Weekly Edition
Return to the Press page
| |||||||||||||
Tridgell speaks out in BitKeeper war (ZDNet)
ZDNet
looks into a somewhat exaggerated
"war" between Andrew Tridgell and Linus Torvalds.
"Andrew Tridgell has made his first public comments on the dispute between himself and Linux originator Linus Torvalds over source code management for the Linux kernel, describing much of the coverage and commentary on the issue as "trivial and crazy"."
(Log in to post comments)
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 21, 2005 15:49 UTC (Thu) by gowen (guest, #23914) [Link] "We have now, in a single line of shell, implemented a BitKeeper client,"Now it's apparent that Tridge's "reverse engineering" consisted of telnetting to a BK server and typing "help" ... why ... it's almost as if Larry was simply looking for an excuse to stop supporting free BK clients. Incidentally, is SCM metadata a derivative work of the source code itself? If so, since BitMover are distributing the Linux source and the metadata to licensed BK users, shouldn't they also make that metadata available in a readily accesible format, under the terms of the GPL?
Linux kernel metadata under the GPL? Posted Apr 22, 2005 6:28 UTC (Fri) by Duncan (guest, #6647) [Link] Incidentally, is SCM metadata a derivative work of the source code itself? If so, since BitMover are distributing the Linux source and the metadata to licensed BK users, shouldn't they also make that metadata available in a readily accesible format, under the terms of the GPL?I don't believe they'd have to unless they distributed the executable binaries as well. Documentation, which is what the metadata effectively is, is an entirely different ball game, as they say, and the GPL wasn't designed to cover it. IMO from what I've read, IANAL and all that. Duncan
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 21, 2005 16:06 UTC (Thu) by huaz (guest, #10168) [Link] So, Linus Torvalds WAS an idiot in this case, as Bruce Perens put it.
I think he owns a big apology to Tridge.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 21, 2005 16:39 UTC (Thu) by aseigo (guest, #18394) [Link] from time to time people have disagreements, doubly so for people who arepassionate, brilliant and under constant scrutiny due to the "largeness" of their presence in a community. no surprises there. it's hard to know the intentions or meanings behind these exchanges unless you know the people involved, the events under discussion and the cultural context within which it all went down. few of us reading these articles fall into that group. at the end of the day, i doubt Linus or Tridge will be losing any sleep over this or so much as scowl at each other next time they pass in the hallway. it's simply not that important. what is important is that the kernel, and for that matter samba, as two very cool and valuable projects continue to move forward in a way that is satisfactory for those involved. they seem to be doing so. the rest is just a soap opera of the media's stirring and not something worth spending time and energy getting worked up over. i'm really not sure why Bruce felt compelled to comment on the whole affair since it doesn't seem he was involved in the least and his comments didn't have much constructive application. if i had to guess i'd say that Bruce probably was asked by a reporter to comment on the situation as an aside and then forgot to be careful as one always should be when discussing matters with the press. in any case, the sooner we learn to laugh about this whole thing and move on to the cool technology these people are working on, the better.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 21, 2005 17:41 UTC (Thu) by StevenCole (guest, #3068) [Link] Very well said. I think you're quite right about the the rest is just a soap opera of the media's stirring part.Here's another angle:
P.T. Barnum: "There's no such thing as bad publicity." Please also keep in mind the possibility that Larry was backed into a corner on this one, and not one of his own making, but of the U.S. legal system. If he had let this go, he might have had trouble defending against future attempts to purloin his proprietary software. Also please remember the tremendous progress made in the past three years, much of which can be credited to Linus' use of BitKeeper. I hope everyone will just "get over it", and move on.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 21, 2005 23:31 UTC (Thu) by bignose (subscriber, #40) [Link] Please also keep in mind the possibility that Larry was backed into a corner on this one, and not one of his own making, but of the U.S. legal system. If he had let this go, he might have had trouble defending against future attempts to purloin his proprietary software.You've just contradicted yourself. If one finds oneself unable to continue locking people into non-free software, that is a corner of one's own making.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 22, 2005 0:22 UTC (Fri) by StevenCole (guest, #3068) [Link] You've just contradicted yourself. If one finds oneself unable to continue locking people into non-free software, that is a corner of one's own making.If any contradiction exists, it's between your view and the facts. Please explain how to reconcile your view of Larry "locking people into non-free software" with the actual fact of the BK to CVS gateway which his company provided, and the free (no whining license) open source BK client which was offered near the end of the BK era. If Microsoft were to take "locking people into non-free software" lessons from Larry, they would have to add Open Office native format (.sxw) converters to Word. (I hope they have to do that some day, regardless)
