FreeBSD Core Developer Thrown Out 681
SlashChick writes "From a discussion on the freebsd-chat mailing list, it appears that one of the FreeBSD core developers, Matt Dillon, has been barred from committing any changes to the FreeBSD kernel. Dillon was one of the developers 'responsible for making FreeBSD 4.x the most rugged and stress-proof free operating system in existence,' and also contributed to fixing the Linux VM. Unfortunately, there has been little explanation from the FreeBSD core team about why Dillon was thrown out, leading to speculation and worries about the future of the FreeBSD kernel. Does the Slashdot community have any more insight into this situation? Would someone from the FreeBSD team care to elaborate and assuage our worries?" CD Update: Greg Lehey from the core team has infact elaborated in this comment.
I know as much about this as the next guy (Score:5, Insightful)
Differences in opinion. Maybe I am wrong (NEVER!) but that would be my guess.
Another Branch? (Score:4, Insightful)
On that note, it's more likely would get adopted by one of the other BSDs, and not really need to start his own. I'm sure OpenBSD can use the help.
Politiburo (Score:5, Insightful)
One thing the BSD developers need to know is that they have no justification in keeping this secret. It is aboput the users after all.
Re:Free BSD Dying (Score:5, Insightful)
Linux boxen can be made just as stable, just as reliable, just as "there" as any BSD. I guess in a round about way I'm saying it comes down to preference and familiarity just like anything else. I'm familiar with Linux (11+ years familiar). I'm not as familiar with BSD (only about 5 years now) but I know a good deal of its strengths and weaknesses and I'm happy to say that there are places I would put BSD right now and not one of the Linux distros (barring Slackware or my own) just because of extra work involved that should be unnecessary.
Ok, done ranting.
Cliff
Re:Free BSD Dying (Score:2, Insightful)
Be careful about generalizing.
FUD (Score:5, Insightful)
Firstly, the FreeBSD Core team (the use of "core developer" in the title of the article could be misleading) have given a lengthy explanation of this decision on the developers private list. This is where the explanation belongs and where it should stay. The reasons and the action are internal to the project and don't need to be aired in public.
Secondly, Matt is not the first, nor the last I dare say, high profile developer to leave the project. It didn't mean the death of FreeBSD then, it doesn't now. No single developer, no matter how talented and hard working, is irreplaceable. While Matt's technical contributions will certainly be missed, the claims of "imminent death, film at 11" are the same baseless FUD that came out when Mike Smith left or would have come out when John Dyson left (had Slashdot been around).
Thirdly, Matt is still free to contribute should he so wish. The only difference is that he will now have to contribute through PRs, at least for the near future, just as every other contributor started off doing, rather than directly committing himself. Whether he chooses to do this once the dust has settled is, of course, up to Matt.
Finally, long live FreeBSD! Can we please get back to worthwhile stories now :).
Re:Little explanation? I think there's enough. (Score:2, Insightful)
In my mind an unwritten requirement for most jobs is "smart and friendly." If you rub people the wrong way you're limiting yourself to small, one-person projects. Not the end of the world if you (or your manager) recognize this and play to your strengths. -IT
Re:Little explanation? I think there's enough. (Score:2, Insightful)
I've read some hella good flames and wars on the linux kernel dev list, I never recall someone being invited not to take part though. Al Viro is especially good and reading your code and then telling you exactly how incompentent your are.
If this guy is the master hacker everyone makes him to be, this isn't enough explanation. Shouldn't the users have some say? Perhaps the mistreated developers should move on to other projects or maybe grow into adults and learn to take the heat, it's just software, it's not like you should be taking the flames seriously.
Re:Little explanation? I think there's enough. (Score:2, Insightful)
I think that's a little unkind. Politics and ideology get in the way of many things. Someone that's a brilliant coder is of little use to a team if they are not prepared to listen to other people.
If I were a hardned cynic of course I would refer the reader to my signature...
Re:Little explanation? I think there's enough. (Score:5, Insightful)
Suffice to say, the ends do not justify the means.
