Okay, so Netscrape 4.7 is a horrendous creeping horror. Why? Six years of crappy hacks, rushed half-ass W3C standard implementations, and instability. It didn't take mozilla long to realize that continuing to sink resources into the beast was much less humane than simple euthanasia. So the project, which has something like autonomy, decides to pretty much start from scratch. Smart move for the browser, but AOL is left with a twelve-month road plan before expecting even beta-quality code.
AOL now is losing ever more ground to IE. It can't get its name in front of enough of the new Internet eyeballs, so it decides to farm out the old tired beast to providers and even allows them to give it a new skin. If they're going to produce by spring, the "new" product will basically be Netscape 4.71 with AIM and ICQ integrated (gee, big improvement to the browsing experience there, guys). AOL is going to screw over customers with this code, but we expect no less of them.
HOWEVER, the threat here is to mozilla. How does AOL justify mozilla's development now? This developing a "Netscape 6.0" sounds like nothing else than an admission that mozilla is a failure. (Which it's not, of course, but from their perspective, they may just want to abandon it.) Why would they fund two development efforts? I sure hope it won't happen, but the lizard might get a salt bath real soon...:(
If you write an application that depends on a GPLed program, and won't run with a replacement, then it's likely going to infringe. But if you write a program, like a compiler front-end, and it simply has the option to use a GPLed compiler, among others, then it doesn't infringe.
This was the case with NeXT -- their front end was specifically designed to use gcc -- they tried to get around the GPL by saying that it was only a frontend and not a complete program.
A user of a program can only infringe on the GPL by distributing a binary without source, or by using GPLed code and not following the license. Anything else they do is kosher. If I wanted to download GCC and write extensions to it, or use it in a non-GPLed wrapper, this would be fine, regardless of how I used it. Only if I released this would it change.
This is correct. It may not have been clear that I intended to release my java stuff under the GPL so y'all could derive something useful from it, if anything useful was there.:) So in my case -- using the Sun libs, writing a frontend, and releasing the *frontend* under the GPL was bad, but if I wanted to do it on my own computer and not give it to anyone, of course no one's gonna care. I just wanted to write a decent java email client.:)
Could you please explain exactly what the problem was with GPL'ing code that used a few Sun Java libraries? I've seen GPL'd java programs out there, so this is very confusing.
I won't copy the email from RMS I got w/o his permission, but I wasn't trying to write one from scratch. I wanted to use the JavaBeans activation framework (JAF) and the JavaMail 1.1 API. These only come as.jar files and you have to pay to get the source to them, therefore not free.
RMS was quite clear that it would violate the GPL to link, even dynamically, with non-free libraries.
On a slightly different note, let's say you have a proprietary library, but with an open API spec. Can you right a GPL'd front-end that uses that proprietary library? Could you distribute the whole thing, front end GPL'd, backend binary only?
This is the same thing as above. The JavaMail API is "open" in that you can download it and write programs against it (we did it here at my job). But the implementation of the API, the source code for it, is licensed only by money.
The frontend I wanted to write would have used the JavaMail IMAP implementation through the JavaMail API. This is not kosher. Again, what NeXT tried to do with obj-c is similar to this: a proprietary,.o-file-only frontend to the GPL'd version of gcc. This did not fly either.
Hope that answers your questions. Again, though, IANAL...
If I understand your project correctly, then you might not be able to release this thing, at least not under the GPL.
IIRC, use of a GPL interface, unless explicitly declared otherwise à la Linus' amendment to the GPL for Linux, generally means that the using code must also be covered by the GPL.
Historical: NeXT wanted to use gcc as their compiler, so they wrote an obj-c frontend to link to gcc statically. RMS told them that that wouldn't get around the GPL's terms, so they put the whole thing under the GPL. Recent: I wanted to use a few Sun Java libraries in writing a GPL'd frontend for an IMAP client. RMS told me this was not allowed -- even dynamic linking is covered by the GPL.
IANAL, but you may not be able to use the GPL. This sounds like a job for the LGPL, but talking to the FSF about this would definitely be a good idea.
I've noticed that sometimes XMMS (when first installed) defaults to the disk writer plugin, which would do exactly what you describe, with no output sound.
This was correct; it was happily dumping my mp3s to.wav files in my home directory!
Useful, to be sure, but a little hard to get used to.:)
If you have this problem, go to Options/Preferences/Output Plugin and select the OSS Driver plugin as your output device. All should be well now.
I saw something very similar. I'm running Red Hat 6.1 on an Intel and was using the 0.9.5 version of XMMS. Now I can't play my mp3s. I can't work w/o me mp3s! Help!
