Basically, the author of cdrecord modified its license to not allow modifications of his code by distros without doing so in cooperation with the author.Many critisized the author as a free software-sellout, however, in my opinion, the author has the right to do these modifications and so I am with him. Usually, when distros are zelously patching packages –any package– they create new bugs. And then, many users (not all) are submitting bugs upstream, to the cdrecord author, instead to the distro in question (no, not just SuSE). This, obviously creates problems to the developer, and additional work load that he should not have to go through. It’s his right to protect his time and work, and I respect this.
Slackware is one of the very few distros that –on principle– try to NOT patch software (if something is stable, is stable, otherwise, it won’t be included at all, not even in the development tree of Slackware). This is the main reason why Slackware is so stable (it doesn’t surprise the user with unexpected behaviors) and considerably fast: no patches. However, on big commercial distros like RH or SuSE, the users are dictating the features they want to see, and so the project managers of these distros sometimes need to add/modify features. Problem is, in the “shared library/dependency” nature of Linux (where even trivial non-system apps are very much depending on the ‘right’ version of some other software), these per-distro modifications lead to problems in the long run (speed, backwards compatibility etc).
In the code comment, the author accusses SuSE of viaolating section 6 of the preamble of the GPL, which of course is impossible (you can’t “violate” a section in a preamble). As far as I can tell, SuSE isn’t violating any clause in the GPL liscense (not even if we were to take the aforementioned clause as if it were enforceable) but the author most certainly is volating sections of the GPL (pertaining to linking free and non-free code), which he does not have the right to do. No real damage, following the liscense itself, the whole work becomes non-free (the GPL is invalid if any part of it is violated), his work will simply be forked and/or replaced as it should be in any distribution that does respect the GPL. Although I do respect the authors goal (protecting his time against being wrongly targeted for support) this really does feel more like an ego battle than anything else to me. And the open source world has dealt with such things (in exactly the same way every time — forking or replacing) before.
Erik
“And the open source world has dealt with such things (in exactly the same way every time — forking or replacing) before.”
Yep, thats the ticket. Nobody can or should prevent him from licensing his code however he wishes, but nobody can force us to make use of it either. If you don’t want to play our way, we’ll find someone else to play with.
To the CDRecord author: Good bye. I wish you luck.
Mr Schilling, the CDRecord author, is a great guy and makes a nice job since a long time ago and he has the right to know about hacks in the CDRecord source to mantains the control about her job. I think that is very funny when people disagree about question that follows the good sense.
but still I agree with the author, the distro’s should help him – now they give him more work. why doesnt suse simply work with him, to improve the *original* instead of patching it?
The cdrecord author has done a great job, but now he is depressed after the “immoral” behaviour of some main distro’s. Of couse, the distro’s have the legal right to modify packages, but is it nice of them to cause so much trouble to the cdrecord author?
I’m a happy slack user 🙂
“To the CDRecord author: Good bye. I wish you luck.”
Your loss.
It seems that all he wants is for SuSE to admit it’s their fault when it breaks. Of course, SuSE wants people to buy their stuff so they aren’t going to admit to it if they don’t have to. So they just gloss over the fact that they ship a patched version, and everyone bugs him because they think it’s his code causing all the problems when it isn’t. Can’t blame him for wanting to try to stop this.
Alas, too many people out there who don’t (or just won’t) understand this and will just fork it from a previous version. (see some of the previous comments for examples!). Maybe, as more people become aware of this, he’ll get cooperation from SuSE and others and they will resolve it. Looks like he’s been trying that, though.
Jorg Schilling, I wish you the best.
As i understand it the Suse patches add dvd-write support to cdrecord. The problem is that Joerg Schilling is also offering dvd-write support but the code is not under the gpl. If you are a private user you can download a key every 6 month for free, for commercial use you have to pay.
From what I have read on the subject contrary to Schilling’s claims the source is available from Suse and Suse specifically mentions that this is a patched versions and Suse should be contacted in case of a problem.
Also note that Suse is not the only distro providing a patched version, Mandrake for example does it too.
You should also note that Joerg Schilling while certainly a great programmer is know to be not an easy person to deal with. So I think it is best to take his accusations with a grain of salt. If you don’t believe me, just search for him on the Linux Kernel mailinglist and you’ll see what I mean.
Why doesn’t SuSe just rename their version to cdrecord-SuSe? Like Alan Cox has the Linux kernel -ac branch? That would solve all this trouble. Tempest in a teapot I say.
Aren’t these the same folks who were fighting against providing pure ATAPI burning as opposed to the ugly SCSI emulation hack in the 2.6 version of the kernel?
If my memory isn’t fuzzy, I remember clearly the cdrtools folks saying that you couldn’t and shouldn’t do ATAPI burning and that SCSI emulation was the way to go, much to the frustration of the kernel developers. Yeah, that confused me too. I thought direct ATAPI burning had been used on Windows for years.
Anyway, how is the libburn project coming? I am personally tired of cdrtools and its accompanying politics. It served me well but it’s still a pain to use. Not mention is destroyed several of my CDs last week, something to do with dma issues. Whatever!
Libburn goes way too slow. And its gui front ends (3 of them that I know), even slower, unfortunately.
Regarding the atapi vs scsi, I am with the kernel guys btw: atapi should be used.
Mr. Schilling has the right to change the license of his software.
However, he seems to have some misconceptions about some things and apparently he can over-react very easily. I have the impression that he often thinks people do things to harm him particularly.
Just look at the ide-scsi vs. ATAPI discussions (“ATAPI on Linux is broken”) and so on.
Best wishes to libburn + coaster.
I completely agree to the fact that distros are overpatching cdrecord and that doesn’t help the user in using it easily as for instance on mandrake I had to overwrite the binary with Mr Schilling’s binary.
The problem is that with a fork, distros will as for XFree and Xorg appropriate themselves the project instead of helping it.
It’s a shame.
Not enough. Schilling and cdrecord have become synonymous. It would have to be named something completely different. Some of the other comments makes me think of XF86, Dawes, license, fork, Redhat and x.org. If this is down the same alley then maybe what’s needed is a cd.org?
See here:
http://www.nongnu.org/dvdrtools/
Downstream packagers break stuff all the time. Look at the mess they make of Wine. I can understand his frustration but it’s something all Linux developers go through …
I read the LKML thread and the articles linked in this article. I found the discussion between Schilling and some of the participants in the LKML to be anything but fruitful-lot´s of recrimination etc. I have also used cdrecord for many, many years now-and with a couple of exceptions it has proven to be a trully great program. I can and do appreciate Schillings’ not liking vendors patching his work and then people complaining to him about something not working. I certainly do not wish to loose the ability to use cdrecord because of this licenscing change-this would be trully tragic.
