Torvalds Unveils New Linux Control System 527
BlakeCaldwell writes "CNet reports: 'Linux founder and leader Linus Torvalds has launched a new tool, called Git, to manage his software project, after a dispute led him to drop the previous system.' He will start using Git instead of BitKeeper to control the flow of updates and track changes in the kernel." We've covered this previously. Relatedly, ChocLinux writes "Jeremy Allison, who wrote Samba with Andrew 'Tridge' Tridgell, is sticking up for his friend in the row over BitKeeper. "
how come (Score:3, Insightful)
Re:how come (Score:5, Insightful)
Re:Based on Monotone it seems (Score:3, Insightful)
I think Larry is going to be unhappy with the end result, because by cutting off the kernel developers he is triggering a lot of work on development of open SCM tools.
What is bad with copying funcionality?? (Score:5, Insightful)
But that's not what Tridge did... He didn't create something new and impressive. He just tore down something new (and impressive) because he could, and rather than helping others, he screwed people over. And you expect me to respect that kind of behaviour?" wrote Torvalds
Come on!, so what if someone makes a program that implements a cool funcionality from another?? I see it in every game that has been developed in the last 20 years!, thats why whe have genres!, also, that would mean that OpenOffice is bad! or what about the same Linux (Unix clone??) or all the BSD's.
I think Linus went to far with that, so also to do SAMBA was a "non respectable behaviour" to him? wtf without SAMBA I bet they would be a really, REALLY big amount of people (and companies) not using Linux these days.
If he does not want to use it, then do not do it, but do not flame the author for doing it, and tell that is not a respectable behaviour! it seems that the most notable figure of Open Source has acquired a Not-So-Open State of Mind.
my 2C
Bitkeeper was a great descision (Score:4, Insightful)
Very simply that time has passed, and NO-ONE other than Linus himself knows what works best for HIM and his direct team
The flaming is useless hes "The Man" and what he wants for us in Linux land is pretty much Law, besides, how many of YOU psting all these nasty comments about his original BitKeeper descision actually were granted access to it directly, NONE.
Linus is a pragmatist not a rabid OS advocate but willing to use closed source tools if its a winning situation for him.
So is most of the world. (Score:4, Insightful)
Yeah, well, so is nearly all the world, except for Linus and Larry McVoy. I'm sorry, Linus' actions are just plain hypocritical here. I can understand how he was pissed at losing a useful tool. I can't understand how he can promote McVoy at the expense of our freedoms, especially to reverse Engineer.
Mod me down, but Linus has too big of a head on his shoulders. He is NOT indespensible, thanks to the GPL. What does go around, comes around. And this action won't be forgotten. With all due respect to him, I think it's one of his biggest blunders in the history of Linux.
Re:Obligatory (Score:5, Insightful)
"Well, after looking it over, we've decided to buy a Bitkeeper liscence. it seems more robust than the competition, plus linus uses it!"
Re:zdnet.co.uk (Score:5, Insightful)
The Linux kernel history was being held hostage to Bitkeeper's good graces. If the business reasons for letting kernel developers do advertisement and Beta testing disappeared, the free version would inevitably disappear and kernel developers would be SOL (as they are now).
If it weren't for the foresight to mirror *some* of the BitKeeper information in CVS, the kernel developers would have no developement history other than what they can dig up in the archives.
Politics is bullshit (Score:1, Insightful)
Linus has always been consistent about one thing: you use the best available tool and screw the politics.
Tridge's clone doesn't work. CVS doesn't cut it. BitKeeper works! So, that's why we're going to use it even if it's not Free.
I'm beginning to feel that all this smacktalk about how Linus made a blunder is a backslash from the rabid zealots who got their feelings hurt when Linus opted for non-GPLd software.
Re:Bitkeeper was a great descision (Score:3, Insightful)
*snort*
Linux is a great guy, who's got both strong technical and strong "people" skills - a rare combination that's made Linux what it is today. But he does get bent out of shape over operating systems awfully easily.
Re:zdnet.co.uk (Score:5, Insightful)
telnet bitkeepermachine
HELP
--seeing the list of available commands--
clone filename.c
seeing a bunch of garbage, then shortening it to:
echo "clone filename.c" | telnet bitkeepermachine > filename.c
wow that's what I call reverse engineering!
