Slashdot Mirror
RMS Calls On Linux Developers To Replace BitKeeper
JakusMinimus writes "The developer of BitKeeper has issued fighting words to RMS and he has responded on the LKML,. I remember the flap about this way back when Linus decided on BitKeeper, now it seems many of the non-free concerns were warranted."
I didn't follow this controversy too closely when it came out (I could care less about kernels), but my impression then was that the best argument for moving away from BitKeeper was that Larry McVoy clearly had some personality issues. Now that he's promised to deliberately break interoperability with any compatible product, RMS looks reasonable in comparison. I know Linus is very agnostic about licenses, but it doesn't seem wise to collaborate with someone who has stated his opposition to one of the main reasons so many people use Linux in the first place. Is it really worth dealing with that asshole?
Oh, and LKML's web server isn't a very good advertisement for free software right now.
Read the replies to the LKML post. It's RMS who is trolling, not McVoy. Most of the LKML posters saw it (used to frequent exposure to RMS no doubt), but the Slashdot crowd obviously doesn't, judging from the comments so far. How unusual.
The Free Software movement needs more people like Linus and fewer like RMS, this is yet another proof of that.
Quality, performance, value; you get only two, and you don't always get to pick.
Err, sorry :) Replace all "developers.slashdot.org with lkml.org :)
... Mike Fedyk Scott Robert Ladd "J.A. Magallon" Alan Cox James Simmons "David S. Miller" (Eric W. Biederman) Jeff Garzik Mark Mielke Loup
Possible replies by: Rik van Riel "Trever L. Adams" Michael Buesch Shawn Shawn Larry McVoy Shawn Larry McVoy Jrn Engel Alan Cox "Trever L. Adams" Svein Ove Aas "Trever L. Adams" Alan Cox Valdis.Kletnieks@vt
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
The most popular free software source management tool (CVS) is a complete p-o-s in many respects and unsuitable for large projects and for those with automated builds.
A proper comparison would be with Subversion. CVS, while certainly not a POS, is showing its age. Subversion is supposedly "CVS brought up to date".
Why did Linus go to BitKeeper in the first place?
Welcome to the Turing Tarpit, where everything is possible but nothing interesting is easy.
It's clear now that you want a real database in the back end. Every major app that had an ad-hoc database in the back end has eventually run into problems. Look at Sendmail, BIND, etc. The replacements that work well have a relational DBMS in back. SCCS, RCS, and CVS all had the same problem.
Now that we have several good open source databases available, there's no need for ad-hoc databases. This simplifies the application enormously - all you need is the logic for version management and monitoring.
One thing I'd suggest for a replacement is heavy use of message digests (MD5, etc.) as identifiers and checks. Everything should have an MD5 of its uncompressed version carried along. I'd go so far as to store files by MD5, not name, so that if two copies of identical bits are stored, they're not duplicated. This makes efficient renaming and sharing, the curse of version control systems, effective.
On the user interface side, take a look at Tortoise CVS. It's very convenient, although the integration with CVS is a bit painful, because the protocol (text messages from the server) is ill-defined.
It's worth a look, if only for the ideas.
Easy, automatic testing for Perl.
I don't know about Linux but these are the problems I've encountered:
1. No way to rename files without losing revision info.
2. Ditto for moving files.
3. Can't handle symlinks.
list..
,
e read the FAQ at http://www.tux.org/lkml/
Message: 16
Subject: Re: Bitkeeper
From: Alan Cox
To: Valdis.Kletnieks@vt.edu
Cc: Larry McVoy , Richard Stallman
Linux Kernel Mailing List
Organization:
Date: 18 Jul 2003 23:16:57 +0100
On Gwe, 2003-07-18 at 22:54, Valdis.Kletnieks@vt.edu wrote:
> Now what's this about the "simply irrelevant"?
"in most of the world"
If everyone spent the time replacing bitkeeper instead of beating up
Larry they'd get a lot further. I appreciate beating up Larry is more
fun but....