I will refrain from speculating on the legal implications of this license change. But in all honesty it does seem as if SuSE has effectively *already* forked cdrecord. SuSE is patching cdrecord to provide dvd burning-this is a substantial change in the code and likely makes up a large percentage of the modified code. This “fork” also competes with the ProDVD version of cdrecord which the author also sells-he has been giving away a license key for non-commercial user -although the ProDVD is for non-cemmerical uses ‘gratis’ it is specifically not ‘libre’.
I can imagine that this has pissed Schilling off. Schilling has now changed his license to discriminate against SuSE-and has, as seen on his web page, stated that he will no longer offer any support. Although I can understand his anger at SuSE I see a number of problems:
1) Not offering any support at all means he effectively has cut off any user feedback concerning cdrecord. Mostly this hurts him-for there is no way for him to independently test all of the various hardware and software configurations under which cdrecord is being used-so this will have a significant negative impact on the quality of the software.
2) His changing the license, single-handidly, make it apparent that he has not accepted any patches for cdreocrd which were not licensed to him-or he has simply not accepted any patches from anyone. Either way this does not speak good about the cdrecord-the future of such open source software is only as good as the community which supports it-I want to know that many eyes are looking at the code and improving it-it gives me a sense of trust and security-which no propietary software can give me.
3) Undoubtedly the man has to earn a living -but- he seems to have chose a path not disimiliar to what many have chosen: eg. David Dawes, who was releasing certain tools for Xfree86, for which there was immense demand, as seperate commercial non-free software. This approach must be labeled for what it is “the carrot-stick” approach, which releases inferior quality open source software with the promise that the good stuff is only available commericialy. This contradicts the spirit of free software and as such should not be the basis for one of the most central pieces of software in modern Linux distributions. Arguably the author did provide the key for the DVD functionality for free(gratis) for non-commerical uses-which is the only reason cdreocrd+dvd functionality could be included in Linux distributions. But when someone pursues this kind of approach it seems to me that conflict with the open source community-as seen in the LKML- is preprogrammed,ie .inevitable. I fully believe that Schilling could have earned whatever revenue he needed from his software by offering support contracts to his commerical clients-avoiding the difficulties mentioned here.
I hope this issue is resolved and I hope that he continues to provide keys for the DVD functionality. If the issue is not resolved or he refuses to indefinitely provide keys for the DVD usage-his work will be forked-and maybe it should be. But then again I would never look to SuSE for a community fork of software- when SuSE effectively forks something it is for their customer base primarily with little and/or no benefits for the community at large-at least this has been my experience with SuSe-perhaps this is changing now, but vendor lock-in is something that SuSE and Redhat have engaged in repeatedly over the years. At the very least this issue makes it abundantly clear how important it is that distributors actually communicate with the developers and that a relationship of mutual trust and rapore is developed.
On a side note one of the articles’ links links (oh dear gawd my english grammar-my mother tongue!!!!) to the LWN article, “The Value of Middlemen”. Those complaining about Gentoo are actually complaing about Gentoo *not* being a middleman, ie. not being like a ‘normal’ distribution-which predefine the usage of source software for their users. The ‘evil’ Gentoo users are actually dealing with the unmolested source code provided by the developers themselves. Oh the irony. LOL…
Oh stop that excessive slackware pimping, it’s just unprofessional and in this story it doesn’t even have relation to the subject. Slackware is great but please.. it’s uncalled for to mention it’s superior l33tness in every other story.
This is about Jorg not liking the GPL, seeing as he refuses to take patches for stuff like DVD burning – while offering a for-pay version of cdrecord that does support this. It’s his code and it’s his right to alter the license, however he is anything but a good FOSS citizen.
Might I recommend this piece of clothing and an effort to do a truly free, none ugly, cdburning tools – maybe related to libburn or the GObjectified version of cdrecord Burgner?
http://www.cafepress.com/mjg59.13063296
Distros should make it clear that bug reports for anything they’ve packaged shoud come through them. They can then forward anything related to an unpatched package and deal with patched packages related issues. Users can be expected to figure out what’s patched and what isn’t. I don’t know about Suse but Mandrake would rather apply patches to a 0.99 release than use the 1.0 version if that conflicts with the distro release schedule.
Similarly, you can’t expect distros to wait on the project leader to agree with all their patch (although they should still submit them). So they have to patch on their side.
Lastly, developpers should give more of a try to Autopackage or Zero-install where possible to distribute their wares, so that they can have a bigger population of people using their binaries rather than the distros’s ones.
But messing around with the GPL is not smart at all. it opens a legal black hole the community really doesn’t need.
I fully agree with Eugenia and Mr. Schilling on this. If SUSE and other distributors break it (over and over), they should rename it to prevent that Mr. Schilling gets all the PRs for the bugs he didn’ t make.
Kudos to Patrick for not breaking stuff.
or use dvdrtools
Mr. Shilling has always been rather vitriolic towards GNU/Linux and has always been extolling the virtues of Solaris. He is absolutely in his right to license his copyrighted work under which any license he wants, but the rest of the world is perfectly in its right to take up the ball and go play somewhere else.
The problem is that with a fork, distros will as for XFree and Xorg appropriate themselves the project instead of helping it.
It’s a shame.
As for projects introducing licensing shenanigans and then expecting that they will get unquestioned support as always… Well, a fork with a motivated team to pick up where the faltering original has left off, that can be a Good Thing ™.
X.org is a breath of fresh air. I plugged X.org 6.7 into my SUSE and it’s been stunning. I’ve never had such a smooth experience before. X.org is up steam to modernize X to a point of usability that was long overdue. I won’t be crying for the loss of Xfree86.
My guess with CDRecord is that this will be the drop that will relegate Mr. Shilling to Solaris. Or CDRecord will be forked or a new project will replace it. Either way, GNU/Linux will roll on.
>Regarding the atapi vs scsi, I am with the kernel guys btw: >atapi should be used.
Eugenia, I think one of the main problems for Jorg is that Linux keeps changing interfaces depending on how the developers feel in a particular day.
Before 2.6 we were all taught that Linux used ide-scsi because atapi was like scsi and so Linux was very clever in doing this (I saw many many posts of “expert” linux users telling this to new users). Now we are taught Linux is very clever in using atapi.
