Slashdot Mirror
Credit and Free Software
Hans Reiser - you're thinking ReiserFS, and you'd be right - has a proposal to slather Free Software with credits for its authors. Good? Bad? This is something the community has generally moved away from, but maybe Reiser has a good point. Newsforge is part of OSDN.
I strongly agree with Reiser on this issue. Although he doesn't
necessarily argue for "slathering" software with attributions, rather
he argues convincingly IMO that the credit for a piece of software
should remain visible to the public users. This can be tastefully
done easily, the point is that leaving the credit for writing the
software in the source code is pointless as most people don't ever
read the source code.
It isn't even so much that someone can't supply a new spalsh screen,
it just needs to include attributions to the original authors. I
think he makes some very interesting and very valid points. It is
interesting to note as he states, that although Stallman is a huge
contributor to many projects, he rarely gets credited on anything.
I feel the same way as Reiser on this, even though Stallman doesn't
want to burden the software with licensing restrictions, it bothers
me that he gets so little in the way of credit for what he has helped
to bring about.
Doug Tolton
"The destruction of a value which is, will not bring value to that which isn't." -John Galt
I'm not saying there's anything wrong with this, mind. I just want to remind some of the zealots that writing Free software is often not the selfless idealistic cause some make it out to be.
OSS belongs to the community. There's really no need for credits. Where would we draw the line if we allowed credits? banner ads? annoying pleas for money? pop up windows?
First, I think that mostly FOSS developers and engineers can appreciate the work that goes into this stuff. So I sortof agree that additional credit might be good as a way of thanking those who have made contributions. Software, especially application software, tends to be like a collage. Do you credit everyone equally? Do some people get more credit? What happens if we forget to thank/credit certain people along the way? I think this could be a disaster and potential hurt the movement.
IMO, the people who are going to care are already seeing the names, either in the source or at the project websites or in CVS. To everyone else, any sort of more obtrusive crediting is just going to be obnoxious, and they're still not going to know any more names then they did before.
The whole point, if anyone still remembers the original goal of the majority of OSS projects, is to write some kick-ass code that's going to be done the Right Way, rather than the short-cutty kludgy way that most programmers are forced to code at work. To me, this includes making the software as elegant and streamlined as possible, and the various methods of ego gratification I can think of (extra splash screens, etc) seem incompatible with this.
Every year during my review, I just pray the words "slashdot.org" aren't mentioned.
legislating developers' name on a screensaver leaves a bad taste in the mouth, honestly. If I had written any significant F/OSS, I would not feel nearly as good about knowing that the license was forcing my name to be displayed on the screen. I would feel nice if someone voluntarily put it up, sure.
Marketers would not want to "un-necessarily'" give credit. Agreed. Not every company selling (services for) open-source code might be doing it for this reason, though.
I can think of two more reasons: (a) they genuinely think that they are reducing information confusing to the (target) user; that their graphic is good; (b) they did not realize that the developers are feeling they are not getting enough credit.
There is merit in the idea that credit to people who write FOSS could be more prominent. There is also a gentler way to do this, IMHO. Like, "Hey Debian dudes! Good work on that release. BTW, my wishlist for the next one is a screensaver that would display names of authors who wrote the packages I installed. Here's a graphic for the background, and here's how I think one could go about it...".
If enough people support this idea and implement it, then the need to enforce it will not be needed. If some notable exception exists, one could consider license as a way to enforce it.
--sdem
Reiser has already lost countless users for his software because he started polluting the kernel messages with "a message from his sponsors".
He should be more concerned with the quality of his software, not with his ego problems. Personally, I find this disgusting. If someone wants to know who wrote the software, he can read the README or ask google.
I don't even have the slightest reference in my free software source code that point back to me, I don't even use huge copyright comments in my software like the GNU project generally does, and yet people have offered me jobs and asked me about my software many times. In general, the people who want to know who wrote the software, do.
Those who try to rub it in their face all the time will cheapen free software for everyone. It's like the "I'm so important!1!!" freeware movement from MS-DOS, and I barely remember a single author from all the software that rubbed their copyright messages it in my face all the time. In contrast, I even learned to know several free software authors personally!
Hans, people are losing data with your file system. I know because I did. Twice. Then I looked at your fsck code and it stunk to the high heavens. You should be concerned with that, not with putting your name in the face of more people.
And what would be the next step? To insert a few seconds delay so people have a chance to see your messages better? Puleeze!
