Slashdot is powered by your submissions, so send in your scoop


Forgot your password?
Programming IT Technology

Tridgell Reveals Bitkeeper Secrets 373

wallykeyster writes "The Register is reporting on Andrew Tridgell publicly demonstrating how to interoperate with Bitkeeper. During his keynote at the Linux.Conf.Au, Tridgell connected to a BitKeeper site via telnet and used the mostly forgotten "help" tool. Ethical arguments of aside, what really counts as reverse engineering anyway?"
This discussion has been archived. No new comments can be posted.

Tridgell Reveals Bitkeeper Secrets

Comments Filter:
  • by bethane ( 686358 ) on Thursday April 21, 2005 @03:01PM (#12304886) Homepage Journal
    Well, I think it's safe to say that Linus Torvalds is wasting his time on his new RCS, 'git'. He may as well just go ahead and write a BitKeeper-compatible system, since he liked BK so much. Oh, wait. That's morally "wrong". So says the guy working on a clone of the UNIX operating system. Something doesn't quite add up here.
  • Recycled Comment (Score:5, Interesting)

    by geomon ( 78680 ) on Thursday April 21, 2005 @03:15PM (#12305044) Homepage Journal
    I've read a few exchanges from the /. crowd, read a few statements by Linus and the gang, have read McVoy's interpretation of the BK saga, and have come to one conclusion:

    No one but the three people involved in this fiasco *really* knows what happened to get this situation to the stage where people begin a verbal free-fire in public.

    McVoy is a business man; true to his heart, he needs to keep the BK user strung out on his code. Hell, I would feel the same sense of outrage that he feels if someone threatened to kill my cash cow. Don't pretend that every one you wouldn't feel the same way if it was *your* revenue stream. To me, anyone who claims an absolute vow of poverty is looking for a monastery to live in. Everyone I know would fight to protect a source of financial income.

    Selfish? You bet. But nature has created more selfish beings than egalitarian ones. Nature favors pragmatism.

    But McVoy could have let this one ride a bit more. It is just a matter of time before someone cracks his model. Then he will have to play the same game as Microsoft and Adobe only on a different level. Too bad for him, though, that his inexpensive advertising scheme didn't last. That is another little detail that goes relatively "un-remarked" upon in the various forums I've read. Larry had one of the hottest programmers in FOSS using his SCM. In fact, this Man Of The Year lavished all kinds of praise on his progeny! You would have to pay more than the "free" license fee for that kind of advertising. Shit, probably A LOT more. If Linus had been paid for his endorsements, that could have added up to quite a sum of money. Larry has wisely kept those funds securely in his pocket.

    Again, I'd do that too. The monks of this world can keep their vows.

    Linus? Well, it was kind of hard to turn down a free license for one of the best SCMs on the market. If I had been in his position, I would have grabbed the product and ran. In fact, I would like to personally thank Larry for helping juice the Linux kernel development. I know SCO has been rummaging around in the Linux closet for evidence that it was their intellectual property that made the kernel advance so quickly. I believe that Larry's BK contribution probably made the significant increase in kernel production possible. Judging from Linus' angst and outrage, I think he believes that too.

    But Linus is being a bit thin skinned. Does he believe he is the ONLY programmer that has been burned by relying on a proprietary product for their work? Didn't he listen to all the people who had been telling him about *their* bad experiences with proprietary lock-in? From what I've read in the past, they had plenty of legitimate worries that this was going to happen. I'm sure that Linus knew it would happen someday too. He's just pissed that it happened NOW as opposed to LATER.

    Boo hoo, get over it, this too will pass, etc. But why attack Tridgell in public? Hmmm.... That does raise some interesting questions. And why get all bitchy about it?

    There is something we are not getting in this little soap opera. Tridgell is silent, probably for good reason. But why would Linus take him to task knowing that he would not be able to respond publicly?