But see here
http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=130197
so a kernel developer is saying that for ide tape drives the opposite is true: they used to use ide and now they’ll use the disgraced ide-scsi in 2.6!!!!
So much for stable interfaces…
… and that’s OSS (Open Sound System), which became a commercial project simply because the author could not handle the support load.
Support services can play a significant role in the success of any software project, and when it comes to supporting rapidly evolving hardware (clearly the case with CD/DVD writers), even more so. Also it should be noted that cdrecord supports more OSes than just GNU/Linux.
My suggestion would be for the “big” distributions (RedHat, Mandralinux and SuSE) to cooperate with Mr. Schilling rather than forking the cdrecord project. Mr. Schilling was effectively “cornered” by SuSE’s actions, it seems, and that’s a rather poor attitude for any Linux commercial distribution, IMHO.
so a kernel developer is saying that for ide tape drives the opposite is true: they used to use ide and now they’ll use the disgraced ide-scsi in 2.6!!!!
So much for stable interfaces…
—-
2.6 interfaces have remained stable and you misunderstood the bug report. its a bug report not from a kernel developer for a beta series
cdrecord licensing has changed due to profit motive because suse, fedora and everyone else basically use a modified version which is clearly labelled as such.
simply run cdrecord in fedora or suse and they will clearly tell you it has been modified. name change cannot be done due to breaking of compatibility. it clearly satisfies the GPL. what other questions do you people really have?
quote:
“I fully agree with Eugenia and Mr. Schilling on this. If SUSE and other distributors break it (over and over), they should rename it to prevent that Mr. Schilling gets all the PRs for the bugs he didn’ t make.
Kudos to Patrick for not breaking stuff.”
Daniel I have to disagree here. Mr Schilling certainly has rights as in copyright, but he has released cdrecord under the GPL. Once released under the GPL it’s forever under the GPL. Sure he can change the license, but anything released prior to that, under the GPL is legally GPL and cannot be just changed willy nilly.
It does appear to my eyes that Mr Schilling is doing a Mr Dawes and that is not good. I’ve spoken to a few XF86 developers that were treated like crap by the core XF86 team and this is why that project is now dead. Not a single major distro uses XF86 anymore. And neither should they.
That said – I do think that Suse et al should be working with Mr Schilling, and maybe helping him out financially to a point. He has, after all done a lot of work on cdrecord. This would most certainly be fair. And distros do make money on others software.
Dave “morgoth” on Libranet Forums
>nd you misunderstood the bug report. its a bug report not >from a kernel developer for a beta series
The reporter is a kernel developer [email protected]. He clearly says that ide-scsi will be used for ide tapes and asks to switch off ide tape support so that problems will be caught early during the beta (final release will use ide-scsi for ide tapes).
>2.6 interfaces have remained stable
but 2.6 is full of problems, see alan cox emails about permissions in SG_IO, forgotten ide cache flush of a few months ago, …
A 2.4 user that has a cdrw and ide tape will see the former switch from ide-scsi to atapi and the latter from ide-tape to ide-scsi; if Windows did something like that we all would be laughing at it.
Dave: I know the legal implications of the GPL. But from a more ethical point of view it is not a good idea to bury a developer in bug reports (as a distributor), by making all kinds of patches. You either try to work with the author, and add the program as the author releases it to the distribution. Or if the author doesn’t want to integrate the patches, and you feel that your distribution needs them, you can rename the program explicitly to make clear that it is *now* a different program. This is not a matter of legislation, but of good manners.
BTW. see you at the Libranet forums
.
Mr. Shilling has said many uncharitable things about GNU/Linux and refuses to accept DVD-related patches because they interfere with his commercial extensions of cdrecord that provide DVD-recording.
Shilling is effectively doing a bait-and-switch on the whole community of GNU/Linux users and that includes Suse as part of this community. Suse is not at fault here, this guy is. Suse clearly labels its version of cdrecord, so I see no problem there at all.
And please, Eugenia, stop hawcking Slackware constantly. Whatever your motivations for doing so, they take away from your credibility. It simply isn’t true that Suse or Mandrake are any less stable than Slackware and anyone who has run all-three in a production environment will tell you so. Considering the immense lack of configuration tools and functionality of Slackware, what else has it going for it, other than your feverish attempts to shout from the mountaintops of Osnews?
And don’t tell me that this was offtopic, because if that is the case, so would be your comments on this article. IN other words, my comments are commentary on your own comments.
It sounds like you think it is morally wrong that he wants to make money with DVD writing software. He is the author of cdrecord, and the maintainer of the project. If he decides that he wants cdrecord only to be used for writing CDs, that’s his choice. In the end we all need something to eat, and if he’d like to earn his bread with DVD writing tools, that’s fine.
If you don’t agree with his personal choice to draw a line of what he likes to release as free software and what not, use the cdrecord code, extend it with DVD writing capabilities, and *rename it* to take of the load of Mr. Schilling for code he didn’t wrote.
I support open source software, but IMO closed software is not morally bad. Sometimes it is pragmatic or just needed to make closed source software.
There ain’t one truth, you know
.
“if Windows did something like that we all would be laughing at it”
If Microsoft did this (Windows doesn’t do anything on its own) you wouldn’t even notice. I bet you only notice it in linux because it’s ” visible” for you. In your case most probably because it’s mentioned on a summary of LKML like kerneltraffic.org. Interfaces matter, but not to such a degree that regular OS users should care if they did or not. Nor should regular OS users speculate about a developer’s motives and opinions with guessing. Neither should an unbiased OSNews.com editor keep plugging her own opinions in unrelated “news” articles.
<sarcasm>
Another reason not to release your brainchild under GPL in the first line. First they f*** with your project instead of cooperating, and when you realize how stupid you were and change the licensing, they take your last version and give you a kick in the backside.
</sarcasm>
A bit more seriously, though, I see so much hostility in the vicinity of GPL licensing that I am really asking myself if that’s really the kind of future you want to build.
I’d mod that up, but this isn’t that other site.
but 2.6 is full of problems, see alan cox emails about permissions in SG_IO, forgotten ide cache flush of a few months ago, …
—
hey. is that why you call 2.6 full of problems. ok i am convinced that its doing great now just because you will have to point to obscure arguments to back your claim.
honestly 2.6 is doing far better than 2.4 ever did
It seems that every tom dick and Harry has a different way to do just about everything in Linux and it is becoming more and more difficult to consider Linux an OS. How long before distros are completely incompatible with each other so that developers will have to work on basically dozens of versions of their software just to keep up?