From the article: I propose that we as a community insist that all distros make the default screensaver be one that randomly displays a different detailed credit for one of the authors of Linux software every 60 seconds.
Erm.. Is Python or Perl or Apache or Emacs - "Linux software"? What about FreeBSD or OpenBSD - that's hardly "Linux software"...
I'm surprised to see someone as knowledgeable as Riser make such a blunder - or is it intentional?
grisha.org
vosbert has a really good point. I like the idea of things belonging to the community more than to any individual person. Yet there is a way that an analogy should be made to art here. If you like a sculpture, or a piece of code, you should be able to find the artist/designer. So, perhaps v would say that having credit in the source code is enough, that anyone who really wants to find the designer, could. But the name of the artist adds to the work (yet perhaps only for marketing reasons?)
I spend a minute being torn.
I thought that I was going to post that while Reiser's suggestion that linux have a mandatory screen saver that flashes credit is totally micro$oftesque in its totalitarianism, but his point is well-taken, and oss designers deserve credit. Instead, your comment really convinced me. Anyone who wants to find the designers can by looking in the source code. What user would be searching for a designer who couldn't get it togeher to look in the source code? And what *other* sort of person would care who wrote linux or anything else? The glory of OSS comes from being a shared project in every senes. Let's keep the focus on that. Kudos to vosbert for convincing me.
Open up Adobe Photoshop. You will se a list of names on the splash screen, I assume they are all contributing producers, engineers, and coders. Am I right? It is also in the about screen of most porgrams, give credit where credit is due.
It seems history, even short term history repeats itself. This was tried in the past by the BSD license and was taken out because it is way too onerous. The problems with requiring such credit are well enumerated by the Free Software Foundation in the essay entitled "The BSD License Problem".
On the surface, it sounds like a good idea until you consider what it means to give prominent credit to all the major people who are involved with a piece of software. The larger a project is the larger the number of active participants. More importantly when a project gets large enough it acquires dependencies that provide significant functionality which also are as deserving of credit as the original application developers.
For example I built a news aggregator that is an now a source code available project on GotDotNet that has 70 developers signed up with about a dozen having been active in one shape or the other. There are also dependencies on three external libraries that also provide significant functionality. If this was a commercial product exactly how feasible would it be for me to give prominence to everyone who provided significant value to the application? What metric would I use?
The users of the software probably won't care and most authors who write software really don't do it for fame or they could just plaster their names all over the software (which I rarely see). Perhaps there is something else motivating people to write software.. for instance.. If I sit down and write a zeroconf enabled server daemon for whatever it's probably because I need it or want to use it. Not for fame, because honestly, I could care less who used the shit so long as it worked for me. The blood, sweat and tears pays off in being able to have zeroconf enabled whatever. If other people can benefit then thats great, if they can help make it better thats another plus and if it helps someone else solve a problem in shorter time or makes their life easier then that's gold right there. Usually you get dumps of email from people thanking you for something you just wanted yourself.. It's great.. You get bored? Feel like moving on?? People who were helping with code tend to take up the slack and so the cycle continues.
If people want to know who wrote the software they'll just look it up. I mean in GUI software there is an "About" dialog that exists solely for info such as stuff in cli utils at the start of the program you can put name of author and email address as most other people do. Or through it into a --help argc or something.
Also the idea of having someones name plastered all over your personal computer doesn't make it feel that personal anymore. A user will just begin to tune the shit out, and if you write shit like BIND or BitchX etc you catch enough flack.
I use OSS precisely because it's not personality and marketing driven in user-land. The day Hans' proposal bears fruit is the day I buy a Mac.
I don't remember who said it on the Debian mailing list, but the sentiment was right:
You can either have control, or you can write free software.
Period.
I'm so sick of people trying to cram ads down my throat just because they feel they can get away with it.
:)
Whether it's pop-up ads, spam, TV inset-credit ads, junk mail, telemarketing, ATM fees, TV channel logos, billboards, etc. The long and annoying list goes on and on and keeps growing.
More and more, I'm getting pissed off about the multitude of intrusions on my time and attention. If I cared about whether brand A was better than brand B, i'd look into it myself, otherwise it's just an annoyance to be so informed.
If anyone is particularly interested, or if the software is remarkable in some way, i.e. small, useful, or innovative, then people will find out who's responsible for authoring that piece of work if they care.
But if they don't, then they don't want to endure YET ANOTHER GOD-DAMNED AD.