As I recall, there was a big to-do with DIVX and rental companies like Blockbuster that the supposed benefit of "avoiding late fees" was something completely abhorrent to them?
Blockbuster made $516 million off of late fees last year, according to an article referenced up above. Does anyone really believe that they're blithely willing to let that kind of money go for an unproven technology that has already flopped bad once in the market? I mean, come on!
There are certainly uses for this idea, but they need to get it through their heads that they can't charge us for every viewing of a movie. Bloodsuckers...
Biotech companies will have the patent on the gene(s) that causes Alzheimer's, assuming there is one. They can not only charge exorbitant amounts of money for possible curing mutations or even (gasp) prenatal prevention, but they can prosecute and stop someone else from doing it!
As RMS and others have so often said, the purpose of the patent system is to foster innovation, not to make inventors/discoverers rich. If one gene causes Alzheimer's, how do you claim a patent on that knowledge will induce others to find "alternate" ways of treating it? More importantly, why should it? It's like someone patenting the administration of drugs in pill form to a sick patient. What am I supposed to do if I don't have any pills, stick some leeches on my head and hope my migraine goes away? (Migraines make you want to roll over and die...)
The idiocy, greed, and outright contempt for human decency and welfare disgusts me more than my acid keyboard can relate.
Anyway, it might not be a bad idea to take active steps towards ensuring that something like this cannot happen. Look at the GPL. It guarantees that no rights are taken away from the user by the software it accompanies. So should we, the "users" of our own bodies (and therefore genes) not be denied any rights to them, especially not if such knowledge can raise the quality of life for all people on earth! I say the human genome should be copylefted.
Jim Carrey is a hell of an actor. I said *actor*, not *comedian*. Though he's not bad at that either.
Anybody who hasn't done so, I highly recommend The Truman Show. Came out last year and was pretty much ignored by the media and the people who give out awards. His performance there is amazing, and he was robbed of a Best Actor nomination.
I plan to see this movie too, even though I have no idea who Andy Kaufman was -- had to ask my dad about him. Since I don't know the history of this guy, I'll have to judge the movie based on his acting, right? It's hard to forget Dumb and Dumber, but we should try a bit harder.
I am no philosopher of science, but it seems to me that putting users in control and allowing code forking is to encourage incompatibility.
Then how do you explain the fact that projects like GCC, GDB, and the Linux kernel have not forked? The only big free software package that I can think of right now that has had a major fork has been Emacs, but I don't believe that anyone has suffered and died from incompatibility between the two. Plus, the GPL ensures that users can take whatever they want from each and make whatever suits their needs best. How is this bad?
ESR writes in the Halloween Document about how projects with the most open source distribution have the least tendency to fork -- I refer you to the commentary there on opensource.org.
The most popular pieces of software, word processors and web browsers, are perpetually crippled with respect to backwards and forwards compatibility, much less interoperability. What the market favours is price, availability, and support. Compatibility, it seems, has been left by the wayside.
This is evidence of the competing closed-source products and the business practices that drive them, not of a failure on the part of the marketplace to demand compatibility. The users have no *power* to enforce compatibility if the source is closed. If the source to both browsers were open, I think RMS would say that the incompatibilities and extensions would be written out or merged. In any case, his point is that the lack of freedom for the users creates an environment where such incompatibilities can arise.
In agonizing over MS's embrace & extend tactics, there is another important factor: MS Office, Windows, IE, etc. is the standard. They are the standard for no other reason than that they are on just about every consumer desktop. The important moral of this little tale, aside from the Linux/open source community's need to focus more on the complete computer-phobic neophyte's needs, is that to beat MS, Linux, Mozilla et al. must in a way *become* MS.
Put down the gun, I'm not crazy. Hear me out.
When the IBM PC came out way back when ('81), there were a ton of other companies making PCs. Most of these PCs were better and/or cheaper than the clunker IBM was hocking. But the IBM name sold so many of those "inferior" machines that soon the "better" ones were so much silicon trash. The market share that IBM was able to grab cemented its place as the standard PC right up until the present day, and it looks like it's going to stay there for awhile.
Then came Compaq, and by reverse engineering the IBM -- i.e, getting 100% compatibility -- IBM lost its preeminence as the PC hardware maker.
I think you see where I'm going with this. Since MS is the standard OS, Office is the standard productivity whatsis, etc., we have to prove we can beat them at their own game. I admit I'm not too sure what this means, or how it could be done without subverting the open standards, but one great idea is a full-fledged, kickass web browser. Mozilla might be that browser; we have to wait and see. But open source guys can leverage their stability/adaptability advantages to out-innovate MS. (I think this can be done without breaking standards. Publishing APIs, standard protocol extensions, etc., will keep someone from 0wning the market.
Slashdot Mirror
They're going to WHAT?