Linux on the Desktop? Not gonna happen, it will be SuSE or Redhat or Mandrake, or whatever distro on the desktop, but by then Linux will have become so forked and so incompatible that it will no longer be a community but a bunch of small OSes all trying to take small bits of the market.
I suppose many think that is great, but from an end users point of view, that looks disasterous.
the distro change the program’s name and add a “poweredy by…” in the logo and the about box, so they can patch the program, but any bug report shall go to the distro and not to the original author?
That’s not an optimal solution, you see.
1) We are going to get different distros supporting different patches and different forks of cdrecord.
2) Most support requests will still go to the original author (inevitable).
3) We put to waste all the accumulated experience of the original author and deprive him of revenue.
4) It increases costs for the Linux distributors themselves since they have to individually support their version of a user program.
Wouldn’t it be more rational for SuSE and the author of cdrecord to work out an arrangement, that would include some payment to the author for his continuing support?
We, users, would continue to get cdrecord with all the features we want, the author would be happy to be paid for his services , and the distributors would share the costs of a single maintainer instead of having the burden of their own maintainers for their forked versions…
By the way, if you read the LKML thread you’ll see that both the kernel maintainers and the author of cdrecord show some inflexibility, but nothing that some reasonable mediation could not solve.
Joerg persists in making cdrecord spew warnings about
the 2.6.x kernel, ads for his DVD software, and cryptic
stuff (that needs change, not removal) about real-time
and locked memory. He also persists in using an obsolete
non-UNIX way to refer to devices. The bus,target,lun
stuff is nonsense on USB, ATAPI/IDE, and FireWire.
Regular device names will work for SCSI on Linux, Solaris,
FreeBSD, HP-UX, OpenBSD, AIX, IRIX, and MacOS X.
(and Windows can use drive letters)
When Joerg screwed up, asking for 16 result bytes instead
of 18 and then complaining that he only got 16, he wouldn’t
really admit it. He’s like that, a lot. He’s always blaming
the Linux developers, but never himself. His usage of the
raw kernel header files is in direct violation of what
Linus declared many years ago.
People have been making the claim of forking forever. It’s been 13 years now, when will people wake up and figure out that claim is more FUD than anything?
I’d say that the linux economy is a good mix allowing some level of self correction. If some developer or project gets out of line after some time there will be a reaction to that project and it gets corrected or replaced by something else. Why is internal competition such a bad thing?
It seems that every tom dick and Harry has a different way to do just about everything in Linux and it is becoming more and more difficult to consider Linux an OS. How long before distros are completely incompatible with each other so that developers will have to work on basically dozens of versions of their software just to keep up?
Linux on the Desktop? Not gonna happen, it will be SuSE or Redhat or Mandrake, or whatever distro on the desktop, but by then Linux will have become so forked and so incompatible that it will no longer be a community but a bunch of small OSes all trying to take small bits of the market.
This is a common missconception about the GNU/Linux community. From the outside it may look like everyone is quarreling about licensing, is starting another project to edit text, is making yet another Desktop Environment/Window Manager and basically is rolling their own stuff. Unity seems to be non-existent.
To an extent this is true. But most of the “outsiders” looking in, miss the opposite movement. Countering all the quarreling and inconsistency, there is a gradual and slow movement that unifies the community. All distributions ship with largely the same Desktop Environments, almost all are shipping with OpenOffice.org. Most of them ship bash, mc, ad infinitum. Distro’s are only different in small things.
Different projects like KDE and GNOME are working together to make interoperability as smooth as possible. Think about Drag ‘N Drop, menu’s, Window theming. Most office applications are going to use the OO.o document formats in the near future. Then we have projects like Utopia, Freedesktop.org, LSB, FHS, that produce standards for use in distributions.
When high profile disagreements like this surface, it can look the the whole thing is falling appart, but this is not the case. Projects that act against the best interests of the community at large are either forked or replaced, leaving the troublemakers behind with a dead project. Continuity is almost built in.
If this licensing thing from Mr. Schilling is to become too bothersome, his project will be replaced with a more amenable solution that causes less trouble. Take the XFree86 debacle as a template. It’s replaced with X.org. The ripples didn’t last longer than two months.
When I use cdrecord in Fedora, it first prints out a nice warning that its using RedHat patches to plain cdrecord and to send bug reports to RedHat.
If the author wishes to end all further support requests, then he should do three things:
1.) Forward those messages to /dev/null, or give a simple robotic reply that he doens’t support such requests.
2.) Make this abundantly clear on his webpage. (mplayer did this with regards to the RedHat gcc-2.96, though they were also in the wrong about that compiler for a while)
3.) Make his mailing lists only accessible to those who have licenses from him. Put a regular post about not supporting the “community” version.
4.) Profit ???
But most of the “outsiders” looking in, miss the opposite movement.
And I think its often overlooked that controversy sells copy, even on community-based sites as this one. A lot of times newbies and even some veterans forget that there’s alot more that binds Linux together than seperates it. As you point out very nicely, most arguements end in only two fashions: everyone eventually agrees to get along, or they got their separate paths. Of course, at the time, it seems like a huge problem, but I think we’ve had several high-profile cases in this manner in the last couple years and the Linux community is stronger for it, not weaker.
As far as standards go: they trail a moving target. If you think a distro should only follow standards, then you’re missing the fact that they often take part in creating those standards, and that the creation generally follows experience. The best standards are often those that specify behavior based on common experience, not those that dictate things that noone knows about.
cdrecord has been a crutch for too long. Every Linux cd recording frontend depends on calling this seperate executable maintained by a hostile author. It’s time to replace this beast with a well-designed library.
Forks can and do happen. Sometimes the old project dies.
Sometimes the new project dies. Rarely, they both survive.
emacs/XEmacs: both survive
gcc/egcs: old gcc dies when egcs renamed to gcc
procps/procps: new fork (2.0.xx actually) died
GNOME/whatever: new fork seems to be dying
XFree86/X.org: old project (XFree86) dies
SLS/Slackware: old project (SLS) dies
NetBSD/OpenBSD: both survive
FreeBSD/Dragonfly: both survive, so far
Mozilla/Firefox: old project (Mozilla) dies
There’s been a few others too, at least involving
GUI toolkits.
The talk of forking should serve as a warning to Joerg.
If he heeds that warning by making a few minor changes,
the fork won’t happen. If he ignores the warning, the
fork will happen within a year.