If the software authors want credit for their work, that's fine, I don't begrudge them that. I'm a software author myself. In fact, I co-wrote one of the most popular ray-tracing programs out there, and my name is on the list of contributors.
The actual software never had my name in it, just in the docs, but people knew me, and had no problem finding out who I was and how to get a hold of me for questions and advice.
I still can list the software on my resume, if I feel that it's relevant to the position I'm seeking. When I do, most people recognise or have heard of it. The fame is still there waiting, bottled up until needed
Anyhow, without being overbearingly egotistical, I managed to get and enjoy my 15 minutes of fame without pissing anyone off and without cramming my name down everyone's throat.
-- You are in a maze of little, twisty passages, all different... --
Hans has done an enormous amount of really high-quality work and deserves fair compensation and recognition for it. He's got every right to have his code display all the credits that he sees fit.
On the other hand, the moment you say that these credits cannot be removed (or suppressed from being displayed by default) then you no longer have a fully free license. That's what the problem was with the old BSD license with the advertising clause (that used to make BSD code incompatible with the GPL until that was removed), and that's the same problem with invarient sections in the GNU Free Documentation License that caused such a stink recently. The GPL doesn't allow any additional restrictions either, and since Hans' code is available under the GPL, the best he can do is ask that people are respectful of the credits. There's no legal recourse if they aren't (other than maybe to get mad, and quit GPL'ing future versions). This leads to the question -- maybe there should be a new free software license that attempts to protect author credits while remaining otherwise free?
That said, I'd have to say that anyone who would remove credits from free software simply because the license doesn't (or can't) prohibit it is being a rude parasite. A good member of the community has more respect for the contributions of others.
try any of these:
or if it has a GUI, go to "Help->About"Have no illusions -- this is what Hans Reiser is worried about, his business. The morality of giving credit-where-due is a red herring.
The debate that sparked this off was Debian removing 20-something lines of crap about sponsors from mkreiserfs. That scares him, because it weakens his power in promoting his sponsors and his brand.
To which I say tough. The GPL was written to ensure that users could make software serve them. If a GPLed program spews unhelpful messages, then anybody has the right to remove them. Incidentally, it's undoubtedly justified in this case, when there's a screenful of rubbish, and the program is regularly used in stressful, recovery situations, potentially on a terminal with no scrollback.
Nobody, I imagine, advocates removing authorship credits entirely, but the GPL does not guarantee free promotion for your company, sponsors, or anything else. If that's what you wanted, you were plain stupid to choose the GPL in the first place.
We're all coders and whatnot, right? We'd all love to think that if we shove our info down peoples throats, they will bow down to our abilities. But realize that if we make our products the best we can make them and then appease our own vanity by placing our credits everywhere, the users will move to a product that doesn't place vanity in front of productivity. Our job is to solve problems and help people get work done, plain and simple.
..with those who take the debian line; as someone anonymously posted to newsforge, "Even the FSF's attempt to require credit in the GFDL is being considered non-free by the Debian project"; and as he or she goes on to point out, Debian ARE usually fairly thorough on principled issues like this. The point, to my mind, of FREE software is that it's free. And whilst the word 'free' has the immediate connutation of lacking monetary compensation, that's not all that the word means. For me, for something to be free requires it not to have certain other obligations attached to it; it goes against my principles - and against the karma of the notion of free software - to tie advertising into freely distributed software in this way. If authors really can't do without this manner of crediting in projects which they've contributed to of their own free will, perhaps they shouldn't have contributed to them for free in the first place?
How many slashdot readers run adware.. and why?.. how long might it be before 'free' software which had advertising in this manner decided that 'trading' adverts with other software authors would increase their user base? Really, it wouldn't take very much bending of the rules before free software looked like free websites. And do we really want geocities on our desktops?
I am a viral sig. Please copy me and help me spread. Thank you
I agree. But why do we need to deal with this issues in the form of license requirements? It's just enough to propose the idea. The actual contributors should be taking care of the details. Additionlly, splash screens don't have enough space to hold all of the contributors. Some names might have *higher* priorities, but who will decide the rankings? Will every developer be happy with the splash screen *policy*? The random display can be a workaround here, but I don't think it's a general solution that will promote the credits of the authors. Actually, we can check the CREDITS file and the Help-About dialog box when we are wondering about the names of the contributors.
However, it is still a good idea to let users know who are actually contributing to the open source software project. But it will be a complicated problem when this credit information display is enforced as a license requirement. Let the project contributors decide what will be displayed in their splash screen; but don't make it a license requirement.