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel"
in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Pleas
Sigs are dangerous coy things
McVoy once talked on lkml about alternatives and uttered sth. like "if any other system can ever come close in features then it will be arch". Afaik an upload of all kernel versions into an arch repository is under way already for 2-3 weeks as a test, so maybe with some heavy code-reviewing-and-feature-hacking a replacement for BK can be there in a month or so. Arch can be found here: arch</ a>
All these things, software included, share a fundamental characteristic: they are all forms of art - a personal expression of the author that can be appreciated and enjoyed by others for nothing more than what other people perceive that it adds to their own life.
I certainly wouldn't ever say that nobody has any rights to give away any their creations, but I don't think anybody has any rights to dictate whether or not someone else should be allowed to seek financial compensation for their endeavors. Stallman is perfectly within his rights, IMO, to encourage, however strongly, the creation of free alternatives to commercial software, but he doesn't have any business to keep telling particular commercial software authors to stop doing what they do after they've already told him thanks but no thanks.
File under 'M' for 'Manic ranting'
I can't believe this wound up on Slashdot. First of all, the vger list admin already shitcanned the thread from LKML because it's just inappropriate there. Now it moves over here for further idiotic discussion.
If you read the original thread between McVoy and Rory Browne, you'll see that Rory started the whole thing by posting a BitKeeper licensing question to the LKML. I'd almost say Rory was just trolling. From there, McVoy's personality took over and he tossed out a worst-case scenario (rewriting the BK protocol to stay ahead of people trying to reverse-engineer it), and that's what spawned RMS's post.
Okay, so I won't disagree that having open protocols and open software is a good thing. But this is hardly a good example for RMS to pick on. There are completely open CVS and SVN gateways into BK, so at no point is the Linux Kernel code at risk. Major kernel hackers such as Alan Cox don't even use BK themselves - they use CVS or SVN to do all their kernel development.
If you read further down the thread, you'll find that even the most rabid of anti-BK people on the list concede McVoy's point - it's his product, protected by his license, and he can do anything he damn well pleases with it. There should be no more upset over this than when the Linux community went after Linksys to get them to obey the GPL for their router software.
The thread ends with a number of posts by people thanking Larry for what he's done by providing tools that make our kernel get better. That and a number of other "we don't need to rehash this again" messages. It's apparent that people are tired of this issue.
First, it's very mature. Second... just check the feature list for yourslef.
Less is more !
Nonetheless, I have jack all sympathy for that. In most projects, sometimes you hit obstacles and have to take a detour for a bit in order to solve your long term goal. We're doing it in my project, because our goal is not to make a nice framework for writing installers and packages, it is to make software installation on Linux easy. That means we have to solve problems with symbol version locking, desktop integration and so on.
When the Gnome and KDE projects decided they were going to write a desktop, they took detours into writing component technology infrastructure. It's not directly related, but it was a problem they needed to solve in order to write a good desktop.
So, the kernel developers need to look at what their goals are. Are they simply to write some nifty kernel components, or are they trying to produce something useful to the free software community as a wider whole. Which is more important, a slight increase in productivity today and getting screwed tomorrow, or long term stability?
If I was Linus, the answer would be obvious, and yes I absolutely would be willing to work on the tools I needed to do my job, if no suitable ones were available. If people didn't like the fact that kernel development slowed down during this period, then they had better damn help me, or take responsibility for the project so I could get on with something else.
In this case, the easy way out was chosen - we'll let this nice man give us his product for free, and that means we can carry on doing what we enjoy. They just forgot to think about the consequences of using this particular tool, and now things have gone pear shaped.
Well, I can't say I have much sympathy. You've got to keep the long view.
Far be it from me to be confrontational on Slashdot, but this isn't actually true. There is, in fact, an article put out by the Free Software Foundation on this very subject.
RMS has stated time and again that he does not care if software is free as in beer-he believes it should be free as in freedom. If I am not horribly mistaken, the FSF themselves sell quite a bit of GPL'd software (and t-shirts and other goodies).
Larry McVoy claims that the Bitkeeper licence agreement will prevent cloning of Bitkeeper, thanks to a clause that "states that you can't use BK if you are developing a similar system, i.e., a clone."