“””I remember clearly the cdrtools folks saying that you couldn’t and shouldn’t do ATAPI burning and that SCSI emulation was the way to go, much to the frustration of the kernel developers. Yeah, that confused me too. I thought direct ATAPI burning had been used on Windows for years.”””
Most good CD-R software on Windows still uses ASPI emulation drivers, implementing Adaptec’s SCSI API, because it works better than the native ATAPI burning functions…
It doesnot matter for linux.
Whatever you use , ide-scsi or atapi they both use SG_IO
interface which is much more better for cd-burning then
what we had in 2.4 Now it you DMA for any block sizes
and not only for one fixed as in ide-scsi in 2.4
Ironically, cdrecord is the only program I’ve used that has killed a Linux box in normal operation. It has been a couple years since it happened, but occasionally cdrecord would trigger some flaw in (I guess, as I never tried to really discover the root of the problem) in ide-scsi that would cause the entire box to die. Usually, this was when I tried to interrupt a burn-in-progress after realizing my iso was wrong. It would either not stop, not die, and couldn’t be killed, and wouldn’t finish either. Sometimes I could get it to die, but similar problems occured when I tried to start a new burn. Usually, the box would lock solid (even to network connections) and I’d have to power cycle the hard way. Sometimes it would flood my terminal with messages about the SCSI layer going haywire.
I’m glad that doesn’t seem to happen anymore
I’ve been using cdrecord for eons. Jeorg has been doing yeoman’s service to the UNIX (note: I didn’t say Linux) community. It’s really sad when the developer has a design for his project and big companies like SuSE run roughshod over his product.
SuSE already fscks the Linux kernel and the way they ship kernel sources (yeah it’s a nightmare to have #ifdef SuSE 8…#ifdef SuSE9….#ifdef SuSE 9.1 in your code! – if someone want’s to find out what they did wrong, grep for
“Stop the kernel madness” thread on LKML back in May-June)
Slackware has been the most faithful Linux distribution – nothing ever breaks and it’s true to Linux’s roots. They ship only stable versions and NEVER EVER patch any program (or atleast without the developer’s consent).
And those who claim Jeorg is protecting his “for pay” version, yes he has a right to make a living!. He’s still giving you cdrecord for free. SuSE doesn’t have a right to modify anything they didn’t write themselves (although GPL allows it).
If this continues, I really hope that FreeBSD and Solaris become popular because as I said in the “Linux standards” article – Linux is a train wreck in the making if nobody follows rules.
best regards
Dev Mazumdar
Jörg Schilling could have also chosen the path of implementing DVD burning into the GPL version of cdrtools and let commercial distributors pay him for the development of that. This way he could have prevented trouble with distributors patching cdrecord, he could have earned a living, and the Linux users would also have been happier.
But at least when it comes to non-programming things, Mr. Schilling doesn’t seem to make very wise decisions, he prefers accusing SuSE and Red Hat of “distributing illegal versions of cdrecord” and “breaking the GPL”.
Sure, it’s his right to change the license, but then he should better not expect any collaboration from the world of software libre.
Mozilla/Firefox: old project (Mozilla) dies
By “dies”, do you mean “continues into the forseeable future”?
http://weblogs.mozillazine.org/roadmap/
http://www.mozilla.org/roadmap.html
Mozilla isn’t going anywhere
And those who claim Jeorg is protecting his “for pay” version, yes he has a right to make a living!.
To what I understand, he is distributing his program under a now modified GPL v2. There is nothing wrong in modifying his version of his program as long as you are distributing the source with the modifications.
SuSE doesn’t have a right to modify anything they didn’t write themselves (although GPL allows it).
If the GPL allows it, where is the problem? If Jörg didn’t wanted this to happen, he shouldn’t have distributed it under the GPL…
Regarding the Slackware rant in this article: Part of journalism is when one writes about an event, he/she adds his or her own opinion. Personally, if a journalist decides to do that, i have no problem with that. I accept thats journalism. Except when it isn’t well argumented such as here. No references are provided by the Eugenia. The rant is far from detailed or in-depth. By the way it is written it can easily be perceived as a flame or troll (a lot of people bite, it seems). Those aspects make it IMO both unprofessional and arrogant.
“On a side note one of the articles’ links links (oh dear gawd my english grammar-my mother tongue!!!!) to the LWN article, “The Value of Middlemen”. Those complaining about Gentoo are actually complaing about Gentoo *not* being a middleman, ie. not being like a ‘normal’ distribution-which predefine the usage of source software for their users. The ‘evil’ Gentoo users are actually dealing with the unmolested source code provided by the developers themselves. Oh the irony. LOL…”
The Gentoo community certainly makes the life of various FLOSS developers hard. Talk to developers of GAIM and Enlightenment. They’re sick of it. Also see the –fuck-upstream part of http://www.funroll-loops.org and yeah that website may not be very friendly to the Gentoo community but its that specific part i’m referring to. Or just check out the mailing lists and/or IRC channels of e.g. GAIM and Enlightenment.
“Slackware has been the most faithful Linux distribution – nothing ever breaks and it’s true to Linux’s roots. They ship only stable versions and NEVER EVER patch any program (or atleast without the developer’s consent).”
Have you ever thought about the reasonable possibility this red line exists because Slackware is a project by 1 person? How on earth would PV regulate all that source he forked and patched? He can’t. That’s why. Because of that it ain’t necessarily better. What if there’s a patch for a software package while there ain’t a new release? For example in CVS? Pat doesn’t patch? Shit happens when that patch solves a security vulnerability in the software. What if the software doesn’t have a developer and/or maintainer anymore? Does Pat patch the software if there’s a useful patch out there? Also, consider
* No Linux distribution is exactly the same.
* No 2 humans agree on everything for 100%.
And you might as well comprehend already why some zealotic “i don’t touch source with a 10 pole feet” is worse than a more pragmatic view on a case-by-case basis.
This situation ain’t a classic “distribution fucks upstream” example either because upstream, in this case, did not behave in a nice manner to the distributor either regarding that DVD functionality. Tis debatable, as some people like Daniel de Kok said, that the software should have been renamed. Yeah, the internals are then still the same and that doesn’t have to be changed. Printing “foobar” instead of “cdrecord” has nothing to do with internals.
Sure, it’s his right to change the license, but then he should better not expect any collaboration from the world of software libre.