Re:zdnet.co.uk (Score:1, Insightful)
Tridge intentionally reverse engineered BitKeeper despite the fact that he knew that his efforts would be discovered and would scuttle the BitKeeper donation.
Before 1972 it was "legal" to dispose of hazardous waste on your property. It was still immoral to do so knowing that various nastiness would migrate across the property line and into your neighbor's well.
Tridge is not in the right. Tridge has demonstrated his immaturity by disregarding the effects of his actions on others. Tridge has the skill to write a source control system from scratch, and intentionally chose not to - all claims to moral superiority died with that decision.
Re:Politics is bullshit (Score:3, Insightful)
That's why he quit developing his shaky kernel (and the Linux kernel WAS pretty shakey back then) when BSD-Lite was released and jumped on the BSD bandwagon.
Whoops, wrong universe.
Re:Obligatory (Score:5, Insightful)
In fact I think BK got a bargain and they've gone and thrown it away.
Re:wow, that has some rough edges alright (Score:4, Insightful)
Obviously what needs to be taken into account is that is isn't necessary to do the delta creation at the point of check-in etc, a background process could promote 'clone' type mods into 'delta' type mods at its leisure, maybe even with a delay so only patches unlikely to get reverted are folded down to delta format.
That way the workload gets distributed as well as the actual development.
Re:What is bad with copying funcionality?? (Score:5, Insightful)
I'm sorry, where exactly has Microsoft been accused of stealing when they copied functionality?
They've been accused of stealing when they've actually used other people's code. They've been accused of embrace-and-extend when they've copied functionality and modified it so the original product they copied no longer interoperates with them. Stealing ideas? Sure, everyone gets accused of that, but nobody in the OSS community with any credibility is going to use that kind of phrase except in jest. And when Microsoft "steals" ideas and they're good ones they often get praised and encouraged for it... all the way back to hierarchical file systems and UNIX style system calls in DOS 2.11...
So you can keep your "double standard" banner under your hat today, it's not happening.
Re:The ZDNet article gets it wrong (Score:5, Insightful)
Only when you have to circumvent authentification (i.e. pretend to be someone else) it gets nasty.
If you connected to my system and got some stuff off my webserver it would be perfectly fine. If you used a rootkit to get a shell you would be in trouble.
Jeroen
Please (Score:1, Insightful)
The reall shame is that intelligent people such as yourself have bought into the myth that allowing people to attempt to patent/copyright files and protocols represents some sort of moral standing.
I think this line of reasoning is absurd.
Re:Doubled file size!? (Score:3, Insightful)
Linus explicitly mentioned that time for commits was a primary consideration, and 'mv' is a lot faster than 'diff'.
Re:zdnet.co.uk (Score:5, Insightful)
Reverse-engineering for interoperability is legally fine, unless you're bound by a license not to do it. Those who've accepted the free BitKeeper client (or who bought BitKeeper) are subject to just such a license.
If Tridgell never acceted the BitKeeper license, then he's not bound by it, and there's nothing illegal about what he did. But you know, you don't have to do something illegal to piss people off. :-(
McVoy got pissed that someone did what he didn't want anyone to do, so he decided to stop maintaining the free BK client. (He's also trying to say that Tridgell should have been subject to the BitKeeper license, since he happens to be a contractor doing some work for a company that had accepted the BK license. I don't buy that one.)
Torvalds got mad that something somebody got McVoy mad, so that now his choice source control tool isn't freely available anymore. He ranted against Tridgell, but that's misplaced, I think. Torvalds isn't fully into the "Free Software" philosophy (despite his use of the GPL for Linux), and so doesn't see any value in Tridgell's work and calls it "evil."
Re:zdnet.co.uk (Score:5, Insightful)
There is a reason why he reverse engineered the file format and not the source control system. BitKeeper stored the code in a proprietary data format, their IP is the only data capable of reading that format. If BitMover ever chose to revoke all licenses to use their IP to read the format it would bar the kernel developers from retrieving the kernel source code stored in the system.
Tridge was not building a source control system to mimic BitKeeper, Tridge simply reversed the file format so that the ability for kernel developers, linus, and the world; to access the kernel source code was not subject to BitMovers good graces.
