's about ? Or did you just want to echo the cry of the "slashdot collective" ?
If the schools are really this bad, they will make life hell for the students, software or no software.
However, I worry what the schools will do with this information -- it wouldn't surprise me at all of they punished students for being considered "high-risk" by the program.
This would draw lawyers the way blood draws a hungry wolf. The schools are more legally conscious than you think.
No, it doesn't.
Even if you are even moderately intelligent, it would not be that difficult to figure out how to lie in order to past the test.
I don't believe this. The "lies" would be fairly predictable.
The best way to do filtering is via procmail.
Good for the newbies (or uncaring), but not very flexible. Especially not for the serious text editor or email fanatic. ;)
Pine allows you to use any editor. Look at your pine config and check "enable alternate editor command" if you want to be able to switch ( with ^_ ) or "enable alternate editor implicitly" if you want it to automatically use your favourite editor ( instead of pico ) . I do this so that I can switch to vim if I'm writing a message longer than 3 lines ...
As for email fanatics ... real email fanatics use procmail for their filtering.
maybe you're all aliens and I just didn't notice.
It's good to read if you're in need of a good laugh.
Cheers,
If you need to use their operating system, you certainly should buy a license for it. If you have an overwhelming need not to support them, don't use their products.
For serving static webpages ? yes. If they really want to measure how well linux meets tomorrow's demands, testing static page serving isn't good enough.
Yes the test was funded by Micorosft.
Hence it wasn't independent. Hence it is misleading to portray it as independent.
If Linux advocates wanted to prove them wrong in a different scenario why not have Redhat or whomever hire them for a different test?
Because independent tests should be just that. Independent. Why should Redhat stoop to MS's level and perform an expensive marketting stunt ? Other independent pundits ( such as PC Week, C'T and others ) will conduct independent tests anyway.
You can't expect to publish statements like "NT is 3.5 times as fast as a webserver as linux" without coming under fire. This statement itself is both biased and misleading. The way they summarised their results would be appropriate for a microsoft press release or marketting document but not appropriate for an independent benchmark.
Mindcraft's results where questioned and rightly so. I don't see why their professionalism shouldn't have been questioned given the circumstances ( a Microsoft funded test at MS HQ ... under the guise of an "independent test" ).
The tests proved the end result, NT wins in that scenario, Linux looses.
Their "end result" was not about "NT winning", their end result was quantitative data asserting that NT was several times faster than linux. Their quantitative data was vastly inaccurate.
BTW, Linux doesn't "loose" or even "lose" ... unless you need to saturate multiple T3s with static content. Wake me up when yopu need to do that.
Yes. Apache development would have gone on whether or not linux existed. Linux was adopted as a webserver because it was a reliable free operating system that could run on X86 hardware.
At one point, Linux wasn't suitable for anything.
By the time Linus made his famous usenet announcement, Linux was free OS that ran on x86 , and had bash and gcc running on it. That in itself was somewhat exciting from the point of view of the free software movement. So yes, it was suitable even at this stage, as a toy for developers to hack at.
What you just don't seem to get is that no one needed to push it. It did just fine on its merits , without some marketeers making misleading claims about what it could do.