What do you call collaboration when the “world of software libre” takes your project and completely ruins it? Think if art was “art libre” and anybody could modify a Monet or a Rembrandt by adding two dabs of paint here and there. Would they still proudly hang a modified painting at the Louvre?
It’s ok to change the frame of a painting (say changing from a metal frame to a guilded gold frame) or the location (say changing it from the Louvre to the LACMA or MOMA) but changing the actual painting? *shudder*
best regards
Dev
I’m glad he’s not-GPL-compatible anymore. I hope this will encourage developers (maybe me) to work on something closer to libburn that doesn’t rely on hackish command line calls. I’m so sick of all the “cdrecord-frontends” out there.
I for one would much rather have apps making calls from dynamic libraries than frontends to the command line.
I hope your analogies are not the pinnacle of your reasoning power because they are tantamount to a meaningless and shallow rant with no bearing on the things you are trying to compare.
Free Software is about collaboration and different developers doing whatever they wish with the code as long as the play by the rules of the GPL.
What a lot of nonsense this “story” is? To those claiming that he has a right to decide what part he GPLs and which he doesn’t. Sure he does, but DVD-RW functionality is now essential and if he doesn’t provide it, someone else will.
And most distributions will not ship non-free software for something so basic as DVD-RW write support.
You chose a very bad comparison.
What do you call collaboration when the “world of software libre” takes your project and completely ruins it? Think if art was “art libre” and anybody could modify a Monet or a Rembrandt by adding two dabs of paint here and there. Would they still proudly hang a modified painting at the Louvre?
The difference in the world of software is that the original remains unmodified. You can copy digital data without effort, contrary to, sticking to your example, a picture.
In software what you described above would be hacking the cvs server, fucking up the code, removing all options to flip back to an older version and deleting all copies of the unmodified source code IN THE WORLD.
People often defend software patents with points quite similar to what you wrote up. No artist ever thought of getting a patent for a special painting technique or for pictures depicting particular things (right, Sir Bill Gates has got all rights for some pictures of well-known spots seen from a certain perspective, i.e. for the Hundertwasser house in Vienna). Remember that in the software world, things are even more trivial, because you don’t have to write _every_ copy of your software, be it free or commercial, you just have to write one.
It’s ok to change the frame of a painting (say changing from a metal frame to a guilded gold frame)
Bad example again, because for art historians it is not OK to change the frame of an image to something which doesn’t fit into the time of origination of the artwork (have you ever seen a Dürer in an aluminium frame?).
Even though you don’t physically damage the picture, you damage the quality of presentation and no good museum would dare to do that unless it’s necessary.
or the location (say changing it from the Louvre to the LACMA or MOMA) but changing the actual painting?
Paintings are restaurated and conserved every 100 – 150 years (depending on what colours and surface are used) and it would be careless not to do so. Adapting a piece of software to be up-to-date, e.g. implementing DVD butning capabilities, is equivalent to this. Now if the patching is done uncleanly and new bugs are caused, that is a mistake and i don’t want to defend SuSE /Red Hat & co. in any way.
Neither do i clame that the open source community did collaborate well with Jörg Schilling or the other way round, but they could have, for example like i wrote in my last post, or by working together with the Linux kernel developers on the ATAPI implementation instead of flaming these people. These things would have certainly been interesting for commercial distributors and i’m sure they would have been willing to sponsor development.
But that’s not my problem, i’m not the cdrtools developer.
But i hope you got my point that you chose a bad comparison in your reply.
For a while, procps was severely hacked up. It’s better now,
but it was broken.
Bold support in top was broken. It’s supposed to be a
toggle. Instead of just changing the default, it was made
to not work at all.
A usage warning with a pointer to the FAQ was removed.
Sure enough, a Google search will reveal that users
were confused over some strange behavior regarding a
user named “x” that was explained in the FAQ.
A title in top was changed. Not bad, but the patch
wasn’t fed back to the maintainer.
An error about a mis-matched System.map file was removed.
Fortunately for everyone, Debian left that in, resulting
in a bug fix. Maintainers need error messages and warnings
in order to track down and fix bugs. When problems are
papered over, they never get fixed.
This is exactly what I wanted to say. This whole ONLY NON-PATCHED SOFTWWARE IS WORTHY line is a load of crap. Patches exist for a reason and are a function of Free software. If every distro just shipped plain Gnome and plain Kernel without and value addons what would be the point? Sorry but Red Hat and Suse add value when they patch software to make a better experience for the user. Saying Slackware is the only stable and “pure” distro where you can be sure nothing is non-standard is a load of BS. That doesn’t make it better than say Red Hat where some no-name like Alan Cox decides that a specific patch will really help. Jeers to Eug for the trolling.
This is all a much to do about nothing and patching and fixing software to suit your specific needs are when Freedom and the GPL are all about. Patching ISN’T evil and everthing shipped “pure” isn’t perfect. Quit the trolling people.
Anyone that says suse and redhat are as stable as slackware deserves a good spanking. FFS the last time i installed redhat even their own redhat-config-network was broken…
No artist ever thought of getting a patent for a special painting technique…
Actually if you look into it you’ll find all sorts of patents for ceramics/glass decoration techniques.
…or for pictures depicting particular things.
Try using the McDonald’s logo on your restaurant sign. It might not be Michelangelo, but it’s still art.
Try using the McDonald’s logo on your restaurant sign. It might not be Michelangelo, but it’s still art.
That’s a trademark, not a patent. There is quite a large difference.
Hmm, that makes me think of the old Eddie Murphy movie _Coming to America_ in which there was a small burger joint called MacDonald’s (not McDonalds), whose trademark was the “Golden Arcs” (instead of golden arches). No point, just thought I’d throw that out there.
That’s a trademark, not a patent. There is quite a large difference.
Not in this context there isn’t. Trademark/Copyright/Patent all prevent people from using things freely. The mechanisms and details might be different, but the end result is just the same.
Guess I should have made the point in a different way. I was trying to say that there are very real restrictions in place on using certain pieces of art.
I personally don’t get why most burning frontends choose to use cdrecord as backend (libburn isn’t quite ready of course).
That is, why would anyone think that wrapping cli gui is good idea?
One is loosing much control over error reporting (what exactly got wrong, and exactly when/where?),
one is getting unnecessary frontend-to-fronted interface (how to feed parameters? do their names/order/whatever change between versions?) etc.