Okay, so Netscrape 4.7 is a horrendous creeping horror. Why? Six years of crappy hacks, rushed half-ass W3C standard implementations, and instability. It didn't take mozilla long to realize that continuing to sink resources into the beast was much less humane than simple euthanasia. So the project, which has something like autonomy, decides to pretty much start from scratch. Smart move for the browser, but AOL is left with a twelve-month road plan before expecting even beta-quality code.
AOL now is losing ever more ground to IE. It can't get its name in front of enough of the new Internet eyeballs, so it decides to farm out the old tired beast to providers and even allows them to give it a new skin. If they're going to produce by spring, the "new" product will basically be Netscape 4.71 with AIM and ICQ integrated (gee, big improvement to the browsing experience there, guys). AOL is going to screw over customers with this code, but we expect no less of them.
HOWEVER, the threat here is to mozilla. How does AOL justify mozilla's development now? This developing a "Netscape 6.0" sounds like nothing else than an admission that mozilla is a failure. (Which it's not, of course, but from their perspective, they may just want to abandon it.) Why would they fund two development efforts? I sure hope it won't happen, but the lizard might get a salt bath real soon... :(
Anybody have a more positive take on this?
Last post on this, I promise :)
If you write an application that depends on a GPLed program, and won't run with a replacement, then it's likely going to infringe. But if you write a program, like a compiler front-end, and it simply has the option to use a GPLed compiler, among others, then it doesn't infringe.
This was the case with NeXT -- their front end was specifically designed to use gcc -- they tried to get around the GPL by saying that it was only a frontend and not a complete program.
A user of a program can only infringe on the GPL by distributing a binary without source, or by using GPLed code and not following the license. Anything else they do is kosher. If I wanted to download GCC and write extensions to it, or use it in a non-GPLed wrapper, this would be fine, regardless of how I used it. Only if I released this would it change.
This is correct. It may not have been clear that I intended to release my java stuff under the GPL so y'all could derive something useful from it, if anything useful was there. :) :)
So in my case -- using the Sun libs, writing a frontend, and releasing the *frontend* under the GPL was bad, but if I wanted to do it on my own computer and not give it to anyone, of course no one's gonna care. I just wanted to write a decent java email client.
Could you please explain exactly what the problem was with GPL'ing code that used a few Sun Java libraries? I've seen GPL'd java programs out there, so this is very confusing.
I won't copy the email from RMS I got w/o his permission, but I wasn't trying to write one from scratch. I wanted to use the JavaBeans activation framework (JAF) and the JavaMail 1.1 API. These only come as .jar files and you have to pay to get the source to them, therefore not free.
RMS was quite clear that it would violate the GPL to link, even dynamically, with non-free libraries.
On a slightly different note, let's say you have a proprietary library, but with an open API spec. Can you right a GPL'd front-end that uses that proprietary library? Could you distribute the whole thing, front end GPL'd, backend binary only?
This is the same thing as above. The JavaMail API is "open" in that you can download it and write programs against it (we did it here at my job). But the implementation of the API, the source code for it, is licensed only by money.
The frontend I wanted to write would have used the JavaMail IMAP implementation through the JavaMail API. This is not kosher. Again, what NeXT tried to do with obj-c is similar to this: a proprietary, .o-file-only frontend to the GPL'd version of gcc. This did not fly either.
Hope that answers your questions. Again, though, IANAL...
If I understand your project correctly, then you might not be able to release this thing, at least not under the GPL.
IIRC, use of a GPL interface, unless explicitly declared otherwise à la Linus' amendment to the GPL for Linux, generally means that the using code must also be covered by the GPL.
Historical: NeXT wanted to use gcc as their compiler, so they wrote an obj-c frontend to link to gcc statically. RMS told them that that wouldn't get around the GPL's terms, so they put the whole thing under the GPL.
Recent: I wanted to use a few Sun Java libraries in writing a GPL'd frontend for an IMAP client. RMS told me this was not allowed -- even dynamic linking is covered by the GPL.
IANAL, but you may not be able to use the GPL. This sounds like a job for the LGPL, but talking to the FSF about this would definitely be a good idea.
Cheers.
I've noticed that sometimes XMMS (when first installed) defaults to the disk writer plugin, which would do exactly what you describe, with no output sound.
This was correct; it was happily dumping my mp3s to .wav files in my home directory!
Useful, to be sure, but a little hard to get used to. :)
If you have this problem, go to Options/Preferences/Output Plugin and select the OSS Driver plugin as your output device. All should be well now.
I saw something very similar. I'm running Red Hat 6.1 on an Intel and was using the 0.9.5 version of XMMS. Now I can't play my mp3s. I can't work w/o me mp3s! Help!
