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."
What RMS is doing his best to ignore is that these restrictions are lifted if you (gasp!) buy a commercial license.
I realise what point he is trying to make, but I think it is unfair of him to cloud the issue like that.
(Please browse at -1 to read this comment.)
It seems reasonable enough that Larry would want to prohibit people from using bitkeeper to compete against bitkeeper.
However I think it is telling that the license goes a step further and disallows any person or entity who ever works on a competitor from ever using bitkeeper. So Larry is essentially helping to see that many people (Linux kernel hackers using bitkeeper) are unable to ever compete with him, even if the kernel hacking and open-source-SCM hacking are in no way related. Way to drive a wedge through the free software community.
As usual, RMS gets little or no respect around here, despite the fact that, as usual, he's right.
/. to listen more to RMS when he speaks and suppress your obvious desire to bash a man because he has a certain set of ideals.
Those of you saying that the restrictions RMS mentions would be lifted if you bought the commercial version are missing the point. The point by RMS is that all of the licenses under which you can use BitKeeper are draconian, as they're EULA's. The problem with EULA's is that they can be changed at any time by the developer, thus creating an unfair situation; BitKeeper could just as easily include such restrictions on its paid-for version. The other problem is that accepting them is mandatory, thus creating another power imbalance.
That said, this is all the more reason for developers to switch from BitKeeper to alternatives. BitKeeper can impose any draconian restrictions on you they wish, and you'd best not wait until you're trapped into using BitKeeper and dependant on it to change.
I'd advise the rest of
social sciences can never use experience to verify their statemen
Firstly, this isn't your project. Mr. Torvalds has made his points and position quite clear, and it's time that you and the rest of the Free Software people leave the kernel hackers well enough alone.
Also, do you have no respect what's so ever? What are you doing posting on the LKML, which is not meant to be political.
Also, it would be nice if you would get your facts straight. Bitkeeper (the gratis version, anyway) only restricts you from using it to develop a competing project, not from using one.
// file: mice.h
#include "frickin_lasers.h"
According to RMS: If you even run bitkeeper, you can't contribute to CVS or other competetors.
That seems to be quite a restriction. Imagine a Microsoft EULA that says: if you run Windows, you can't contribute to Linux.
RMS has a point. Licenses like these are there to kill free software alternatives.
Goddamn, but what has happened to slashdot? Judging by the posts from the majority of the slashdot crowd, I think that they'd be happier if slashdot started reporting every new Microsoft Update patch instead of new Kernal releases.
We will never know the answer to this puzzler because he is the only person in the world to get slashdot headlines by posting flamebait to the wrong news groups. Anyone else would either be ignored, flamed for cross-posting, or deleted by the moderators.
I do not fail; I succeed at finding out what does not work.
Normally I don't mind RMS spouting off about something when he has a decent leg to stand on or is using his own forum. In this case, he really doesn't.
First, he didn't seem to choose the right forum to speak in. A listserv for kernel development is not the correct space to bring political speach into. RMS's post was very possibly off topic to the list.
Second Linux is not his project, and he is not managing it. Torvalds has expressed his opinions on the Free Software movement. He doesn't believe in Free Software as an all important political idea, thus he has not don anything wrong by using Bitkeeper. Torvalds chose Bitkeeper, and that's what the project uses. Period.
RMS should attempt to open a serious technical discussion directly with Torvalds. RMS should say "What do you need?" and then deliver it. Or RMS should violate the license in a clearly absurd manner and let Bitkeeper take him to court to test the validity of the license.
I understand why it's a slap in the face, too; but comparing software licensing to slavery is a bit overboard. Software is not sentient by any stretch of the imagination. Further, no one is restricting your rights toward software development - only your rights toward the use of their system.
A far better analogy is to compare a software licensing agreement to a lease--some landlords are perfectly respectable, while others are more than a little shifty. In any event, as long as you live on their property, you abide by the terms of the lease.