OTOH, most GUI “archivers” wrap gzip/bzip2 cli tools, instead of using libraries, so it’s common “mistake” (I would call it bad programming practice, or “comes directly from scripts”)
If distro’s like Slackware can use non-patched code, than so can the other’s. I certainly wouldn’t want to be doing extra work, Im not going to build my house twice. I will build it once and build it right. If other distro’s keep doing this they should change the name of their software. Like Redhat, before mind you, with KDE. I’ve run into so many distro related problems because of patches. Who do I ask for help, KDE, or cdrecord, when its Redhat who is causing the problem.
If your going to change the code, than build it yourself, and change the name!
Clean up your own mess! Take responsibility for your actions!
Thanks Dad
Free Software is about collaboration and different developers doing whatever they wish with the code as long as the play by the rules of the GPL.
Wasn’t GPL designed to prevent forking?. It’s quite ok for developers to hack other people’s stuff in the confines of their own cubicle or house. But when a distro (who’s main job it is to distribute other people’s work and make money) does it, sorry, you play by the rules!. What is to prevent SuSE (or Novell) from sidelining Linus Torvalds in the future?. Then what will you do?
SuSE should atleast show respect by talking to Jeorg first.
It’s the same situation that happened when Redhat broke KDE when they shipped BlueCurve. KDE guys were pissed but Redhat has more muscle and power to make the KDE developer bow down to their demands. SuSE pulled similar shit for KBUILD and they are just wrong!.
And somebody else said:
If every distro just shipped plain Gnome and plain Kernel without and value addons what would be the point?
Well they should have thought about that before going into business selling other people’s software. This is precisely why LSB cannot ever become viable. Just stop backing SuSE and Redhat and start backing developers!
There are generally accepted rules and if you don’t play by the rules you don’t sell products – for instance, in the US, every car must have the steering on the left. In the UK ever car must have the steering on the right and no exceptions.
best regards
Dev
I am the author of the blog post linked in this article, and I want to make it clear that I did not call Mr Schilling a “free software-sellout”. In fact, the author of the article entirely missed my point.
Please see my updated blog post on the issue: http://weblogs.mozillazine.org/gerv/archives/006272.html
Gerv
…bitching about people modifying his code he released under the GPL? Or did he miss this, straight out of the GPL:
2. You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions:
a) You must cause the modified files to carry prominent notices stating that you changed the files and the date of any change.
b) You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.
c) If the modified program normally reads commands interactively when run, you must cause it, when started running for such interactive use in the most ordinary way, to print or display an announcement including an appropriate copyright notice and a notice that there is no warranty (or else, saying that you provide a warranty) and that users may redistribute the program under these conditions, and telling the user how to view a copy of this License. (Exception: if the Program itself is interactive but does not normally print such an announcement, your work based on the Program is not required to print an announcement.)
Now, maybe I missed the point of what he was bitching about, but, if you don’t like the GPL, do whatever you want, but it isn’t GPL.
“Trademark/Copyright/Patent all prevent people from using things freely”
Oh no, they don’t. At least it depends on the definition of “freely”. Linux is a trademark. You may use the Linux kernel freely. GPLed, MITed, BSDed code may be used freely in some extend; its not the same as Public Domain code, but some of these licenses come near that. Some patents are merely “defensive” patents like the one RedHat has; compare that with for example the GIF patent. I argue here to rather view it on a case-by-case and i’d agree that there is a lot of “prevention” used by these mechanics of law but it ain’t true when you state that they only take away freedom. Not even from the point of view of a standard proprietary software patenting trademarking group.
—-
@ Dev
“Wasn’t GPL designed to prevent forking?”
Did you stopped beating your wife? : The GPL _allows_ forking thus the GPL itself doesn’t prevent it. If the GPL was specifically designed to prevent forking, it wouldn’t allow forking. Just because the GPL recommends or pushes people to collaborate, does not mean it should not allow forking.
Better yet, i’d argue the possibility of forking is like a Sword of Damocles above the developer’s head, saying: “we better remain progressing, collaborating” is a Good Thing for similar reasons as “QT becoming BSD licensed if Trolltech vanishes” and “BitKeeper becoming GPL when BitMover Inc. vanishes”. It means progress will happen anyway as there is no (or less) authority in the way. Because if progress doesn’t happen, we see a fork. Or if people fundamentaly disagree on what road the project should continue to drive.
Examples of that disagreement on roadmap:
NetBSD – FreeBSD (“unfriendly” fork)
Samba – SambaTNG (“friendly” fork)
The former was not a fork in which both parties left with a content feeling. There was damage done in the relationship between the 2 groups.
In the latter example however, there was a consensus and both groups were okay with what was happening.
And more “friendly” forks exist; e.g. when a developer stops working on the software, abandons it, and another developer picks it up (FreeS/WAN, Licq) then that’s what i consider a “friendly” fork.
“What is to prevent SuSE (or Novell) from sidelining Linus Torvalds in the future?”
With freedom comes responsibility.
Did you stopped beating your wife?
Who the F**K are you? Stop hiding behind “dpi” and post real name and a real website! How old are you and do you even write open source code?
Best regards
Dev
“Did you stopped beating your wife? ”
dude that was sick, why wud u even put that?
To put a fallacy of interrogation in the same manner as the original author did except that mine was a tad less friendly. Apparantly i should have explained that more futher to make my point. Excuse me. Anyway, that wasn’t the actual point of my post.
A bit off-topic but…
The Gentoo community certainly makes the life of various FLOSS developers hard. Talk to developers of GAIM and Enlightenment. They’re sick of it. Also see the –fuck-upstream part of http://www.funroll-loops.org and yeah that website may not be very friendly to the Gentoo community but its that specific part i’m referring to. Or just check out the mailing lists and/or IRC channels of e.g. GAIM and Enlightenment.
Some newbies that use shitloads of “optimisation” flags and seem to be unable to understand that bugs should be sent to the Gentoo bug team (and they will send the bugs upstream) but many long-time users are wiser. I use Gentoo but I only use -O2 (or whatever the package gives me). Of course I could simply use Debian but I just plain hate it (the way packages are handled, etc).
There is something a little odd here. The reason given for the licence change is that all the misdirected bug reports are a pain, but on the website it says …
“Personal support for cdrecord has been terminated!
This is not because I am a bad guy but because I am receiving far too much mail that only wastes my time and prevents me from doing serious work. It seems that far too many people are antisocial and just too lazy to read the documentation or are using an account that has been set up incorrectly.”
This message has been there as long as I can remember. And there are no forums or newsgroups.
So, if he isn’t accepting bug reports from random users anyway, why the need for the licence change, other than to make the point he does not like people fiddling with his code?