    And Perens? This is a slugfest that only Gates, Darl, and RMS would love - all for differing reasons. Why does Perens feel compelled to call out Linus over his treatment of Tridgell?

    I thought the points made by some posters about just how Tridgell was sniffing packets to see the metadata protocols is extremely insightful. To have BK protocols running on his network would require that he be operating a client and server somewhere where he could see it, no? What network was he sniffing if he didn't have a license?

    What amazes me is that the attempt to get BK's protocols didn't happen *sooner*. With all of the pissing and moaning that erupted when Linus started using BK, I would have thought there would have been someone doing what Tridgell was accomplishing years
  • by FidelCatsro ( 861135 ) <[moc.liamg] [ta] [orstacledif]> on Thursday April 21, 2005 @03:21PM (#12305118) Journal
    It would be moraly wrong if he himself had done it after promising not to , but he didn't and there is no law i know of that restricts reverse enginering(not that companys don't try with some though).

    for example Wine would be a moral linux no-no , samba also , many many other things including as parent said, the kernel.

    What this ammounts to is Linus saying "linux is immoral" ... I respect the fact that we all make mistakes , but forgivness comes after an apoligy.
  • by Anonymous Coward on Thursday April 21, 2005 @03:23PM (#12305137)
    Seems the Darcs guys are talking to Linus [].

    The distributed tools are way more powerful than the centralized systems; so I think it's great to see the Darcs and Monotone groups both interested in the (probably much more performant) Git backend.

    (PS: yeah, I know about Arch, but damn that thing's confusing. I'm guessing they borrowed the usability team from clearcase. If you like Arch, it's definately worth checking out Monotone or Darcs. (personally I lean to Darcs because of the cool language it's written in; but like monotone as well)

  • by Elshar ( 232380 ) <> on Thursday April 21, 2005 @03:27PM (#12305185) Journal

    He's not working on a clone of the unix os. He's working on a kernel that in all actuality is nothing at all like unix. The gnu tools that people use with it make it feel like its kinda sorta unix-ish though. Sorta. :)

  • by geomon ( 78680 ) on Thursday April 21, 2005 @03:28PM (#12305195) Homepage Journal
    Not if you were invited.

    Scenario: Bob is forced to buy a client for a SCM he doesn't like. Bob invites Ted to come over to his house and poke around on the client. Bob has permission to use the client AND interact with the server. Ted is looking at the server from the client that his friend purchased.

    I don't think that would be something that could be construed as "illegal". It might be "actionable" in a civil tort sense.

    That might be why Tridgell is keeping quiet.
  • Re:Recycled Comment (Score:5, Interesting)

    by swv3752 ( 187722 ) <swv3752&hotmail,com> on Thursday April 21, 2005 @03:40PM (#12305332) Homepage Journal
    McVoy and Torvalds are friends. It has been posted earlier that McVoy went to Torvald's home to pitch the use of BitKeeper.

    Linus speaking out against Tridge, is simply that of someone backing up his friend. Unfortunately for Linus, it makes him a hypocritical git.

  • by Bamafan77 ( 565893 ) on Thursday April 21, 2005 @04:15PM (#12305887)
    "I hope he has something more substantial to back himself up than a weak joke."

    What's suprising is how many people actually believe that's all Tridge had to do reverse engineer the BK protocol. While I'm willing to buy that he didn't need a BK client, the demo is obviously at best a trivial first step.

    Unlike most people, I can actually respect McVoy's decision to remove the free client (though not necessarily in the angry way he did it). The SAMBA and BK situations aren't exactly identical. Tridge's reverse engineering for SAMBA is not *that* big a deal to MS. So what if a Windows server gets fooled into thinking that some Linux or VMS box is a Windows machine? While this service is immeasurable to many of us, we represent a small part of MS's customer base. It's unlikely that such a thing will enable anyone to budge MS in it's golden goose OS or office productivity markets.