Re:FUD (Score:2, Insightful)
Thank you for providing, in a single sentence, the best reason to use Linux rather than FreeBSD. With FreeBSD, you basically have a bunch of whiny elitest developers writing code behind closed doors and releasing periodic snapshots for the dirty masses to consume. With Linux, you have a bunch of whiny, elitest developers writing code in the open, warts and all. Since pretty much all developers are whiny and elitest, I'd much much rather have the second situation when I am using code for important tasks.
Idiots. Marshall Matt Dillon came first. (Score:1, Insightful)
Free BSD (not) Dying (Score:5, Insightful)
First, remember that there is no magic bullet. There are always tradeoffs with anything. Linux has definate strong points (new hardware support usually hits linux first; there are more developers for linux). FreeBSD has fewer developers, and doesn't support the newest hardware as quickly - but the (FreeBSD) network stack is extremely solid, and the system design is very clean.
So, you have to evaluate your goals in these kinds of situations. Are you out to get the newest hardware and features, or are you looking for a clean design and good performance.
There is a reason many sites (like Yahoo [yahoo.com], imdb [imdb.com], cr.yp.to [cr.yp.to]) use Open/FreeBSD to run their servers.
If that's not one of your priorities, but you're still curious: I'd still take a look at FreeBSD; the overall design is quite pleasant to work with.
Also, many of the exploits produced are usually done on Linux, at least initially. This could buy you a little extra lead-time when something malicious is released. It's not security by obscurity, but it is a fringe benefit.
As always, if you're truly curious as to which OS would suit you best, you should put a little effort into it, and do [lemis.com] some [linuxworld.com] research yourself [tek-tips.com]. I'm not saying you shouldn't use Linux, and I'm not saying you should use FreeBSD. FreeBSD is not for everyone. Linux is not for everyone. Do the research, decide for yourself, and next time - when you feel the urge to ask "why use *BSD?" -- you'll be able to at least discuss what you do or don't like about either. Otherwise, you end up contributing nothing to the discussion.
Re:Little explanation? I think there's enough. (Score:1, Insightful)
I do. Sounds to me like his ego to brain ratio was way, way too high. It's a consistent problem with many developers (not all, so don't even fucking think about modding this flamebait, flamers). Writing code is not necessarily easy, but you're not Jesus Christ or anything. I have yet to meet a developer that doesn't make mistakes. Ever.
Re:FUD (Score:2, Insightful)
Please explain how a public CVS repository (and CVS commit mailing list) is "writing code behind clsoed doors".
While you are at it, you may want to explain the logical leap from internal developer politics to code openness.
Re:Similar to the Net/OpenBSD split (Score:3, Insightful)
dillon will, in fact, leave the project for good: nothing in this matter requires that. But it's certainly good to note that the BSD world is better for OpenBSD, not worse. Maybe dillonBSD would be another good one.
Re:I know as much about this as the next guy (Score:2, Insightful)
He was probably overly nice and diplomatic.
If FreeBSD is going to die becasue of this.... (Score:4, Insightful)
If they let this project revolve around one guy then the project was doomed from the beginning.
Actually this is a good test of FreeBSD, if it survives its because its bigger then this one guy, which is the way projects should be.
Re:Little explanation? I think there's enough. (Score:5, Insightful)
The only difference between the bitchy whining children in open source development and the bitchy whining children in closed source development is that the latter have signed NDAs that keep the pissing contests out of the public spotlight.
Re:Kumbaya... (Score:2, Insightful)
It is a sad day when one person's arrogance and abusiveness excludes them from technical discussions.
Communities have a right to choose with whom they wish to associate. I don't HAVE to put up with your abuse, just because you're smart.
Re:dillon leaves the FreeBSD project (Score:3, Insightful)
Re:Little explanation? I think there's enough. (Score:3, Insightful)
Open projects who have a decided leader (Linux w/ Linus for instance) can also work this way, though I doubt Linus wants to arbitrate every argument
Re:Little explanation? I think there's enough. (Score:3, Insightful)
Apples, oranges.
Matt Dillon is not banned from any mailing lists. He's only been removed as a "committer". In linux there's only one "committer", Linus himself. Others send patches, to the mailing list or to their pet maintainer upstream. Matt can still do both with FreeBSD, or simply use the send-pr command. What he can't do now is make changes directly to the source tree.