Personally, I like distros that make as few changes as possible. Yup. Slack.
I wish Joerg Schilling would actually change the license and not even try to call it GPL. That would be less confusing. As best as I can tell, though, he still calls it GPL even though it really looks like it isn’t.
If the code is entirely his, he has a right to license it however he pleases. If he released it under the GPL, then deliberately left out a key capability so that he could sell a proprietary enhanced version, and someone else comes along and patches it, he has no cause for complaint. At least in SUSE 9.1, the cdrecord program’s banner makes it clear that issues should be sent to SUSE. That should take care of any legitimate complaints.
It’s not really practical to rename cdrecord, because it’s invoked by a lot of other programs, such as k3b.
I certainly agree that distributions should work with developers where possible. I’m the project lead for Gimp-Print, and I was a bit unhappy when SuSE 8.1 distributed 4.3.3, which was a development version, rather than whatever 4.2 version was around at the time. That’s not out of any sense of pique; it’s simply that our intent was to maintain 4.2 and support additional printers while developing 4.3 toward 5.0, which I knew would be a long process due to the extensive changes I anticipated. The issue has been long since resolved (and I’m a loyal SUSE user), but it would have been helpful for them to have asked us about our plans. No harm was done — this was very early in the 4.3 lifetime, before compatibility was broken — but it underscores the point that distributions should work with upstream providers *when practical*. But I don’t feel like it showed a lack of respect or anything.
More recently, SUSE patched a bug in Gimp-Print 4.2.6 rather than upgrading to a 4.2.7 prerelease. I completely agree with their action in this case. Prereleases (even in the 4.2 line) aren’t necessarily quite as stable as full releases, and the particular fix was important. It doesn’t bother me in the least that they did this, and I think it was the right thing to do.
That’s not the issue here, however. In this case, Schilling intentionally refuses to add a key capability to cdrecord to protect the market for his proprietary enhancement. There’s no issue of distributions trying to work with him to get this capability in, because he’s not going to add it for reasons having nothing to do with technical merit. This puts distributions in the unenviable position of either having to apply a major patch, denying their users an important capability, negotiating some kind of license with Schilling (thereby driving up their cost), or forcing their users to download new license keys every 6 months. I think that the decision to apply a well-tested patch is absolutely the right one in this situation.
Dev asked what would happen if SUSE tried to sideline Linus in the future. The simple answer is that the market would decide. Linus’s authority derives from the respect he has earned for what he has done and for his *continued* good judgement. If Linus starts doing stupid things on an ongoing basis, the kernel will be forked. Indeed, changes have been forced on Linus against his initial will. That’s the joy of open source — you aren’t constrained by decisions the author has made.
BTW, it’s true that distributor mods can make life a bit of a nuisance for people who like to install the latest and greatest on their machines. I’ve had to be a bit careful with Gimp-Print installations to avoid conflicts. However, this actually led me to a good technical decision about supporting the concurrent installation of multiple versions of Gimp-Print, so that actually worked out for the best.
[quote]
Examples of that disagreement on roadmap:
NetBSD – FreeBSD (“unfriendly” fork)
Samba – SambaTNG (“friendly” fork)
[/quote]
NetBSD and FreeBSD are not forks of eachother. They started independendly, at around the same time, without either project knowing about the other. OpenBSD forked from NetBSD, though.
I agree with the reasons of the writer, and he could only say that any changes maded without send informations to the main developers (him) need to change the program with another name.
like suggested cdrecord-suse ou suse-cdrecord.
I think it’s funny that various comments here basically disclaim the author’s right to make money… free as in free beer, indeed.
As for the “beating your wife” thingy, that’s the standard example for a “loaded question”, a question which you have no chance of answering correctly. Saying “yes” implies you did beat your wife before, saying “no” implies you’re still beating her. That’s what dpi tried to point out: That the question already was invalid. Stop flaming him and engage your brain.
As was mentioned in the article, Slackware has always used non-patched and standard versions of software right down to the kernel – if you need to patch, then it shouldn’t be there!!!
There is an accepted deal with opensource code and this isn’t it. Fork the software and kill the original. Play ball or get out of the game.
Question – if this stuff was GPL’ed isn’t he prevented from going retroactive with his new license? Shouldn’t he have to rewrite his code from scratch?
He can’t go retro-active for the copies out there. He can stop distributing old versions (as long as he stays within the limits of the GPL) and can start to license old and new version under any license he likes (or not license them at all). He can also sell the rights to the source code.
I can understand why the author did that, but I expect the effect to be the exact opposite of what he expected. Most likely someone is going to take the latest version that was released under the GPL and start to maintain it in a different repository. Instead of giving the original author more control over the source, that’ll give him less control. That’s very exactly the point of the GPL: the author can’t prevent anyone from doing it, even retroactively.
His only hope to regain control over his software is probably to proves that it infringes on a patent he owns, but that theory has yet to be tested in court.
The patched versions already claim that they are patched and that the original author should not be bothered with bug reports. UW-IMAP has a similar behaviour, whereby the authors request that patched versions be marked as such.
So, I think Schilling is overreacting, and some dumb users are just not reading the disclaimer that cdrecord is patched.
Also, there is a clear conflict between these patches and the commercial version Shilling sells, so it’s understandable that he doesn’t want these patches to become popular.
He has the right to make money, but since he licensed it as GPL, he no longer has the right to prevent others from doing it for free.
thats the key, he has the right to potentially make money (it isnt gauranteed), but at the same time he is not the god of his own software anymore. he gave that up, he made sure that he would never have absolute control over it. now he complains when that decision comes back and interferes.
my take, tough luck buddy. you made the decision, you cant expect everyone of them to be right.
I run LFS. I couldn’t get the blessed thing to compile for love nor money. I had to get the (heavily patched) DEBs from debian. The docs basically say if you’re a GNU user and you’re having trouble it’s ’cause GNU sucks. I think it even uses a custom make program (which has the same name as a different make program…). He doesn’t seem to believe in autotools.
The website makes it clear that it’s basically shareware. The only reason the distros use it is because there’s nothing else. I hope this license change makes them put some money into dvdrtools.
I agree with Schilling. It’s in his right to protect his code that after all it works very good. So i don’t see where the problem is. When i write a program which is well coded and works well i don’t see why some hot shot from the bigger players have to patch it adding more bugs than before. It’s a matter of respecting the work of the others, which is exactly the spirit of the GPL licence.