Re:Bitkeeper was a great descision (Score:5, Insightful)
Re:zdnet.co.uk (Score:3, Insightful)
Downloading something from a publicly accessable server using no authentification whatsoever is perfectly fine and has nothing to do with the bitkeeper license.
If I download a html file from a http server the Apache foundation can say whatever they want but they couldn't stop me...
Jeroen
The Great Rift (Score:3, Insightful)
I think that attitude marks a great rift between open source (OS) advocates. There are those who support both OS and proprietary software (PS), and those who think all PS is wrong. Judging from their public statements, Linus is in the first camp, and Richard Stallman is in the second.
Myself, I think free people should be able to to choose whatever approach they want, and good luck to them. And I'm bloody tired of all the fanatics in the world who take a good cause and elevate it to a mindless religion.
There is no "appeasment" here because there is no enemy.
It will be interesting to see how it all plays out.
Re:So is most of the world. (Score:2, Insightful)
This chain of events was triggered by Tridge's work. That explains the intensity of the venom spat out at Tridge.
Tridge did nothing wrong cos globally, reverse engineering is completely ethical and legal for interoperability.
McVoy is a scumbone who has abused everyone involved including his friend Linus, in order to get free publicity for his bitkeeper product.
Linus is a fool to get manipulated so easily by McVoy. It is Linus' inability to distrust his "friend" that is making him say all these weird things.
Re:zdnet.co.uk (Score:5, Insightful)
Re:Doubled file size!? (Score:2, Insightful)
Comment removed (Score:2, Insightful)
Re:zdnet.co.uk (Score:2, Insightful)
Re:The ZDNet article gets it wrong (Score:5, Insightful)
Where's the agreement that Tridge had to sign to use it? The port's open to the public. After all, the BitKeeper server could've "closed the door". It didn't. It could've required authentication. It could've been encrypted. It wasn't.
I don't have to sign an agreement to telnet to www.slashdot.org, port 80. If they suddenly put up something on the web that said "You have to sign this license and agree to these terms in order to do this", and I never see that license, how can I be bound by it? Given that there's a thousand ways to prevent unlicensed access, and they did't do any of them, how can it be negligence on my part, and not theirs?
If that isn't either illegal or in violation of a license, then be a man and post your IP. I'll "reverse engineer" your machine and see what interesting thing I can do.
If you think that's illegal, you're out of your mind. You don't have to sign a license agreement to "use" a server just because the server exists. He wasn't presented with a license, he didn't agree to a license. The fact that you're trying to claim someone toying around with a random port is 'illegal' or 'in violation of a license' is just ludicrous.
Re:Is this for real? (Score:4, Insightful)
Perhaps it's all about ego (Score:4, Insightful)
Re:zdnet.co.uk (Score:2, Insightful)
That's ridiculous. If people didn't want to use BK, then they just wouldn't. Linus did not approve of Tridge's actions.
If BitMover ever chose to revoke all licenses to use their IP to read the format it would bar the kernel developers from retrieving the kernel source code stored in the system.
Now that's ironic. Tridge's actions are what caused BK to revoke the license. If he had done nothing, then we'd all still be happily using BK.
Re:zdnet.co.uk (Score:5, Insightful)
You and Linus should get together and have a big "completely missing the fucking point" party. I have always respected Linus Torvalds greatly for his technical and societal contributions until this whole flap.
Anyway, everyone wanted a way to get ALL data in and out of the repository to make it easier to merge in third party patches. The only differences here between (say) Linus and tridge are that Linus is McVoy's buddy who is pissed off that he no longer gets to use software, and tridge is a champion of Open Source who feels that we should be in charge of our own data. Remember, Linus does not care if software is Open. He wrote Linux that way based on a pragmatic decision.
Now, I actually pride myself on being a creative asshole, but sir, you take the prize. Tridge was trying to do one thing: INTEROPERATE WITH BITKEEPER. He was NOT writing a replacement. How could he be, he was only writing a client? This cannot possibly hurt bitkeeper; they sell a service, and there are already other clients. He was trying to develop functionality that did not exist, and that bitmover does not want to provide.
