| |||||||||||
How can anyone respect a changing license?How can anyone respect a changing license?Posted Jun 3, 2005 21:44 UTC (Fri) by lm (guest, #6402)In reply to: How can anyone respect a changing license? by felixfix Parent article: Linux dispute boils over to MySQL, other projects (ComputerWorld)
The "changing license" story is a myth:
$ ./bk help bkl|grep version
As for Tridge, I'd love someone to explain to me how he managed to snoop
We then watched the connects from his domain, it's clear that someone was
(Log in to post comments)
The myth is real Posted Jun 3, 2005 21:57 UTC (Fri) by felixfix (subscriber, #242) [Link] I didn't follow it closely, but I remember the license changing, or at least what McVoy said was the license. At some point the source code was withdrawn, at some point you couldn't use it to develop competition, at some point you couldn't use it if your company developed competition, and so on and on. It kept on getting more and more restrictive. I am not going to quibble over details, because I don't remember them, neither the order of the restrictions nor exactly what they were. But there was a definite feel of the noose tightening every time McVoy got a bug up his backside. He had a Do What I Mean, Not What I Say license, and what he meant changed over time. Companies were denied licenses which they could have gotten a year before. People were denied free usage for non-competing free source products because they were employed by a company who had even the slightest connection to other version control products, free or not.
Tridge says he didn't use the bk client, I believe him. Unless you saw him using it, what you saw on your end of the wire is immaterial.
The myth is real Posted Jun 4, 2005 13:43 UTC (Sat) by fergal (subscriber, #602) [Link] Posting a rant based on details you can't remember and won't bother researching is hardly productive.
I don't remember the details either because I looked at the license a long time ago. Even back then, those restrictions were there and I don't believe the source was ever available (certainly not for the server). It also had things like posting your changelogs on their servers. I never used it.
Oh please Posted Jun 4, 2005 14:26 UTC (Sat) by felixfix (subscriber, #242) [Link] That was not a rant. You want rants? See Larry McVoy for that. Don't call me on not researching something that you won't research either.
Yes, the licensing, or at least McVoy's interpretation of the license, did change as people kept on doing what had been legal but annoyed McVoy. There are any number of web sites with little stories about how they could not do what they had done before, because McVoy changed things. I am certain google or archives would show this if you cared to search. I won't, because I remember this. You won't, because you don't remember it. Which makes more sense?
And yes the client source was available, for a while. If you think not, you have not been paying attention. McVoy had a nice rant when he withdrew that.
Oh please Posted Jun 4, 2005 17:46 UTC (Sat) by fergal (subscriber, #602) [Link] I am certain google or archives would show this if you cared to search. I won't, because I remember this. You won't, because you don't remember in Right so, I've searched, I can't find any mention of the open source client. I did find a message in my old email: From: "Theodore Ts'o" <tytso@valinux.com> To: bitkeeper-users@bitmover.com Subject: How does one get the sources to bitkeeper? Date: Fri, 07 Jul 2000 10:01:47 -0700 OK, stupid question. Suppose I'm paranoid and I want to see the sources to bitkeeper (the license seems to imply this is possible, since it has all sorts of words about source distributions, and compliace with the test suites, etc.) How do I get access to the bitkeeper sources? They're not on the download page..... is there some other place I have to go to get a hold of one? Say, a place where I can get a clone of the repository for bitkeeper? :-) Does that sound to you like the client was open-source in 2000? I also tried to find the source (I think that's why I joined that list) and it wasn't there. What about BitMover announces Open Source Client for BitKeeper from March this year? Does that sound like the client had been open source before? I didn't see any mention of it in the LWN comments. As Larry's reply to you shows, the license hasn't changed since 2003. Larry's interpretations may have changed but it's the judge's interpretation that really matters. I'm not trying to defend Larry, but comments based on vague memories of things you weren't really following anyway aren't very helpful.
Easy way to find out what port to connect to. Posted Jun 4, 2005 2:10 UTC (Sat) by flewellyn (subscriber, #5047) [Link] It's called "nmap", and it works quite nicely. You just portscan, and then look for anything that looks likely.
And he used telnet, of all things, to connect. TELNET! And it worked.
If that qualifies as reverse engineering, it can only be applied to sloppy forward engineering.
How can anyone respect a changing license? Posted Jun 4, 2005 19:33 UTC (Sat) by GreyWizard (subscriber, #1026) [Link] Great, more classic Larry McVoy misdirection. Please forget that kernel developers adopted BitKeeper a wee bit before 2003. Pay no attention to this post by a certain Larry McVoy claiming that the license was... well, changing. Also please ignore the 404 error you'll get if you try to follow the link given in that post to the text of the license itself. Inconvenient evidence like this might lead you to the conclusion that the "changing license" story is true. I don't know how that dirty, rotten scoundrel we call Tridge figured out the protocol changes on bkbits but I'll take a wild guess: he ran nmap and applied common sense. (See flewellyn's post above.) Care to tell us what cunning protocol change you made on bkbits that he couldn't possibly have worked out using elementary methods? No, of course not. So much more convinent to spread half-truth and innuendo with so little substance it's impossible to disprove, isn't it?
How can anyone respect a changing license? Posted Jun 5, 2005 18:27 UTC (Sun) by dmaxwell (guest, #14010) [Link] Dr. Tridgell telnetted to a bk server. Typed the 'help' command and used netcat to dump the contents of the "clone" command to a file. Gosh but Tridge is 'leet evil reverse engineer isn't he? At the conference where he revealed what he did, the audience was shouting out the commands ahead of him. He used a bog standard UNIX tools in a one liner shell command to dump a repository. That is "echo clone | nc thunk.org 5000 > e2fsprogs.bk"
At this point, I have a difficult time believing you haven't gotten the straight of this story. Why are you still implying that Dr. Tridgell did anything improper?
It wasn't terribly difficult to analyze the clone dump and create a tool for working with it.
|
|||||||||||