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 22, 2005 14:33 UTC (Fri) by man_ls (subscriber, #15091) [Link] I rather see it as a classical embrace-and-extend situation: you can use your client of choice for basic access, but you only get full functionality with my proprietary version. Someone more knowledgeable should answer this anyway.If Microsoft were to take "locking people into non-free software" lessons from Larry, they would have to add Open Office native format (.sxw) converters to Word. (I hope they have to do that some day, regardless)Obviously, Microsoft can teach Larry and everyone else a few lessons in how to play embrace-and-extend. Once upon a time, the import capabilities of Word were probably the best among word processors, and that was a decissive factor among many managers to standardize upon it. Of course, it gave read-only access to foreign documents, so once a document entered the Word treadmill there was no way out. Nowadays they seem to be too arrogant to care about OpenOffice.org, not a strange thing since they have a near monopoly, and anyhow its import and export filters are so good that there is rarely any need for that.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 22, 2005 21:08 UTC (Fri) by sbergman27 (guest, #10767) [Link] > Please also keep in mind the possibility that Larry was backed into a corner on this one, and not one of his own making, but of the U.S. legal system. If he had let this go, he might have had trouble defending against future attempts to purloin his proprietary software.
IANAL, but I believe you are confusing U.S. trademark law with... something else. (Copyright? Pattent? Contract? Or whatever covers licesnes?)
As far as I know, it is only trademark law that takes a "defend it or lose it" stand. Not sure how it works with licenses, But with patents, for example, you can stand and watch your patent rights get violated till the cows come home, let lots of companies become dependent on those violations, and then move in for the kill. And the fact that you stood by and watched for years makes not a whit of difference.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 23, 2005 0:32 UTC (Sat) by StevenCole (guest, #3068) [Link] IANAL, but I believe you are confusing U.S. trademark law with... something else.I am also not a lawyer, so it's entirely possible you are correct. This might be a good question to ask Pamela Jones of Groklaw. Maybe Marbux (who is a lawyer) or someone similar can provide an expert opinion. As far as software patents go, that's a subject that a far greater majority of us can agree upon.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 21, 2005 17:44 UTC (Thu) by error27 (subscriber, #8346) [Link] > at the end of the day, i doubt Linus or Tridge will be losing any sleep> over this or so much as scowl at each other next time they pass in the > hallway.
I think Tridge is in .au so probably they don't pass in the hallway all that often.
But yeah, the register needs to hire fewer drama queens to write for them. Or actually, they shouldn't because it's a lot of fun to read. But you can see where I'm going with this.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 21, 2005 18:32 UTC (Thu) by huaz (guest, #10168) [Link] In fact, I find there have been amazingly TOO FEW journalist articles about this issue, in which Linus himself was not on the totally bright side.
So I like theregister's way of making the bold comments and criticizing the King who couldn't be wrong.
And I can certainly understand why some people want to get over this as quickly as possible.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 21, 2005 18:53 UTC (Thu) by smitty_one_each (subscriber, #28989) [Link] There is a fine line between <i>making bold comments</i> and blowing a lot of smoke. When I navigate to The Onion, I know I have entered the realm of fiction. The Register seems unclear on the distinction at times.
we all love The Register Posted Apr 22, 2005 0:00 UTC (Fri) by xoddam (subscriber, #2322) [Link] That's what makes the Reg the Reg. The masthead reads "Biting the hand that feeds IT", and in biting Torvalds' hand it's in good company.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 21, 2005 21:49 UTC (Thu) by dlang (✭ supporter ✭, #313) [Link] if that was all he did it seems rather strange that it would take him a couple weeks to do this.