As I recall, there was a big to-do with DIVX and rental companies like Blockbuster that the supposed benefit of "avoiding late fees" was something completely abhorrent to them?
Blockbuster made $516 million off of late fees last year, according to an article referenced up above. Does anyone really believe that they're blithely willing to let that kind of money go for an unproven technology that has already flopped bad once in the market? I mean, come on!
There are certainly uses for this idea, but they need to get it through their heads that they can't charge us for every viewing of a movie. Bloodsuckers...
As RMS and others have so often said, the purpose of the patent system is to foster innovation, not to make inventors/discoverers rich. If one gene causes Alzheimer's, how do you claim a patent on that knowledge will induce others to find "alternate" ways of treating it? More importantly, why should it? It's like someone patenting the administration of drugs in pill form to a sick patient. What am I supposed to do if I don't have any pills, stick some leeches on my head and hope my migraine goes away? (Migraines make you want to roll over and die...)
The idiocy, greed, and outright contempt for human decency and welfare disgusts me more than my acid keyboard can relate.
Anyway, it might not be a bad idea to take active steps towards ensuring that something like this cannot happen. Look at the GPL. It guarantees that no rights are taken away from the user by the software it accompanies. So should we, the "users" of our own bodies (and therefore genes) not be denied any rights to them, especially not if such knowledge can raise the quality of life for all people on earth! I say the human genome should be copylefted.
What I'm about to say may shock and amaze y'all:
Jim Carrey is a hell of an actor. I said *actor*, not *comedian*. Though he's not bad at that either.
Anybody who hasn't done so, I highly recommend The Truman Show. Came out last year and was pretty much ignored by the media and the people who give out awards. His performance there is amazing, and he was robbed of a Best Actor nomination.
I plan to see this movie too, even though I have no idea who Andy Kaufman was -- had to ask my dad about him. Since I don't know the history of this guy, I'll have to judge the movie based on his acting, right? It's hard to forget Dumb and Dumber, but we should try a bit harder.
I am no philosopher of science, but it seems to me that putting users in control and allowing code forking is to encourage incompatibility.
Then how do you explain the fact that projects like GCC, GDB, and the Linux kernel have not forked? The only big free software package that I can think of right now that has had a major fork has been Emacs, but I don't believe that anyone has suffered and died from incompatibility between the two. Plus, the GPL ensures that users can take whatever they want from each and make whatever suits their needs best. How is this bad?
ESR writes in the Halloween Document about how projects with the most open source distribution have the least tendency to fork -- I refer you to the commentary there on opensource.org.
The most popular pieces of software, word processors and web browsers, are perpetually crippled with respect to backwards and forwards compatibility, much less interoperability. What the market favours is price, availability, and support. Compatibility, it seems, has been left by the wayside.
This is evidence of the competing closed-source products and the business practices that drive them, not of a failure on the part of the marketplace to demand compatibility. The users have no *power* to enforce compatibility if the source is closed. If the source to both browsers were open, I think RMS would say that the incompatibilities and extensions would be written out or merged. In any case, his point is that the lack of freedom for the users creates an environment where such incompatibilities can arise.
In agonizing over MS's embrace & extend tactics, there is another important factor: MS Office, Windows, IE, etc. is the standard. They are the standard for no other reason than that they are on just about every consumer desktop. The important moral of this little tale, aside from the Linux/open source community's need to focus more on the complete computer-phobic neophyte's needs, is that to beat MS, Linux, Mozilla et al. must in a way *become* MS.
Put down the gun, I'm not crazy. Hear me out.
When the IBM PC came out way back when ('81), there were a ton of other companies making PCs. Most of these PCs were better and/or cheaper than the clunker IBM was hocking. But the IBM name sold so many of those "inferior" machines that soon the "better" ones were so much silicon trash. The market share that IBM was able to grab cemented its place as the standard PC right up until the present day, and it looks like it's going to stay there for awhile.
Then came Compaq, and by reverse engineering the IBM -- i.e, getting 100% compatibility -- IBM lost its preeminence as the PC hardware maker.
I think you see where I'm going with this. Since MS is the standard OS, Office is the standard productivity whatsis, etc., we have to prove we can beat them at their own game. I admit I'm not too sure what this means, or how it could be done without subverting the open standards, but one great idea is a full-fledged, kickass web browser. Mozilla might be that browser; we have to wait and see. But open source guys can leverage their stability/adaptability advantages to out-innovate MS. (I think this can be done without breaking standards. Publishing APIs, standard protocol extensions, etc., will keep someone from 0wning the market.
Someone please, PLEASE port this to NT.. I've got a five-day-old Java process that could use a few pounds of lead upside the head...