This clause sounds suspiciously unenforcable. Are licence agreements powerful enought to allow vendors to specify/limit product use in this way?
---
From: Larry McVoy
To: Richard Stallman
Subject: Re: Bitkeeper
Cc:
Date: Fri, 18 Jul 2003 13:44:05 -0700
I'm trying hard to stay out of this, I think Richard may be trolling,
but I need to make sure that people understand that what Richard is
suggesting is violation of our license and copyright.
On Fri, Jul 18, 2003 at 03:51:36PM -0400, Richard Stallman wrote:
> I think it would be appropriate at this point to write a free client
> that talks with Bitkeeper, and for Linux developers to start switching
> to that from Bitkeeper. At that point, McVoy will face a hard choice:
> if he carries out these threats, he risks alienating the community
> that he hopes will market Bitkeeper for him.
Our license states that you can't use BK if you are developing a similar
system, i.e., a clone. Without using BK it's impossible to reverse
engineer BK to create the clone. So your message seems to be saying
"it would be appropriate at this point to violate the BitKeeper license
in order to write a free client which talks with BitKeeper".
Are you really instructing people to go out and violate our license?
This is already the case in many non-US jurisdictions. German law, for example, does not allow you to sign away your copyright to anything you create. Copyrights are never transferrable but, of course, you can get a license to use the material. Similarly, you can't give away your right to a court hearing (making binding arbitration an impossibility in such a jurisdiction).
I think that some stuff you ought not be able to give away. Otherwise, you as an individual might find yourself at a disadvantage at the bargaining table. If, for example, software company A had a large market share and as part of its sales agreements stipulates that everything you create with that software automatically belongs to company A, you wouldn't have much power to bargain here. Either you accept the terms or your don't use the product. That kind of unfair advantags is what such legal restrictions are supposed to prevent.
"Light is faster than sound." - "Is that why people tend to look bright until you hear them speak?"
Please somebody point out the weak link in my reasonning. I can feel it is warped but can't fail it.
And Section 6 of the LGPL states
Considering this, can the BitKeeper license really forbid reverse-engineering, and still be allowed to use the GNU C Library ?
I am but a lowly CS student trying to follow the subtleties of software licenses battles, not a lawyer, and my not being a native english speaker may have made me misunderstand the problem.
> Personally, I think this is such an important issue, and I can't believe people are walking into this
They traded convenience for liberty. And the 'They' is Linus.
I followed the BK thing from the beginning. It is 100% certain that it will clash sometime in the future.
First because RMS is right (you should not use non-free software even if it is 'easier')
Second, because larry changed opinion a lot of time. He clearly cannot be trusted. He just want to use linux as a marketing tool for his product.
Third, because no onelooks at the long time implication. The development process is hugely important (aggraved by the fact that most kernel developers are not even aware of the existence of a process). And now, a commercial entity control the process.
BSD development is so much mature than linux development that it is not even funny anymore (how funny. [The irony is that the BSD folks are supposed to be less free(dom) than linux folks].
As a side point, I think that newscomers to Free software take freedom as granted. They don't recall time where you had to pay to get a compiler, to pay to get an operating system, to pay to get manuals. And sometimes, you were refused the access to the dev tools because you were not a professional. And when the tools were free, you could get them removed from you at anytime (ie: becoming obsolete). I learned to value my freedom. I suspect that most new hobbyist developers take the freedom of the tools as granted, so closed-source but free is free enought.
I only hope that when it'll clash (not *if*, but *when*), people have learned something. This is why I hope for a huge clash (like BK getting bought by microsoft, which is probably Larry business plan).
--fred
You think McVeigh is OK? You sound dangerous...
Anyway, regarding McVoy, he is basically defending some of the least reasonable licensing terms I've seen.
If he expects any respect from the free software community while licensing software under the condition that you're not allowed to make competing software...I don't think he's going to get much.
Whether I wanted to or not, I for one would, out of principle, never use a product whose license said that if I use it, I'm not permitted to write some particular type of software.
I have the freedom to write any kind of software I want to. I may temporarily give up some of that right under the terms of employment, but I sure as hell am not going to accept limits on that right just in order to use somebody else's product.
