The BitKeeper Ghost

Written by: Shlomi Fish
Date: 17-October-2005

BitKeeper is a commercial closed-source version control system that was once available under a gratis license for use by open-source projects. After he was criticized for a slow patch merging speed, Linus Torvalds, the originator and project head of the Linux kernel, decided to use it as the primary version control system for the Linux kernel. Several other developers (but not all) joined him in making use of this system.

BitKeeper accelerated the speed of kernel development considerably. However, in April, 2005 BitKeeper's parent company decided to end-of-life the gratis version, which left the Linux kernel developers and developers of other projects in Limbo. The kernel developers tested several version control systems, and ended up developing their own tool called git, which has since then undergone some active development. Many other open-source alternatives to BitKeeper have been developed before it or afterwards.

On 30 September 2005, Bryan O'Sullivan, a developer of Mercurial, a version control system similar to git, published a note to the Mercurial mailing list where he explained why he no longer could be working on Mercurial:

As I mentioned the other day, I will not be contributing to Mercurial development for a while. Several people have asked me why.

At my workplace, we use a commercial SCM tool called BitKeeper to manage a number of source trees. Last week, Larry McVoy (the CEO of BitMover, which produces BitKeeper) contacted my company's management.

Larry expressed concern that I might be moving BitKeeper technology into Mercurial. In a phone conversation that followed, I told Larry that of course I hadn't done so.

However, Larry conveyed his very legitimate worry that a fast, stable open source project such as Mercurial poses a threat to his business, and that he considered it "unacceptable" that an employee of a customer should work on a free project that he sees as competing.

To avoid any possible perception of conflict, I have volunteered to Larry that as long as I continue to use the commercial version of BitKeeper, I will not contribute to the development of Mercurial.

As such, Mercurial can stand entirely on its own merits in comparison to BitKeeper. This, I am sure, is a situation that we would all prefer.

Following is my analysis and conclusions of this latest development.

The Issue

First of all, I'd like to note that I'm not worried about the future of Mercurial due to O'Sullivan's leave of absence. Mercurial has many other active users and developers. So do many other open-source version control systems. That's not the issue here.

The real issue is whether Larry McVoy, the CEO of the BitKeeper parent company, can and should prevent the employees of his customers' from contributing to competing version control system. Unfortunately for McVoy, the Law is not on his side. Someone who bought the product, is allowed to contribute to a competing product under the laws of most countries. While it may be marginally legal to impose a non-compete clause for a gratis version of a program, a paying customer should be allowed to do so. (Including a company that makes a competing product).

Judging by what O'Sullivan said, though, Larry McVoy did not imply that he was obliged to do so, and O'Sullivan did it out of good will. Still, the question still remains: why should BitMover care that employees of its customers are contributing to competing products?

Paranoia

The answer is simple: Paranoia. Larry McVoy is positively paranoid and possessive about BitKeeper. He believes that it is the best product in the market, and that in order to stay ahead others must not be able to borrow features or ideas from it. During the entire course of BitKeeper history, there were more and more measures taken to ensure that people would not abuse the license, or attempt to compete with BitKeeper. I wrote about it earlier.

As we all know, such paranoia only brings a lot of frustrations, and doesn't actually help you to keep ahead. BitMover should try to stay ahead by making sure the development of BitKeeper is rapid enough for BitKeeper to become better and better, rather than harass its customers and competitors.

So What Should I Do About It?

BitKeeper is, from experience, a fine piece of engineering. However, the CEO of its parent company has too much bad attitude, and doesn't understand the value of a free and open competition. While I respect him as an engineer and as an entrepreneur, his personality and behaviour are extremely annoying and are doing a lot of harm, both to him and others. Furthermore, BitMover is practically entirely his entity, and as such is entirely affected with his attitude.

As such, I believe any self-respecting businessman should not buy his products. There are enough other alternatives around, both open-source and commercial, and some of them are superior to BitKeeper in many respects. BitKeeper has its share of short-comings, despite what BitMover would like you to believe. If you are already using BitKeeper, you are hereby being recommend to switch to a different version control system.

BitKeeper was probably never meant to have any long-term significance on the version control world, at least not with the personality that its lead developer and CEO of its parent company possess. We may see BitKeeper concepts and ideas in other version control systems, but we can hope that BitKeeper would be dead, and no longer actively used.

A completely alternative route is for BitKeeper to become open source or something very close. This is a decision only Larry McVoy can make. I wish he would do just that, but it seems unlikely at this point.

Links

• LWN Coverage of Bryan O'Sullivan's Email.