Unfortunately, software manufacturers don't require notice before changing the terms of their licensing agreements. I think they should - and I think that people who purchased software licenses have a right not to have the license changed on them arbitrarily. That sort of fly-by-night treatment is my principal objection to BitKeeper's practices: it's unethical. And unethical behavior is not limited in practice to companies which distribute proprietary software.
Their treatment of their own customers deserves a response - and the best response is to cease doing business with them. That couldn't work with slavery, which was far too widespread. . . but when you're dealing with a single corporation, a little bit of financial pressure goes a long way.
!#@%*)anks for hanging up the phone, dear.
RMS is certainly entitled to his opinion; I don't think that is the issue here. The issue here is his choice of forum. The Linux Kernel Dev list is for technical discussion related to the Linux kernel. If RMS has issues with the bitkeeper license he should post his opinions somewhere else, like the GNU website or Usenet. It doesn't matter who you are or what you've accomplised, offtopic is offtopic.
If you read the entire story, not just the slashdot post. It looks like Bitkeeper is getting a little microsoftian about what type of project they 'allow' you to use with their product.
I am not a huge RMS fan, especially with all of his "it is a GNU/World!!" assertions, but i think he has a good point this time.
Maybe RMS should not cry wolf so often; so we might learn to listen.....
Ghandi had a great point
First they ignore you...
Second they laugh at you...
Then they fight you...
And then you win!
There are plenty of open source systems for version control and configuration management. Furthermore, they way open source works, if you need an unusual tool for your project, you create it yourself and share it.
It's funny how much people will bitch when they're not the ones that have to deal with the inadequacies of $OSS_SM_TOOL when it comes to kernel development.
There are plenty of huge open source projects, and they work fine with CVS. GNU Hurd is being developed with CVS. BSD is. To me, the real question is: what is going wrong with Linux kernel development that CVS is not sufficient?
You obviously don't know Larry McVoy.
I first met Larry McVoy on usenet. He was highly cogent in his arguments, and generally a very intelligent guy. Unlike a lot of the idiots who came in after the NSF quit running the Internet, you could actually hold a technical discussion with him, after which you would have a solution, or an approach to a solution, for the problem at hand.
I've talked with him on the telephone on several occasions, when issues have come up that merited a telephone conversation; I've called him, and he's called me, though it has been a while since our last voice conversation.
When Larry McVoy left Sun, he wanted to take the SunOS 4.1.3_U1 code (U2 has not yet come out), and release it under the GPL. This was quite visionary, given the amount of competition that Linux is now giving Sun, even internally, within their own engineering staff, these days. Sun would not do the release, because it would cannibalize their SVR4-derived "Solaris" market.
Larry's motivations in this case were, I think, base... in that he wanted to "rescue" the important work which had been done on the BSD dervice Solaris (SunOS) code base. He saw the GPL as a way to do that.
Larry was an early GPL advocate, in this sense. Frankly, I'm glad he failed in this endeavor; it wporbably would have meant the end of BSD derived OSs, which generally exist only because the GPL is too draconian for people who need to do business.
Larry became an outspoken Linux advocate; he authored the "lmbench" suite of micro-benchmarks, all of which show Linux in a good light, compared to its competition. One can argue that these tools drove a number of the important design decisions in the Linux kernel itself, which, among other things, led to the current threading model and code, which *depends* upon the fact that process context switch overhead is minimal, and there is very little difference between it and thread context switch overhead.
Larry advocated Open Source software, in general: BitKeeper, by it's nature, *from the beginning*, offered free licenses those people who woul publish their source archive, as the cost of the license.
Thus, by its nature, BitKeeper encourages free software by providing economic incentive.
But, like the GPL itself, it is an instrumentality, and the instrumentality must not obey the same rules as that on which it acts.
The GPL carries a prohibition against modification: it is not itself under the GPL. Ask yourself "Why?".
For this same reason, good or bad, BitKeeper can not itself be Open Source software. Yes, there are economic issues. Despite people's intentional misinterpretation of the word "support" in Larry's statements to mean "technical support" rather than "economic support", Larry's correct: the Open Source model is not economically self supporting for stategic projects... it only supports itself for tactical projects.