"Judge, I want to ignore this license which I never agreed to"
The BK license is a click-through EULA like any other. Just like all of them, it's not valid.
Unless your country decides to pass a law making them binding, it won't effect you. (Even if your nation allows electronic contracts, the software vendor would have to take some steps to record the agreement in a valid manner. BitKeeper, like most publishers, does not do that)
If a person permits you to download his software without obtaining an agreement as to how you'll use it, you can do whatever you want (execute the code) so long as you don't violate copyright (make copies of the program, which is unnecessary if you've already got a copy from the publisher)
What makes software so special? Is he also against people getting paid for painting, drawing, making movies, music, or writing books?
All these things, software included, share a fundamental characteristic: they are all forms of art - a personal expression of the author that can be appreciated and enjoyed by others for nothing more than what other people perceive that it adds to their own life.
RMS's argument (as I understand it) is that software is a rigorous description of how to perform some task, and so should not be eligible for copyright.
In support of his argument: you cannot copyright a recipe or a formula. Copyright generally applies to a particular expression of an idea, and not to the idea itself.
In opposition to his argument: nowadays, copyright law explicitly allows computer programs to be copyrighted. But then, RMS is not arguing what the law is, but what it ought to be.
And in case it isn't obvious, he does think that literature, pictures, music, etc. should be eligible for protection by copyright.
Don't forget the youngsters to computers that don't recall a time when software or hardware came with a full, detailed manuals Normally a set of them that required a reinforced bookshelf to hold the weight of them. And detailed ment a lot of source code and schematics. You had to pay for it of course, but you got what you paid for: good stuff. Well not good in all senses, there were bugs in the software, but in general it was better than software from MS in the '90s. (MS has improved quality or so I hear)
I got modded as flamebait, so perhaps I was being a bit abrasive. Let me address this now.
RMS (and alot of his followers, me included) seeks to make proprietary software obsolete and non-existent.
When you make something non-existant, you destroy or remove it. I stated this and you reaffirmed it.
Does mean that noone can charge for software? No, it simply means that all software would be free as in freedom, price is mostly irrelevant.
This is the single biggest lie in the GNU/Free Speech movement. As long as software is "Free" (meaning modifyable or redistributable) it will always be "free-as-in-beer". Always. People don't pay for software that isn't given a redistributal liscence (Check Sharereactor), why will they ever pay for anything that can be given away freely? Lets say we are selling Linux. I sell you a copy for $25 and say it is redistributable. You then give away a million copies to anyone that asks. Am I ever going to sell another copy of Linux? Hell no, people will just get it off you for free. Now I know you're going to mention Redhat or IBM, but that's not the same. Redhat sells technical support - not very helpful if you're a coder with poor people skills. IBM does much the same except they sell complete hardware/software packages - kinda sucks if your program can't be linked to some special hardware.
Does it mean that all software would be forced out into the open for everyones grabbing? No, you're perfectly entitled to write your own private software and not show it to anyone. You're equally entitled to make a private branch of Emacs and do whith it what you will as long as you don't distribute.
Ok, here you've even further validated my point. What if I write software for a living? "Writing my own software and not showing it to anyone" won't put food on my table for very long. The biggest flaw with Stallman is that he hates a system he's never been involved with.
It is an order of magnitude easier to criticize non-free sofware when you have never had to rely on it for your living. Stallman has never had financial survival riding on how many copies of emacs were purchased, or when HURD is released. He makes a living hawking his ideas and getting people to buy into his utopian fantasies and donate money to the FSF which pays his salary. He's like Terry McAullife, he'll make his living peddling the products of other hard work. Terry McAullife will never fight against a Republican congress, but he'll collect a paycheck for telling you how great the Democrats are. Stallman will never make a living of free/Free software, but he'll do just fine telling you how good it is. I'm not bitter that he is able to make a living off the goodwill and donations of others, but he is the exception, not the rule.
How about all of you take a much nicer tilt on this, and ask McVoy (who's already giveing you the software free) his price to GPL bitkeeper.
McVoy is giving the software away for free, as in no cost. Let's not confuse that with Free as in freedom.
