Slashdot Mirror
RMS Weighs In On BitKeeper
An anonymous reader writes ". . . and boy, is he pissed! The BitKeeper license, he told the Linux kernel mailing list, is 'the whip hand' of proprietary software. His brief but pungent comment is carried by Linux and Main."
Well spoken, Rik. Until RMS can propose an alternative to BitKeeper, he's just pissing in the wind.
Of course, he has every right to piss in the wind as much as he likes.
--
pants ahoy
But what it comes down to really is why are they using a commercial product at all to develop one of the most sophisticated open source products.
While (most|some) of us dont always agree with what RMS says - he almost always does have very valid points - and this is something that I personally agree with him on.
Maybe its time for someone to start developing a OSS competitor to BitKeeper (without using BitKeeper of course!)
Code something GPL that performs equal to or better than BitKeeper.
I'm not familiar with the arguments of CVS vs BitKeeper. If it is a philosophical argument about a way to do things, then fine. Someone take the CVS code, fork it, and modify it to do what BitKeeper does.
It is a question of the "Software as Religion" vs "Software as Tool".
I'm going to go out on a limb and assume that Linus and the other kernel hackers were pretty proficient with CVS and knew what they were doing. If they are more productive with BK, then there is something wrong with CVS.
Productivity is what counts. This isn't an addiction -- if people want to they can switch back to CVS at any time.
Learning HOW to think is more important than learning WHAT to think.
I preface this by saying that I'm a big fan and proponent of open source but anyhow...
:)
:).
If free software really is better then why do people like RMS need to come out ranting about giving into the evils of proprietary software. Either it's a better model of development or it isn't, and if you have to brow beat people into using it, maybe it's not a better way of doing things. It always seems to me that the people most afraid that their beliefs are wrong are the ones screaming the loudest that they are right.
As long as we have the freedom to write free software and use free software, then I don't really see the point of such ranting. Let the software itself speak quietly of the benefits of being free rather than screaming about the evils of proprietary software.
I'm sure that Linus would be happy to use a more free product if he felt it was suited to the task. Linus, unlike RMS, has never, to my knowledge, been much of a politician. He wrote open source software because it made sense to do so. He's made his choices for the same reason anybody should, because they made good sound technical sense. (if I'm wrong here, please don't hesitate to point it out
Either it will survive and thrive because it has benefits or it will shrivel up and die. Evolution marches merrily on. I mean the process, not the mail client. Though I LOVE the mail client
This sig has been temporarily disconnected or is no longer in service
BK is not preventing Redhat, Suse, Mandrake (et al) employees from using BitKeeper, only those who actively develop Subversion or arch
It's not that they are, it's that they could.
EULAs that dictate terms of use of software (as opposed to EULAs that only address distribution), and companies that support those kind of EULAs, and a very dangerous thing.
That is Stallman's point.
I've had enough abrasive sigs. Kittens are cute and fuzzy.
It's not that BitKeeper shouldn't have the right to choose to whom they give away their product for free; it's just that many feel that it's not appropriate for something intended to be used to maintain an Open Source project such as the Linux kernel.
A man's accomplishments do not put him above critique, questioning, or insult.
I am quite aware of this distinction, however you are missing my point. The license says more than "you cannot use the free version to develop the competition." It also says "if you, the person, do any kind of unrelated development on a competitor (perhaps even submitting a bug report!) you no may no longer use the free version of bitkeeper. Which means that any kernel developer who has become accustomed to using BitKeeper will retstrain himself from aiding competing free software projects at all.
This divides the world of open-source developers into two mutually exclusive groups: those who use bitkeeper for kernel development and those who can ever work on free alternatives.
Isn't it true that for any proprietary software you could presumably get code if you satisfied the owners, paid them enough or whatever? It sounds very much like you see the point as being able to get code, but completely don't 'get' the idea of taking authority out of the original developer's hands.
'Open' source and proprietary source can both be transmitted to different coders given that you satisfy the owner, get their okay. Maybe sometimes they'll agree. Maybe they won't. Maybe they're dead, or out of town, simply unavailable.
Free code, you don't have to satisfy the owner- you have to qualify to BE an owner, at which point you get all privileges of ownership to the extent that the license permits. You're not dealing with a PERSON, you're dealing with a set of legal rules. It's set up in the only way it can be so that if you see code you can HAVE it. It defines a commons that is to some extent self-protecting.
The reason it's that way and not simple public domain is, a certain person saw years ago that the public domain naturally tended towards the proprietary. Stagnation, walling off IP. That person saw fit to arrange for something else to be available.