Re:get some perspective, please. (Score:4, Insightful)
Huh? Whose right is being taken away? The legal IP owner who decides to license his software under the GPL? The person that wants to use the software without abiding by the IP owner's license?
Exactly what private right is being taken away?
Re:well..... (Score:2, Insightful)
HOW COULD IT DAMAGE FREE SOFTWARE?
And what's the difference if BSD code used by Microsoft or IBM or Sun or Apple ? Or any other company or organisation or government?
Re:dillon leaves the FreeBSD project (Score:3, Insightful)
Re:dillon leaves the FreeBSD project (Score:5, Insightful)
The FreeBSD core team has informed the development community in detail about the reasons for Matt's removal. We don't think it's appropriate, nor fair to Matt, to wash dirty linen in public.
When it comes to Free software, isn't "the development community" the same as "the public"?
This stuff raises more questions than it answers. Granted, I really don't care about the precise names behind FreeBSD, and I'll forget about this promptly I'm sure, but I hate to think the influential folks in the BSD community aren't engaging in some kind of power play or the keeping of dirty little secrets. There should be openness in the mechanics of the BSD as project, just like the code is open.
I'd like to think Free software is above the stuff that goes on in corporate boardrooms, but maybe not?
Re:dillon leaves the FreeBSD project (Score:4, Insightful)
I'm sure this will get modded down, but that's a pretty gutless statement to make, and really isn't supported by the commit logs. Though, when it's time for Core to toss someone under the bus...
To be sure, there is plenty of history with Matt, much of it not great. He's simply not a team developer. However, I honesty hope there's more to this [freebsd.org] and this [freebsd.org] than there appears to be.
One wonders when Core is going to stop acting parents and start acting like leaders.
Re:dillon leaves the FreeBSD project (Score:0, Insightful)
What's sad, Greg, is that the FreeBSD Core Team -- of which you are perhaps the most elitist member -- is acting like a bunch of immature children acting as if they were a Grand High Cabal... and ejecting one of their peers from the clubhouse.
By acting like a secretive club and making decisions such as this one behind closed doors, you are damaging the reputations of FreeBSD and the BSDs in general.
FreeBSD is dying (Score:0, Insightful)
One more crippling bombshell hit the already beleaguered *BSD community when IDC confirmed that *BSD market share has dropped yet again, now down to less than a fraction of 1 percent of all servers. Coming on the heels of a recent Netcraft survey which plainly states that *BSD has lost more market share, this news serves to reinforce what we've known all along. *BSD is collapsing in complete disarray, as fittingly exemplified by failing dead last [samag.com] in the recent Sys Admin comprehensive networking test.
You don't need to be a Kreskin [amazingkreskin.com] to predict *BSD's future. The hand writing is on the wall: *BSD faces a bleak future. In fact there won't be any future at all for *BSD because *BSD is dying. Things are looking very bad for *BSD. As many of us are already aware, *BSD continues to lose market share. Red ink flows like a river of blood.
FreeBSD is the most endangered of them all, having lost 93% of its core developers. The sudden and unpleasant departures of long time FreeBSD developers Jordan Hubbard and Mike Smith only serve to underscore the point more clearly. There can no longer be any doubt: FreeBSD is dying.
Let's keep to the facts and look at the numbers.
OpenBSD leader Theo states that there are 7000 users of OpenBSD. How many users of NetBSD are there? Let's see. The number of OpenBSD versus NetBSD posts on Usenet is roughly in ratio of 5 to 1. Therefore there are about 7000/5 = 1400 NetBSD users. BSD/OS posts on Usenet are about half of the volume of NetBSD posts. Therefore there are about 700 users of BSD/OS. A recent article put FreeBSD at about 80 percent of the *BSD market. Therefore there are (7000+1400+700)*4 = 36400 FreeBSD users. This is consistent with the number of FreeBSD Usenet posts.
Due to the troubles of Walnut Creek, abysmal sales and so on, FreeBSD went out of business and was taken over by BSDI who sell another troubled OS. Now BSDI is also dead, its corpse turned over to yet another charnel house.
All major surveys show that *BSD has steadily declined in market share. *BSD is extremely sick and its long term survival prospects are very dim. If *BSD is to survive at all it will be among OS dilettante dabblers. *BSD continues to decay. Nothing short of a miracle could save it at this point in time. For all practical purposes, *BSD is dead.