As for buying out BitKeeper and GPL'ing it vs. developing a replacement, it's purely a strategic decision, based on -- I think -- two factors.
(1) Which will produce a GPL'ed product that Linus and other developers who like BitKeeper's features as fast as possible? Buying out BitKeeper may or may not be the choice. It might take longer to raise the money to buy out BitKeeper than it would to develop a Free alternative with comparable features that Linus and other's would switch to. There's no reason why both can't be done in parallel. Indeed, the best strategy is to do them in parallel, so that McVoy feels pressure -- once a Free alternative is developed that Linus and other's would switch to, the effort to buy out BitKeeper is off.
(2) Which will be cheaper? Financial considerations are important. It may or may not be cheaper to develop it ourselves.
social sciences can never use experience to verify their statemen
To which McVoy response:
Well, I believe that only your free (as in beer) license says that. And that may very well be upheld by a court, because you're giving it away for no money. However, a similar provision in your for-money BitKeepr license might not be held up. Court's have ruled that -- despite what licenses say -- reverse engineering is acceptable for the purposes of developing products with interoperability. It is highly unlikely that McVoy's license on the pay-for version of BitKeeper would hold up, as that grants no extra rights not given by law, and in fact tries to deny rights to reverse-engineering given by law.
And it is precisely for this reason that RMS wants people not to use BitKeeper. Just as MS has taken measures to lock people into using MS Office, so too can McVoy take measures -- both technological and licensing-wise -- to lock people into using BitKeeper.
social sciences can never use experience to verify their statemen
And apparently, some of them are are so mesmerized by RMS's charisma that they downplay their own contributions.
Source
Sony was suing Connectix (now owned by M$) and Connectix won because the reverse engineering solely for the purpose of interop was considered legit.
IANAL, but one of the issues is that copyrights protect expression but the actual useful ideas are not subject to copyrights but rather patents. So I see no reason to think this is a copyright issue per se.
The license however is one which contains restrictions which *may* be valid under contract law. So this may be an argument of "You agreed not to by using our product and now you are violating your contract and hurting us." This is a completely different issue, and quite frankly I don't know who would win....
LedgerSMB: Open source Accounting/ERP
Dude, quit with the purple prose. Software is not an extension of body or mind. It's some instructions to a CPU. That's very different.
I can type really pretty fast, but the keyboard is not an exension of my body, it's well, a lump of plastic with loads of buttons on.
Software may be a tool like Photoshop, or it may be art in itself, or it may be something really frigging boring like a printer driver.
It's just not special in any way, not matter how much you personally love it.
There is nothing wrong with handing it over to proprietary interests per se. I couldn't care less who writes printer drivers and whether the code is open, I want the frigging printer to work and that requires hardware, firmware, software, mechanics and some guy in the power station making sure the electricity supply is on. Why the hell is the software the ultra-important part of this chain?
-----
He actually got a lot of sympathy on the LKML. One of the problems is that people hate RMS so much that they will rally around whomever RMS is debating, no matter the merits of the argument.
I find it funny that RMS often predicts doom and gloom, people call him an idiot, and a few years later something happens (like this) and he's proven right. RMS said BK was a Faustian bargain. Bingo. Right again, RMS.
The GNU foundation's people? Are these the same people who have a strong dislike of LGPL? Are these the same people who have strong political feelings that all software should be free?
Here's the first paragraph of section 5:
So you define "program" as "source code"? An executable is not a program? Do you think you could convince anybody of this in court?
The only time a program which links to glibc falls under section 6 is in the second paragraph of section 5.
There you have it... In order for section 6 to cover your program, your program must contain portions of the library. Now when you dynamically link a program it could be said that your program now contains portions of the library. Section 6 says:
So if you try and distribute the memory image of an application dynamically linked to LGPL you are right. But only the copyright holder of all applications running in that memory space could authorize distribution to a third party of that memory image.
But paragraph 2 of section 5 is pretty clear... If you do not contain code from the LGPL library, you are free and clear.
Yes, this means you are allowed to reverse-engineer Maya, or any other commercial software that runs on Linux according to the reverse engineering terms in the LGPL.