That's the person you want to get rid of.
That's why I say- go RIGHT ahead. I already know what will happen. You wouldn't be reacting the way you are, if you didn't have some underlying assumptions about how coders should have say over what happens to their code- the idea that people should make their own 'open' licenses, should get to determine their own rules for their openness.
This is no different from very permissive proprietary licenses. The thing about the GNU sphere that is different is that it is set up to balk any effort to impose authority, up to and including RMS's. If he did a new version of the GPL in which he got to come into your house and eat your Fritos and microwave all your nonfree software CDs, GNU would promptly fork and nobody would adopt his new license at all.
Because the GNU/GPL sphere goes against the grain in balking authority from outside itself, it isn't the instinctive choice of anybody- it's giving up your power to something bigger than you. Not a person or authority, but a concept- but it's still giving up your power, to use the GPL. You have legal rights that you intentionally let go of when you use GPL, because legally you are allowed to be proprietary and take ownership of your code.
RMS may or may not have known that what he was creating in Free code would be collectively successful- in fact it's led to huge things, much grander than cooperation among non-Free independent developers (given that Linux is grander than POV-Ray ;) ). However, since it is collectively successful as a result of individual coders giving up some of their power to support the common interest, it makes sense to be interested in the reasons for this.
And it makes sense for RMS to be tireless, unbending, even maddening in his demands that people respect and understand this collective.
You only see him, the figurehead, loudly heading up this thing, and because you don't know better you see him as being in control. You see this guy as the leader, the one in charge.
The difference is, what he's leading, he has no real control over whatsoever- and he likes it that way. He can parade his ego all he wants and it won't do any harm because he HAS no authority over what he's supposedly leading.
Shouldn't you be reconsidering your hostility to the guy in light of the fact that he's not the boss of you, not the boss of Linux, not the boss of GNU, and not the boss of the GPL? You gotta make some allowances for the visionary type. Name one more socially acceptable person who's innovated something like the GPL universe, while not leaving a loophole for himself to direct and control it.
It's amazing that the same people who put themselves in a bad position with their pictures and music would again willingly do so with their kernels.
Bitkeeper could easily change their license with after an acceptable amount of market share, just as GIF, JPEG and MP3 did before them. RMS, as usual, is dead on target here.
-Chris
--an unbreakable toy is useful for breaking other toys--
Well I don't like RMS due to many reasons, mainly that he follows his ideas like a religious fanatic altogether with some very faint ideals on how the world should be. It reminds me those hippy, extreme left revolutionaries that raised Che to the level of a saint, considered Mao's sentences the Bible and USSR a traitor to communist ideals... Meanwhile there are many things on RMS that deserve some high respect. Here is one of them, because, RMS is absolutely right...
I may understand the reasons kernel developers point to hold up to BitKeeper. However they can't and shouldn't ignore the consequences of ignoring the legalities of their move. The fact that BitKeeper is factually an EULA much worse than M$ is something that may put into question the future of Linux. Yes, it is much worse because ethically violates some principals of market, things that were formed not yesterday but millenia ago. It is much worse because it is clearly not a commercial license but a typical feudal decree of the worst kind, in common terms: "you can't do that or do this while you are under my service". This is what is inside this license and it is so superficial that any deviation may turn it void. For some you may have had the chance to read documents from the V up to IX centuries, one may note that feudalism started this way. First they said "while", then they said "because" and later they didn't say nothing as everyone considered it natural...
And to consider things worse:
Larry McVoy of BitKeeper:
"Our position:
"1) No free licenses for our competition, they can buy them if they like.
"2) The software is not open source because the open source business model doesn't have a prayer of supporting the development costs.
"3) If you had built a decent system instead of sitting around and whining, we could be doing something else instead of sitting around listening to your whining."
If we look at the reaction of BitKeeper's owner, we can see that we are really going the worst way. He is ultimative, he is arrogant, for him OSS lacks prayers. But this is not the worst. The worst is that we are a mass of apatic whinners, but he stands higher and listens to the crowd of gentiles. And he only can listen whinning, nothing else, because the brilliance and crystal sound of BitKeeper's castle blinds and deafens everything else.
This is not OSS, this is not Free Software. This is not even the M$ Empire. This is the Black Cathedral.
Normally I do not agree with RMS at all. In this particular case, though, he is flat out right.
BK is infrastructure for the kernel development. This is not something that you happen to rip out and replace on a moments notice in case the license changes to something unacceptable. changing infrastructure is a painful process in which more often than not valueable metadata is lost and in which you have to change processes and retrain people.