    But I think the situation is a little different with the BitMover guys. It's probably not THAT incredulous to imagine someone coming up with a free BK client that is better than the free version, and at least competitive with the paid-for version. If such a client is released, then no one would have a reason to buy BitMover's non-free client, thus putting a dent in BitMover's income. After all, it's not like BitKeeper has the channels to force their product down people's throats while threatening vendors who dare to sale someone else's product. They don't have the power to articially manipulate the "free" market for their benefit.

    So, IMHO, the difference is this: The Free Open Source development community doesn't have the resources to affect a goliath like MS in any significant way. But a smallfry like BitMover? The FOSS bandwidth is there to bring a company like this to its knees and McVoy knows it. No doubt he's reaped many benefits from the free BK client and his company's association with Linux, but now the other shoe has officially dropped...

    Now, I'm not arguing that BitMover doesn't deserve such a fate or that I'm siding with them. It is an open market after all and may the best man win. However, I can at least understand why McVoy and crew would be threatened by a free product competing with their non-free product. Yes Tridge building a new client *does* release Linux source from propietary SCM lock-in which is good for the *rest* of us. But let's at least admit that it's also a valid economicthreat to BitMover as well. Again, why buy their non-free client, if I can get a good enough free client off sourceforge? On a purely economic and pragmatic basis, both sides can be right.

    For the record, I don't think Tridge is in the wrong and I don't think he's "out to get" BitMover or McVoy. However, I think an unintended side-effect of his development could be the downfall of BitMover.

  • by Anonymous Coward on Thursday April 21, 2005 @04:21PM (#12305972)
    Hasn't anyone told Linus that the name git is already taken? Hasn't anyone notified the GNU Project?

    git (GNU Interactive Tools) is a screen-based console "filer" with command line and extreme flexibility and key mapping, etc.

    Freshmeat: tivetools/ []
    Homepage: []
    GNU Page: []

  • by DA-MAN ( 17442 ) on Thursday April 21, 2005 @04:27PM (#12306061) Homepage
    Amazingly, yes. Apparently, the original license did not specify a time frame, so it could be interpreted as valid for undetermined time. The revised license seems to specify 1 year as the no-compete clause.

    But Larry McVoy revoked the license. Doesn't that mean that the no-compete clause has been revoked with the rest of the license?
  • Re:The Linux Life? (Score:3, Interesting)

    by JoeBuck ( 7947 ) on Thursday April 21, 2005 @05:45PM (#12307274) Homepage
    Yes, person 1 (Linus) liked a tool. But person 2 (Tridge)'s actions (basically stumbling on a way of getting the data out of BitKeeper) should not have led person 1 to lose rights to the tool. It was person 3 (Larry)'s decision to demand that either OSDL fire Tridge or, Linus quit OSDL, or Linus lose the right to use BitKeeper. This ultimatum was completely unnecessary; he thought he could bully OSDL. It didn't work.

    But the breakup is at a good time. Larry's managed to bootstrap BitMover into a viable company, and the world knows what BitKeeper is without a massive advertising/PR budget. He'll do OK. The Linux kernel will soon lose its dependency on a proprietary tool, and there's been a renaissance in version control systems as many groups have come up with new and promising systems. These systems aren't clones of BitKeeper; there are a variety of approaches.

  • by DarkOx ( 621550 ) on Thursday April 21, 2005 @08:19PM (#12308889) Journal
    If its morally wrong to reverse enginerr you'd better stop speaking. Figure that you could not possibly understand the spoken word when you were first learning to speak. Your infintile brain would have been forced to study the protocol and make some guesses, then try some stuff produce some sounds that it thought had some meaning and see how the people who can speak respond. Reverse engineering is part of human nature. I would bet most of us learn more by what is more akin to reverse engineering then, serialized learning. Could really ever learn anything from a text book if you could not turn around and at least look at the broader contexts of its application that is, how it interacts with the larger system.

Reactor error - core dumped!