Oh really? Not even David Turner agrees with that.
I'm sorry, who was the moron armchair idiot?
Education is a better safeguard of liberty than a standing army.
Edward Everett (1794 - 1865)
In general, RMS's position with regard to "free software" (as in freedom) is RARELY off-target. This is because his motivations are simply an outcome of the reasoning that code, like any other form of expression, is crystallized thought, and as thought should be cherished and spread to as many as possible to share it's benefits (and, in some cases, drawbacks). The GPL is one tool that helps ensure the spread of thought in code form. It truly is a disease to think that thought products (affectionately known as "intellectual property") can truly be contained. So why not embrace the fact that we as human beings are tremendously adept at copying, processing and synthesizing thought products.
The other tool at his disposal is a relentless pursuit of the goal that all software should be "free" (again, as in freedom). Software that is not free is always suspect: the "owner" that reserves rights to restrict others in their ability to utilize "their" software is prone to an "absolute power" syndrome. Their intentions may start out noble, but that can change at any moment. Sad to say, but it appears the Mr. McVoy may be on the verge of this. I remember reading the mailing-list archives when this first blew up. Mr. McVoy really has drawn a mental line in the sand that he's not willing to cross. It is unclear if he has the strength of character to let go of his "closed-source" ways. It would really be nice if he did.
Unfortunately, in pursuit of his "free" software goal RMS is likely to piss people off, hence the established rancor at him. Fortunately, RMS has the resolve to see past this and move on. It will be a rocky road to enlightenment for most (it certainly is for me, and I can't even begin to claim any form of enlightenment), but it certainly is liberating.
A replacement for BitKeeper should be developed post-haste, subversion looks promising, but needs many features to completely replace BK. I personally use subversion in my own projects and found it to be quite workable.
Food for though (pun intended): Are your thoughts yours? Do you own your thoughts? If so, how do you exert thought ownership rights? If you never existed would your thoughts exist in others? Now, granted, there clearly are expressions of my thought process that others recognize as "me", but at work, if others were confronted with the problems you work to solve every day what is the likely-hood that others would produce the same or similar solutions as yours? I find it amazing looking at all of the various Web site building technologies out there built by many others how similar many of their solutions are to the ones I produce. The evidence that elements of "my" thoughts are going on in other people's heads oh so clearly demonstrates the fact that "intellectual property" are hollow words. Pursuing the task of squirreling away your thoughts and jealously guarding them is the labor of sick-minded people. Even writing what I writing now is not new. Most of you have read "drivel" like this before, but it is still worth mentioning. Just think about it.
The only excuse some individuals have for not listening to Richard is personal prejudice against Richard's way of life.
Yes, I am prejudiced against a man supported by a MacArthur Foundation grant and the facilities of MIT who devotes his life to attempts to thwart the efforts of those who invest their own money in software development.
Perhaps you are not familiar with a story from the early days of the MIT AI Lab? A group of RMS' colleagues left to set up a company. Everytime they released a new feature in their software, RMS, from the financial safety of MIT, reverse engineered it and released a free version. He drove them out of business, for no other reason than resentment that they had quit the lab.
The reliability issues with subversion are not so much repository corruption issues as program failures -- I've had it crash on me or refuse to touch my working copy more times than I'd like to admit. In any event, however, the design issues are larger ones; see Lord's analysis on arch-users for that -- to make a strong argument I'd be simply parroting his arguments.
darcs has a very nifty merge algorithm -- but it's not practical for immediate use, and not nearly so actively maintained as arch, which has a substantial active user base, constant maintenance, a nontrivial suite of 3rd-party tools available (see cscvs for converting CVS archives to arch or tla format, perspective for a nifty web-based repository viewer, and assorted other stuff as well); further, in its tla incarnation, arch is far faster. I wish darcs well, and hope to see its merge algorithm integrated into arch (and indeed, I understand that Tom is doing some work to make those pluggable, and I understand that Robert Collins may end up working on some new ones taking advantage of said pluggability) -- but don't see it as much as a serious revision control tool as a very nifty proof-of-concept of a damn cool merging algorithm.