Why do they not want to provide this functionality? Why do they want to hold the interface to YOUR DATA hostage? So that they can lock you in to bitkeeper. In other words what you and Linus are both completely missing in your ire is the fact that bitmover is entirely microsoftian in this regard. They want to lock down the standards and prevent people from using them (McVoy threatened to repeatedly change the protocol in order to stay a step ahead of tridge - I'm sure that's going to provide a lot of benefit to bitmover's customers!) so that it is extremely inconvenient to move to another SCMS. This is 100% the same as Microsoft's file format strategies; hell, if they open DOC, in some ways they'd be LESS restrictive than bitmover - fact not fiction.
Reverse engineering for the purposes of interoperability is a time-honored tradition, and protected at least here in the U.S. by federal law including the oft-reviled DMCA. Perhaps you should adjust your attitude, I think it's poking out of your ass.
The CalvinBall License (Score:5, Insightful)
Re:zdnet.co.uk (Score:5, Insightful)
Tridge reversed engineered [groklaw.net] BitKeeper in the same way I "reversed engineered" SMTP:
There are a bunch of developers saying "I told you so" to Linus; BitKeeper may have been a wonderful product, but it was a train wreck waiting to happen. If Tridge had done nothing, the result would have been the same except that Linus would have to find another scapegoat to take his frustration out on.
Re:Tridge Speaks (Score:2, Insightful)
1. OSS community doesn't like proprietary formats/software.
2. The OSS community is very good at reverse engineering formats.
And ultimately yes, the BK format would be reverse engineered for more freedom. It is like giving a beer to Homer Simpson and expecting him not to drink it.
That is all.
Re:Perhaps it's all about ego (Score:2, Insightful)
Arch and Monotone have progressed a lot since the decision to use BK. That might have been a good decision to avoid them back then. That doesn't mean they shouldn't be considered at all today.
I really don't see why a brand new tool had to be written for this purpose. I bet any of the OSS tools' development teams would be eager to jump in and work on any of Linus' gripes. I wonder how much time the kernel guys will spend working on git when they could be working on new kernel features/fixing bugs. Starting *completely* over is rarely the right decision in the long run.
My hope is that git serves useful in the short run to keep the kernel from completely stalling, but that a switch to something more standard happens within the next 6-12 months.
Re:zdnet.co.uk (Score:5, Insightful)
The problem is that Larry is an insane person. Let me draw up an analogy for you.
Let's say there is a park close to your house and you like to walk your dog there.
Larry one day calls your boss and says that you are ugly, your dog is ugly, and he wants your boss to fire you unless you stop walking your dog when you get home.
OF course both of you are a little bewildered because a) you are doing something on your own time and b) what you are doing is perfectly legal. So your boss tells him that no he is not going to fire or rebuke you in any way.
Larry then calls back and says that if you don't stop walking your dog he will beat of on his friend linus. You know linus and think he is a good guy but you figure linus can take care himself and indeed is a pretty good jodo expert so you tell him no again.
Larry then lashes out at his friend and his friend is now mad at you!.
This is what happened. Larry stabbed his friend in the back in order to make tridge stop doing something he had legal right to do.
Larry is an immoral insane person and linus had no right to yell at tridge for larry's actions.
You missed some of the viewpoints. (Score:5, Insightful)
#2. Linus chose to use BitKeeper knowing all of that. He still chose it because it seemed to be the best product around that would meet his needs. Linus did not seem overly concerned about the potential for losing the "free" client.
#3. Tridge did not break any laws when he started to reverse engineer the packets.
So
Where's the problem?
Well, Tridge should have known that his work would piss off McVoy and that it could result in the loss of the "free" client. Yet he did it anyway WITHOUT writing a SCM that was as good or better than BitKeeper.
So, the only thing that Tridge is guilty of is not having a replacement ready for when everything blew up.
McVoy decided that he didn't want to deal with Tridge's work and just pulled the "free" client to stop what he viewed as a threat to BitKeeper.
So the only thing McVoy is guilty of is attempting to protect his own project.
Which leaves Linus suddenly without an SCM and he blames Tridge for wreaking a working situation without having a replacement ready.
So, the only thing Linus is guilty of is venting publicly.
So why is everyone picking sides? That comes down to each person's values.