In fact, this is precisely the reason while Linus is still using BK despite the controversial license: Convenience vs. da pain of changing.
Now, consider the license change BK put through, and what it mandates: It basically says that you cannot use the free license for Bitkeeper to use Bitkeeper for anything including Kernel development if you or your company happen to work on a version control system.
Can Bitkeeper legitimately impose such license restrictions? Yes, they can - it is their product and they can do whatever they fucking want with it.
Is such a license acceptable for Linux kernel development? Not at all. Despite the fact that there are Bitkeeper-to-CVS and Bitkeeper-to-Subversion and Bitkeeper-to-tgz-Gateways all over the place now, Non-BK users are second class citizens in Linux kernel development. They do not have realtime access, and they do not have proper access to BK metadata at all. Also, patch submissions that do not come in via BK are treated worse than patches that come in via BK - Linus and friends may say they aren't, or they aren't intentionally, but they are - again matters of convenience and infrastructure working against Non-BK users.
Switching from BK to a different systems becomes harder the longer you use them, because the systems accumulates metadata and processes center around BK - the infrastructure mechanics I explained above at work.
Finally, is this situation sane at all? That is, is there at least one party that wins because of this license change?
Well, the people who cannot use the free BK license are certainly not winning. They are shut out and have to use second-class systems to contribute to the kernel. Their enthusiasm and energy that could be used for Linux kernel development is diverted to integrating into a principally broken infrastructure.
The Linux kernel development process certainly is not winning at all. There is war and blodshed on the list, and people are switching or turning their back on kernel development out of principle, others defend Bitkeeper out of purely technical reasons, shutting out all the process problems and political outrage BK creates.
Bitkeeper, the company, certainly isn't winning, too. They created this license in order to discourage the development of BK alternatives. Well, that backfired big time. I couldn't have devised a better Subversion promotion campaign than this particular license change.
Basically, BK has pulled a Qt. That is, they created a great product that is important infrastructure for other projects, with a license that is unacceptable for many people. This is just what Qt was in the beginning of the KDE project: a truly great product in an pivotal infrastructure role for an important project (the Linux Desktop). And just like the license to Qt forced many people to create an alternative to Qt, spawning the Gnome desktop, the BK license change will be a great incentive for many to work on Subversion, Arc and related projects.
So BK actually achived the opposite of what they intended with the license change.
Is there any other party that might be winning? Well, yes. Microsoft. But you already knew that. And hey, they even have a version control system.
RMS doesn't give a damn about convenience, especially for his own sake. He cares about ideals, and challenges everyone else to do the same. Most of his life has been spent trying to make living with his ideals more convenient. Most of us losers spend our lives rationlizing about our ideals until our ideals are convenient. I hope this explains where RMS is coming from and why he says the stuff he does.
Linus Torvalds is more like the majority of people whose ideals meld (by design or otherwise, I don't want to guess) with convenience. This is probably part of why he is fairly popular and seen as a regular guy (unlike RMS). For instance, Torvalds feels the "best" tool for the job should be used regardless. Unfortunately, Torvalds has never taken the time to describe for us exactly what "best" means. I'm sure he doesn't mean anything so sinister as "if unpaid child labor makes the tool cheaper, then it's better!", or so naive as "I'll pay anything for the best, screw value/dollar!".
RMS doesn't give a damn about Larry McVoy's company succeeding with propietary software. I believe (I don't want to put words in RMS' mouth) this is because RMS feel propietary software is worthless in the long run, and hence a waste of society's time, energy, and money. I believe he could win this argument, should he choose to make it. Torvald's feelings on the subject are useless until he defines what he means by "best".
My second personal interjection for this post is about competition versus scratching itches. It's not clear to me that the "competition" from BitKeeper is what will spur the creation of a Free package with similar features -- Torvalds doesn't seem motiviated by competition with Microsoft (he's said such several times), so it's not obvious to me that software authors will be motivated to compete with BitKeeper for the sake of competition alone.
The people in the Free and Open Source communities most affected by the lack of Free BitKeeper alternatives are the Linux (i.e. kernel) developers. Most others seem to be happy enough with existing tools, with just enough disgruntelment that subversion is finally emerging. Thus BitKeeper is providing non-Free itch-relief for the only people in the communities who are have this particular itch. Rik and Linus are dreaming if they think I care about their itches more than I care about mine. Essentially, the only people who are likely to produce BitKeeper alternatives are those kernel developers who refuse to use BitKeeper because of their ideals (if they don't use BitKeeper because they don't like source control, or because they're lazy, then they're unlikely to write a replacement).