Fact: *BSD is dying
Re:dillon leaves the FreeBSD project (Score:2, Insightful)
To top that off, having used FreeBSD extensively in the past, I must say that FreeBSD is one of my favorite free/OSS operating systems, and I enjoy and appreciate it's refinement and clean implentation on a level that no Linux distribution has managed to equal (yet).
Surely these count for something.
Why resort to cheapening this excellent operating system with ignorant comments like "it's not linux anyways."
Regardless, the "core team" voting and commit logs tell the tale more clearly then any teenage pissing contest that may brew on
Re:dillon leaves the FreeBSD project (Score:4, Insightful)
very few people, especially on this board, contribute to any open source projects (I would say far less than 1%).
So its hardly like the great unwashed masses are making linux what it is; the unwashed masses DO, however, make Slashdot what it is...
Re:dillon leaves the FreeBSD project (Score:2, Insightful)
Maybe so. However, in a controversy the persons involved do feel the pressure to make some kind of general statement. It's respectible that the leaders of the FreeBSD development team do not want to drag Matt's name through the mud, despite the fact that something bad obviously happened causing the leadership to take action and revoke his commit privileges.
It is a sad situation on all sides. We should not blame the FreeBSD leadership for being noble and attempting to keep Matt's life private and not defame him.
Core team == Elitism? (Score:5, Insightful)
I kind of read the "elitist" things as assuming a bit much of the FreeBSD core developers as well. Did someone catch a member of the core team slipping and admitting "(Evil maniacal laugh) I have POWER I can commit to CVS, and YOU CANNOT!!! The world is MINE!!" I think you're reading too much; a core developer is just someone who can be trusted to code right and keep things consistent. I don't know were these assumptions of ego trips are coming from. If you want ego trips in BSD, I think OpenBSD with the one guy at the top who started it because he didn't work and play well with others and rather take his ball and play on his own court is a better place to look.
Oddly enough, I didn't see one post about Apache. Apache has pretty much the same development style, with a core of developers that have to approve or can veto a patch. Why isn't anyone complaining about the way that Apache is managed? It's usually held as the shining star of OpenSource, so why is the model for Apache good when a very similar model for FreeBSD is "elitist" and needs to be changed? Hmm, I'd say that some folks say this because FreeBSD is "competition" for Linux and Apache is not, but that would mean I think people waste their times battling one free stable x86 Unix workalike against another free stable x86 Unix workalike and I'd like to assume better than that.
One reason I like FreeBSD is because of that cohesiveness. You lose some speed in getting new features, but you also don't get some of the hiccups. In the 2.4, stable kernel, there's been a VM switch, a scheduler switch, and a change to the VFS subsystem that caused a corruption in the default journaling filesystem (though not with the default configuration of it). You get new features (Linux has had better SMP for forever) but you might bump into problems. The choice is yours. Who the hell said it was a zero-sum game? Oddly enough I haven't read "If you run FreeBSD on any of your servers you can't run Linux and RMS is going to come over to kick your ass" in the GPL, nor have I read "download one RedHat
And if anyone says that the Linux model has proven itself better by virtue of it's success, I'd just say: 1) what's your definition of success, the FreeBSD developers seem quite happy thank you and 2) Linux didn't have to fight a licensing suit (and win it) against ATT/Novel just when it was picking up steam.
devlopment community vs. public (Score:5, Insightful)
The two are not the same. I am a consumer of FreeBSD, by virtue of having a Virtual BSD Server from aplus.net. My use of that operating system in no way entitles me to know what transpires between the developers of that OS.
If I want to know the nitty gritty details of OS development, then I need to subscribe to the general mailing lists, read the code, and submit my own work.
Since I'm not prepared to do the above - I am quite happy to be a mere consumer in this case - I don't have any objection to people saying "this is a private matter, it doesn't concern you."
That the source code is available for your perusal is completely unrelated to the behavioral dynamics which govern the production of that code.
Re:dillon leaves the FreeBSD project (Score:5, Insightful)
Free Software is not necessarily developed publicly. Every project is going to involve at the very least a few e-mails back and forth in private between people. And maybe that evolves into a more formal core team.