remember that the other side of the story was that after they found him pokeing at bitkeeper there were several weeks of negotiations, during which bitkeeper thought that there was an agreement to halt work, and Larry blew up when he found that Tridge was continueing to work on his client.
this doesn't jive with 'it's just a single line of code' or 'asking for help from bk'
I suspect that what tridge is saying is how he got started, but I'm also pretty confident that he went far beyond simply fetching things and started trying to figure out how to put things in, this is the part that would trigger alarms (from bad data, etc) and raise hackles.
in other words I don't believe it's nearly as simple and trivial as tridge is makeing it sound.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 21, 2005 23:49 UTC (Thu) by bojan (subscriber, #14302) [Link] > if that was all he did it seems rather strange that it would take him a couple weeks to do this.
No, it didn't take him a couple of weeks to figure out the telnet bit. He was trying to figure out the format of the metadata that he just checked out with "clone" command.
> remember that the other side of the story was that after they found him pokeing at bitkeeper there were several weeks of negotiations, during which bitkeeper thought that there was an agreement to halt work, and Larry blew up when he found that Tridge was continueing to work on his client.
What is the legal foundation of Larry asking for reverse engineering to stop? You will also find that Australian copyright law has explicit provisions that render any licence invalid if it prohibits reverse engineering for interoperability purposes:
http://www.austlii.edu.au/au/legis/cth/consol_act/ca19681...
So, even if Tridge accepted some kind of licence agreement, it would not count. And he didn't. Tough luck for Larry, eh?
> I suspect that what tridge is saying is how he got started, but I'm also pretty confident that he went far beyond simply fetching things and started trying to figure out how to put things in, this is the part that would trigger alarms (from bad data, etc) and raise hackles.
And so what if he did? Reverse engineering is not illegal. At least not down here. And not yet (we imported some of the U.S legislation with the FTA, but that's not been enacted fully yet).
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 21, 2005 23:57 UTC (Thu) by dlang (✭ supporter ✭, #313) [Link] there is the moral question of someone giveing somethign away for free that normally costs a bunch of money, with the two conditions being 1. it's used for opensource software and 2. don't reverse engineer it.
you may have the legal right to reverse engineer it, but the moral right and respect for the author should weigh in on the question as well.
the fact that the legal letter of the license affects everyone at your company (to prevent having two people working side by side, one with the free version and one on it's competition) and many of the people at OSDL do work in the US where there aren't clear reverse-engineering exceptions should have also make a person think twice, and consider the effect that the actions will have on others.
tridge may (or may not) have had a legal right to do what he did, that doesn't make it moraly right.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 22, 2005 0:15 UTC (Fri) by pyellman (guest, #4997) [Link] I'm truly baffled by the repeated invocation by some of a "morality" element to this affair. Has it occurred to you that your position that what Tridgell did is immoral is a minority view? Most fairly, these events have little moral quotient at all. Even among those who have consistently maintained it was a bad idea to use proprietary SCM software - free or otherwise - as the core of Linux kernel development, morality was rarely mentioned; nearly always, the issue has been the consequences of a change of mind on the part of the owner of the software. Please, leave the big M for more appropriate discussions.
Peter Yellman
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 22, 2005 1:17 UTC (Fri) by rickmoen (subscriber, #6943) [Link] dlang wrote:you may have the legal right to reverse engineer it, but the moral right and respect for the author should weigh in on the question as well. The author has the moral right to compete in the marketplace with others, and they with him. He/she has the right to independently reimplement other people's designs and ideas (patents permitting) with his/her own creations. For example, he/she has the right to try to create his/her own client software to interoperate with someone else's server -- which is what Tridge is said to have been doing. Those are important rights, and our community has gone to bat for them. We'd do so for Larry, and we certainly would for Tridge. In this case, Larry would deny them to Tridge -- yet you're ascribing moral failings to Tridge. I find that very odd. The fact that the legal letter of the license affects everyone at your company (to prevent having two people working side by side, one with the free version and one on its competition)... According to Andrew's account -- and I've never known him to be anything but extremely honourable -- after BitMover objected to his working on the project under OSDL auspices, he did so solely using his own time and resources. I believe him, implicitly. However, since you've raised the issue of moral claims, I find morally troubling BitMover's claim that the kernel repository's metadata is proprietary to them. The change semantics stored in that metadata are the essence of the kernel project's workflow, so telling us we have no business understanding how it's stored strikes me as extremely questionable, and a loss of developer rights that the developers never would have consented to, had they been told that as an a-priori condition back in 2002. Rick Moen
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 22, 2005 1:32 UTC (Fri) by bojan (subscriber, #14302) [Link] > [...] and consider the effect that the actions will have on others.