A.) Those who value Open'ness more than functionality support Tridge because they believe Linus was wrong to push a proprietary product.
B.) Those who value functionality more than Open'ness support Linus because the system was working and it was helping development and there isn't an equivalent system to replace it yet.
But those are simply judgement calls based upon each individual's value set. Neither is more "right" or "wrong" than the other, except in your opinion.
Re:You missed some of the viewpoints. (Score:3, Insightful)
What about those of us who value Openness AND functionality? I value functionality, so I think tridge should have been able to develop his addon, which does NOT replace or compete with BK (McVoy is primarily guilty of being a whiny baby who wants the world to be different than it really is) to provide that functionality. I also value Openness, so I doubly think that he should have been able to develop his tool.
Plain and Simple, McVoy claims to be upset that tridge was creating a BK replacement. He was doing nothing of the sort. He was trying to use the existing system, but develop additional functionality. McVoy took his toys and went home; There is no way in hell I would do business with bitmover knowing what I now know about McVoy.
I will agree that it is not pragmatic to piss off the person giving you software. However, if you need functionality, you develop it.
Re:You missed some of the viewpoints. (Score:3, Insightful)
Re:You missed some of the viewpoints. (Score:1, Insightful)
Yes, that's all true. I don't fault McVoy for being an assmunch, it happens. In fact I recommend BitKeeper to my non-open source using friends, because it really is the best system available. But it's not open source. Licenses matter. Politics matters. Legal quibbles matter. More than the code itself, in many cases. People like Linus think they can avoid this stuff, and other coders lap it up, but the reality is more like how RMS sees the world.
Think about it: the reason people use Linux (and free software generally) is to AVOID this kind of crap. That's all. So when Linus comes along and says he going to go "closed source" for his own benefit, you just know there's going to be a conflict someday down the road.
Just for that reason alone, it was a mistake for Linus to use BitKeeper, and to give McVoy all the free press and "beta testing", with no benefit to anybody but Linus.
Maybe people will think twice about this stuff in the future and stop trying to score "street cred" by being "pragmatic", whatever the hell that means.
Re:zdnet.co.uk (Score:3, Insightful)
No, using software from a company run by a megalomaniac who wants to control your data and lock you into his solution is the part that is stupid. It was stupid when Linus decided to do it in the first place, and it is equally stupid today.
Linus' refusal to admit that it was stupid to use McVoy's system is the real problem here. He says he's an egomaniac; I guess it's true. His ego won't allow him to admit that it was a ridiculous decision. He seems to be buddies with McVoy, so that may be part of it as well.
Linus chose to use an unfree system because it provided him with certain advantages. He claims he is a pragmatist and not an open source champion, but first of all taking the easy way out and using BK was NOT pragmatic, because it was inevitable that something like this would happen eventually; Second, clearly he has the ability to write the kind of software he needs, which will necessarily have licensing terms he can live with. This makes much more sense than being locked in to proprietary software.
The software was not self-defeating. It solved the very problem it set out to solve - being able to manipulate Linux data (which doesn't even BELONG to bitmover!) in the way that the developers need to manipulate it to be effective. Granted, it did cause many other problems, but I think it can be seen as a success for open source.
Again, Linus says he's a pragmatist, but one of the primary goals of Linux was extensibility and redistribution. You cannot have that without Open Source! So really, by supporting closed-source software, Linus was shooting himself in the foot. Since Linus made that decision, and this is hurting all of us, Linus is shooting us ALL in the foot. Granted, the gun went off when tridge bumped Linus, but he is the one who had his hand on the grips and his finger on the trigger.
Re:zdnet.co.uk (Score:2, Insightful)
Tridge was just minding his business doing something he had full legal right to do.
Last time I checked, I have the full legal right to sleep with my best friend's girlfriend. That doesn't mean that a) I should do it, or b) it won't affect my friend.
Re:You missed some of the viewpoints. (Score:1, Insightful)
While I'm more than willing to compromise on openness vs. functionality when choosing software to use, what I'm much less willing to compromise on is my freedom to write whatever programs I want using legal means. For something that I'm involved in myself (such as my day job), I can accept restrictions on that.
But if you're passionate about programming like I am, and Tridge seems to be, anyone trying to tell you that you shouldn't work on some particular program is offensive.