So what? Not everything is meant to be aired in public. You act as though you have some "right to know." Who or what twisted your mind into thinking that it is your "right" to stick your nose in everybody's business?
I think you need to go turn off the television and stop reading Slashdot for a while.
NOTE: Of course I was hoping someone kindly leaked the details in the comments, but I'm just curious. I don't expect to be given the details just because I have a pulse.
Re:dillon leaves the FreeBSD project (Score:3, Insightful)
Now you make it sound like you are friends ("It's always sad to have to make these decisions.") and this move is something radical and/or important ("It's even more difficult to defend them when our hands are tied behind our backs.") when coding for free distros is freetime hobby for most of us. "..hands are tied behind our backs." yeah, right. Your hands are never tied behind your back unless you do it to yourself. Just do what you like and explain why if you want, but don't come out telling that your hands are tied. That's lame attempt to put the blame on "the system". We have enough people around already weeping that same I'm-doing-my-job, my-hands-are-tied song.
"Nothing is as important as gardening and even that isn't very important" -some wiseguy somewhere
Matt Dillon is a hack (Score:2, Insightful)
"If/when Luigi fixes the ABI problems with IPFW, we can remove this 'hack'. Until then, I do not consider the hackiness nature of the patch sufficient reason to not put it in."
I never saw one brilliant remark from this guy and by reading his interviews and his design documents I could only conclude that Mr Dillon was so incredibly blind and biased towards "there is only one solution to all problems and that's my solution". This remark quoted above is the most utterly stupid thing you can do as a developer, because it proves that the no.1 reason why software sux so much most of the time so true:
Nothing lasts that long as a temporary solution.
Think about it. When was the last time you hacked something in "just to patch this problem for now, I'll fix it later", while you never fixed it properly? The 'solution' Mr. Dillon wants to commit into the tree is a temp-solution. You shouldn't commit temporary CRUD into a development tree. Period.
Re:Matt Dillon is a hack (Score:4, Insightful)
When you're working on a large project with tight deadlines, sometimes you have to put in hacks. Sure, it hurts to do it, but sometimes there is just no choice. The quote you use suggests one of these situations: there's a bug in a piece of code you have no control over, and no influence over. You however are dependent on it to provide some funtionality you _must_ deliver. Only choice is to put in a workaround, and of course lean on the appropriate developer (very) heavily to fix their code pronto.
Now I've never worked on a large O/S project, so maybe he could just have gone in and fixed the IPFW (whatever that is) himself, in which case this argument is moot, but in the world I work in people own code and they don't like other people fiddling with it.
Re:Matt Dillon is a hack (Score:3, Insightful)
I would have to heartily support that.
For a critical feature in a large project, having that feature in a workable state is more important during on-the-fly development and testing than having it in a perfect state. Perfection can be achieved later with a little discipline. Having it non-functional is just time-consuming.
The fact that it's considered a crude hack is all the more motivation to fix the code so you can remove it.
Re:The flamewar is here: (Score:2, Insightful)
Now I can understand why the CORE developers don't want to have us dogs snooping on their list. Because it's obvious that Matt was getting extremely frustrated and when he decided to install a kludge to prevent himself from losing hours to recompile time, he was shot down for it. Or offered the magnanimous option of "fixing it himself."
If my air bag went off every time I went driving, and I couldn't figure out why and neither could Mazda, but I found a way to at least prevent it from going off (but not fixing the real problem), should Mazda prevent me from buying a car again? Yeah, it's a Straw Man. It's an analogy. Get over it.
Re:dillon leaves the FreeBSD project (Score:4, Insightful)
The Linux kernel developed in a more "freewheeling" way. Some say that Linux had no formal "core," but it did. A core of one. Linus himself. He used to maintiain the kernel himself (without version control, I might add).
Different development groups do things differently. We can bicker and argue all day over whether we would have included or excluded a given developer for doing a given thing. It doesn't matter. Dillon did what he did. The "core" did what they did. That's all there is to it. If this really offends you, take the FreeBSD source code and launch your own BSD project (it has happened before). I doubt you will succeed, but you may. And someday people will be talking about OpenBSD and GeekBSD or SlashBSD and no one will remember FreeBSD. Until then, all us slashbots carping do not mean a thing.