Well, Linus clearly didn't think of that one when he decided to employ a proprietary tool, licensed under a draconian "no compete" clause, for kernel development. Where was his morality then? Why didn't he think of others? The whole thing was a disaster waiting to happen.
Linus may not agree with FSF's strong "freedom" position, but he said many times that he prefers "share and share alike" type of open source work. How did BK fit into that I cannot understand. I have no doubt Linus is a brilliant guy, but sometimes he tends to come up with really strange stuff.
> tridge may (or may not) have had a legal right to do what he did, that doesn't make it moraly right.
Well, as others pointed out elsewhere, what is different here to reverse engineering of various Microsoft protocol bits for Samba? Nothing. Should Tridge have some moral obligation to Microsoft because he dared to reverse engineer that? I just don't see it...
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 22, 2005 22:50 UTC (Fri) by sfeam (subscriber, #2841) [Link] Well, as others pointed out elsewhere, what is different here to reverse engineering of various Microsoft protocol bits for Samba? Nothing.Not true. There is at least one substantial difference between reverse engineering SMB and reverse engineering BitKeeper. SMB could be set up on a captive machine, endangering no one else's data while it was being prodded with exploratory queries . BitKeeper, however, was running live on a set of central repository machines containing actual data. I would hope and expect that it was designed to tolerate faulty protocol requests cleanly, but I don't think that hope or expectation is enough to remove an ethical taint from playing games with someone else's live database.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 23, 2005 2:03 UTC (Sat) by bojan (subscriber, #14302) [Link] > BitKeeper, however, was running live on a set of central repository machines containing actual data.
You mean PUBLICLY AVAILABLE machines, right?
And, the main work was with figuring out the metadata of the checked out stuff, which didn't involve the server at all.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 23, 2005 6:48 UTC (Sat) by jamesh (guest, #1159) [Link] Are you implying that someone could upload data to a bitkeeper repository via an unauthenticated connection?
I'd be surprised if the bitkeeper server processes providing anonymous access to the Linux bk repositories even have write access to the data they're serving. I'd be surprised if Larry designed something so fragile.
Do you consider it unethical/immoral to test new web browsers against existing public web servers? They also contain live data, that may be modifiable ...
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 23, 2005 19:32 UTC (Sat) by sfeam (subscriber, #2841) [Link] Do you consider it unethical/immoral to test new web browsers against existing public web servers? They also contain live data, that may be modifiable ...Well, let's suppose you are working on a set of scripts to simplify large-scale editing of data on a wiki. Fine, no problem. But if you decide to debug those scripts by testing them against, say, Wikipedia rather than debugging them first on your own private wiki installation... then yes, I'd say that was not an ethical choice. You should not risk other people's data, or their valuable time to clean up after damage, if you can accomplish the same goal without imposing any such risk. I have no problem with Tridge or anyone else purchasing a BitKeeper license and using it to set up a private site for use in reverse engineering the protocols used. That, I presume with no real inside knowledge, is how SMB was reverse engineered. Do you know otherwise?
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 24, 2005 0:25 UTC (Sun) by bojan (subscriber, #14302) [Link] > I have no problem with Tridge or anyone else purchasing a BitKeeper license and using it to set up a private site for use in reverse engineering the protocols used.
But that would be no good at all. The terms of BK license agreement (i.e. the contract) specifically include clauses that would have prevented him from doing this. Now, he may have chosen to go down that path (given that the laws are actually on his side) and risk being litigated against, but that would be really expensive for him (purchase licence, then fund defence). Not to mention inconvenient.