That RMS is complaining about BitKeeper now is, I think, sour grapes. That's the kind interpretation. The unkind interpreation is that BitKeeper is a more effective mechanism than the GPL itself for achiving the goals of the GNU Manifesto, of which the GPL is an instrumentality.
So before you call Larry an idiot, or blindly GPL or even BSDL your next set of source code, understand the long term consequences of the license.
Frankly, I'm glad he's let go of the understandable bitterness that comes from pouring your soul into a product, only to have it hidden away in a vault by an employer with goals other than advancement of the art and science of computer science.
I think this license demonstrates that he's come to his senses, on strategic issues -- a painful lesson. Would that RMS would so the same.
Thanks,
-- Terry
It already did. If you are on a campus where anyone (not you - anyone) is working on patches to CVS you are not allowed to use Bitkeeper in the same way I am. What's not arbitary about that?
TWW
"Encyclopedia" is to "Wikipedia" what "Library" is to "Some people at a bus stop"
Yet, thats where Larry posts liscense issues. Thats where the discussion was taking place. Who are you to tell them where they can discuss kernel issues?
His point is that Free Software is better because it gives freedom to its users: freedom to use it as they choose, to understand how it works, and to modify it to fit their needs.
Question: is the average Word user made more or less free by having the source code to Word?
proprietary software is so bad- it doesn't respect the needs of its users
On the contrary, proprietary software must respect the needs of its users, otherwise they won't buy it. There are no such incentives for free software, which doesn't have to respect the needs of anyone but its author.
This is not FUD and it is not panic. It's a problem and a serious one.
The fact that RMS is frequently one-sided is a fact and I agree with partially with you on the case of single criteria. However you should note that a larger part of the people here is worried not about this but on the fact that someone is being more papist than the Pope. And it's not RMS but BitKeeper. These license policies may lead to the fact that, in one point of time, BK may hinder a lot the development of Linux. Note that Linux is not ONE product for ONE objective on ONE SINGLE environment. Inside of it, along with it and beyond it there are TONS of programs, applications, drivers and other stuff which depend on Linux and which Linux depends of. To understand how BK may hinder this, try to get a deeper look at their license.
This is a license directly against best value. It is a binding that forces people into conditions where they may be unable to find that same best value. The license is even anti-commercial as it forbids people to sell third-products, that may have nothing to do with BK, except some similarity on functionality. No matter the qualities of BK, such term is enough to put it more dangerous than M$.
If BK is sincere on being a good company, willing to receive a direct reward, they should choose three ways - turn the license into a genuine commercial license, make a license in terms near to BSD,or separate it into components with different licensings. Probably this would hinder kernel developers, but there is a problem on playing with half-agreements, not seeing dropped nets, accepting broad middle-terms and forgetting about consequences. In one way or the other this may one day turn into the bad corner. Much like M$ did since its advent, let's not forget that 10 years ago BG was Luke against the Empire. Frankly I would not like to see Torvalds being compared to Dart Vader...
Consider that the open source software often chases the coattails of proprietary software, and it is like an erosive force against a software monopoly. Rather than let a given company build an invincible fortress of refined, polished, peerless software, they are constantly forced to accept that their current innovations will eventually become basically public domain through free software. This is an incredible incentive to keep in touch with their market, make real and substantial improvements to their product, and avoid heavy handed dictator style behavior.
If closed-source proprietary software blazes the trail, open source paves the highway. Making a practical public domain out of so much software ensures that innovation in proprietary software is a process, not an end point. It's competition, it's checks and balances, and it benefits everyone who uses the software.
Open source does hardly any damage to commercial products. It does ensure that the #1 commercial product has a competitor close on its heels that cannot be driven away not by competitive pricing but by smart business and new inventions. I wouldn't argue against having competition around here, I get the feeling that Microsoft isn't loved. At any rate, this is a very sustainable situation and the checks and balances benefit the users substantially.