That is, only people with strong ideals about Free software are likely to write a replacment for BitKeeper. That means people more like RMS and less like Linus.
-Paul Komarek
I would expect that in addition to this post to the kernel list, RMS is also doing what he can to support BK alternatives. But that probably would not be as public or controversial.
That's precisely the problem. He's making himself look like more of an ass by doing this. It would be 5000% more effective for his post to lkml to say "I've decided to support the (subversion|arch|aegis|pcrs|whatever) project as an official GNU project, and give them funding and development help so that the kernel team will have a viable alternative to BitKeeper. The project page is at http://www.gnu.org/projects/gnukeeper. Any input from kernel developers about what features are needed to replace BK are welcome at the mailing list, gnukeeper-devel@gnu.org"
*THAT* is constructive. Sending a post that says that he's doing something about his situation. Otherwise he looks like every other crybaby group, like the Group of university students earlier this year, or whoever started last weeks huge thread. Saying "I Hate that you're using this closed product!" is all good and well, but if you don't follow it up with "So here's what I'm doing to remedy the situation", you look like an idiot. If you care that strongly about something, especially if you're in charge of or high up in the GNU foundation or the Free Software Foundation, you should put some backing behind your words.
I expect more out of RMS as this; Others have pointed out that the linux environment is kind of a "flagship" of opensource development in GNU/FSF's eyes. You'd think they'd be committed to providing an alternative if they cared. They've done it with lots of other stuff in the unix world, and they've done a very good job. But, it's been since April or so since bk started being used for kernel devel, and not a single project has stepped up with GNU/FSF's backing and put some real work into doing everything bitkeeper does and more.
Or, the reason could be that Larry McVoy's estimates on the time and money required to create another such beast are shockingly accurate, that's why nothing's stepped up.
get 0wned. irc.w30wnzj00.com
Also, the current Perl 5 development also takes place on a Perforce repository, with public-accessible rsync mirrors available.
CVS is painful to use for many common tasks required for large-sized software projects -- its shortcoming on atomic commits, directory versioning, copy-on-write branches, etc. are widely acknowledged.
It is a good thing that, for a relatively small software sector where the neccessary designs are hardly well-understood, proprietary version control systems could use its customer's funding to experiment with advanced features. As long as nobody gets a monopoly on those ideas (read: software patents), they create a pool of ideas that related free software projects can learn from it -- it is really more like a symbiosis, not antagonism.
It is all a very healthy process of ideas in the 'niche' market, first commissioned by paying customers, then trickle down to the low-end market (think iMovie), which makes enough people to appreciate and understand how it should work like, and finally appears as a full-fledged free software -- and everybody can just move forward and play with new things, proprietary or not.
The remark on lkml that the new BKL is 'pulling a Qt' is probably right on the mark, though: The new Perl pumpking (Hugo) wishes to migrate Perforce to Subversion, and help building the missing pieces that people needs. No doubt that many people are doing the same thing right now, myself included.
Let's see.. a few years back Linus just used CVS. Developers began complaining about shortcomings since Linux became too big to be handled in plain CVS etc.
Since CVS is a different beast then the Linux Kernel they did not try to develop something themselves. Besides, that would only distract them from kernel development. To make things worse the competitor, subversions, wasn't much usefull eather since it was still in development (and still is).
So Linux chose BitKeeper. Not because he was pushed or otherwise but because it was, and still is, the best alternative offered then and now. Not because out of convenience but to keep the development going in a strong pace lest it become, like the HURD, a product which seems to be never finished.
BitMover provided the software for free to kernel developers. The only real restriction is that you ae not allowed to use BitKeeper to create a competitive product. Why? Because that's the way BitMover earns it's money to pay it's employees and to fund development for BitKeeper.
So, to be able to use a good product (BitKeeper) which in it's turn made it possible to create a good product (linuxkernel) Linus agreed on the terms that were layed out. If they should have to go back to CVS (which is technologically still possible although people claim otherwise) they would also have to go back to all the problems there were before they started using BitKeeper.
And may eventually a better product come by which is OSS then i would have no doubt everyone would switch to using that.
I can understand RMS's opinion. He sees the flagship of the FSF being "corrupted" by closed source software. This is of course a big blow to him. The FSF always tried to create the tools to do the job themselves. Apparently their jobs were not as big as the Linux kernel though. So their tools became inadequate. However, instead of arguing he could also try to understand the forces that work here. If he could start up a project aiming to replace BitKeeper i think alot of people would jump right in. But if he does not then please be a bit more polite against the people working on the Linux kernel.
So far for this not entirely coherent post...