Instead, he has chosen the only legally and morally available path to him - to work with what's out there. And as others pointed out, I cannot believe BK repository would be open for writing to just anyone. So, even if he wanted to, he couldn't have damaged any data (i.e. your wiki example has no bearing on this).
So, the main problem here is the stupid "non-compete clause" of the BK licence. That's why Tridge couldn't have worked with it and that's why he had to reverse engineer things.
As for SMB reverse engineering, I'm sure you are correct. This was most likely done on private servers (nobody in their right mind exposes those on the Internet). And, only the bits that Microsoft slipped in and then failed to disclose were reverse engineered. The rest was from the specs. I'm not aware that Microsoft ask from people to sign off the competition rights through EULA, so it's not the same as with BK.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 24, 2005 3:29 UTC (Sun) by sfeam (subscriber, #2841) [Link] I concede that point. The purchase-and-reverse-engineer path also has ethical issues, because of the non-compete clause, enforcible or not. I had lost track of the fact that this clause was in the commercial license also. When faced with two bad paths to a goal, it may be time to re-think the goal rather than take either one. But this is all tangential to my original observation, which was just that reverse engineering Samba and BitKeeper are not identical cases. Each must by justified on its own.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 24, 2005 10:57 UTC (Sun) by bojan (subscriber, #14302) [Link] > Each must by justified on its own.
I think we'll probably have to leave it at "agree to disagree" here. I find no justification is necessary for any reverse engineering effort. When I initially said that there was no difference between BK and Samba, it was in the ethical sense. The technical process may be different, but as long as all the laws are obeyed, I see no issues.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 22, 2005 5:07 UTC (Fri) by huaz (guest, #10168) [Link] : there is the moral question of someone giveing somethign away for free: that normally costs a bunch of money
I can understand Larry's position and frustration, but Larry didn't really give BK away for free - there is a license, and one important piece is you'd lose your right to work on any other SCMs. And the fact that Larry is helping Linux Kernel team did not really put any obligation onto others.
Imagine one day Linus decided to use some Microsoft free tools to do kernel work (because it was the best). Should everyone in the open source community thus stop pissing off Microsoft so it wouldn't withdraw its support for that tool?
Replace Microsoft with BK here, now what's the difference?
I think neither Larry or Tridge is guilty, at least not more so than Linus.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 22, 2005 0:52 UTC (Fri) by rickmoen (subscriber, #6943) [Link] bojan wrote:Reverse engineering is not illegal. At least not down here. And not yet (we imported some of the U.S legislation with the FTA, but that's not been enacted fully yet). Even up here, behind the Disney Curtain, reverse-engineering for the purpose of compatibility remains lawful. For now. Best of luck fighting off the effects of the FTA. Rick Moen
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 22, 2005 17:00 UTC (Fri) by zabriska (guest, #29486) [Link] "Even up here, behind the Disney Curtain, reverse-engineering for the purpose of compatibility remains lawful. For now."
With the apparent exception of agreeing to a shrinkwrap (by buying) or clickwrap (by "Yes" or similar) license that says otherwise.
http://www.sethf.com/infothought/blog/archives/000173.html
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 23, 2005 6:56 UTC (Sat) by rickmoen (subscriber, #6943) [Link] That's so far only a First District position; the court may well be determined over time to have erred. E.g., I recall there having been other decisions that conflicted with the ProCD v. Zeidenburg precedent cited in support. Also, one notes some serious limits mentioned, e.g.: "In making this determination, this court has left untouched the conclusions reached in Atari Games v. Nintendo regarding reverse engineering as a statutory fair use exception to copyright infringement." But it does illustrate why consenting to a contract purporting to ban reverse-engineering is unwise and should not be tolerated. ;-) Rick Moen
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 27, 2005 6:38 UTC (Wed) by daniel (subscriber, #3181) [Link] "remember that the other side of the story was that after they found him pokeing at bitkeeper there were several weeks of negotiations"
Remember that the other side of the story came from Larry McVoy. Take that how you will.
Tridgell speaks out in BitKeeper war (ZDNet) Posted Apr 22, 2005 15:00 UTC (Fri) by ordonnateur (guest, #6652) [Link] seehttp://sourceforge.net/projects/sourcepuller/
|
|||||||||||||