Recent Features
| |||||||||||||
Ubuntu and governanceUbuntu and governancePosted Oct 5, 2007 18:19 UTC (Fri) by rickmoen (subscriber, #6943)In reply to: Yet another male perspective on women in free software by tajyrink Parent article: Yet another male perspective on women in free software tayjrink wrote: The reason I like Ubuntu is not just the sleekness and polishness (Fedora and openSUSE do better in some/many ways, not least with the system configuration tools), but the fact that it's not just "hey whee source code to the latest kernel", but "we're building a more equal world where people help each other and do stuff for the common good". Plus the Code of Conduct that may come to rescue whenever someone is being an idiot on the forums / mailing lists. There are actually two Codes of Conduct, the regular one and the leadership one. Both are vague feel-good suggestion lists, rather than the functional code of conduct that they purport to be. Is Ubuntu going to enforce the "Be collaborative" and "When you are unsure, ask for help" items from the regular code against members, e.g., lock someone out of an IRC channel for failing to be collaborative? As an actual code of conduct, i.e., a set of governing documents for participants, it's a joke. The relevant Web pages do allow one to discover the actual governing structure, if one disregards the fluff: Technical decisions are vouchsafed to a Technical Board, and member-conduct discipline to a Community Council. Both meet on IRC; neither is particularly transparent to the public (i.e., the public cannot readily track what their history of decision-making is, or their real process and criteria for doing so), for three reasons: 1. The IRC sessions are not logged and made available in the general case. 2. Members of both committees reserve the right to retreat to a private channel if in their judgement "an open meeting becomes too noisy". 3. It is likely that most of both groups' business occurs on their respective strictly private and confidential mailing lists. And who are the members of those groups? They are appointed by Mark Shuttleworth, confirmed by majority vote of the maintainers, and serve for one and two years respectively. To whom are they accountable? Nobody in particular. (I'm sure they're good people, and are not criticising them. I'm just pointing out the process.) The end-result is that Ubuntu has two private committees under the founder's proximate control that have roughly zero public transparency, one of which is fully empowered to restrict or ban Ubuntu participants without needing to explain themselves to anyone but each other. Anyone trying to find a reliable record of what they've done and why will be stymied by the surrounding information void. To the extent there's anything, there are just vague and, frankly, deceptive-in-context handwaves about the Ubuntu Code(s) of Conduct. Now, it's entirely possible if not likely that the end-results are mostly desirable, e.g., incorrigible personal flamers and posters of large amounts of offtopic drivel being exiled temporarily or permanently. However, the process is not (per above) particularly honest. Me, I prefer honest. Rick Moen
(Log in to post comments)
Ubuntu and governance Posted Oct 5, 2007 22:39 UTC (Fri) by mjg59 (subscriber, #23239) [Link] The IRC sessions are not logged and made available in the general casehttp://people.ubuntu.com/~fabbione/irclogs/ Members of both committees reserve the right to retreat to a private channel if in their judgement "an open meeting becomes too noisy" I'm not aware of this ever happening - it certainly hasn't for the technical board. In the event of an open meeting becoming too noisy, it'd be far more likely for us to moderate the channel and continue in public, before unmoderating the channel to allow for general feedback (I have a vague recollection of this happening at some stage) It is likely that most of both groups' business occurs on their respective strictly private and confidential mailing lists. It doesn't. The only issues generally discussed on the mailing list are either procedural (approving decisions made by other bodies) or private at external request (in the technical board case mostly upstream authors asking for decisions about their software). They are appointed by Mark Shuttleworth, confirmed by majority vote of the maintainers, and serve for one and two years respectively. It's true - Ubuntu isn't a democracy. I don't think we've ever tried to hide that. Nor do I think anyone could ever accuse Mark of always making good decisions. But that's not really the point. The structure of Ubuntu was designed precisely to avoid the issues that keep on arising in more democratic organisations like Debian, such as protracted arguments about what sort of social standards are acceptable or the precise meaning of the wording of a document written by someone who's left the project years ago or what if I were a lesbian amputee from Mars or whatever. We have bodies of people who are chosen on the basis of competence and agreement with the existing social and technological standards, with the explicit aim of ensuring that those same standards are perpetuated. This sort of arrangement certainly isn't for everybody, and I'm certainly not going to try to convince you that you're wrong about Ubuntu. I'm just interested in clarifying why we have this set of arrangements. The end-result is that Ubuntu has two private committees under the founder's proximate control that have roughly zero public transparency, one of which is fully empowered to restrict or ban Ubuntu participants without needing to explain themselves to anyone but each other. While that's certainly theoretically true, I suspect that it would never happen. If we're unable to explain precisely why someone was excluded, the community would be asking serious questions. On the one hand, the large scale community involvement with Ubuntu aids the distribution by producing huge quantities of free advertising - on the other hand, they also provide a very public counterreaction if we fuck up. So really, we're left with the case where someone gets thrown out, there's an inadequate explanation and the community doesn't care. That kind of suggests it was the right decision, though the lack of explanation would be justifiably criticised. Is Ubuntu going to enforce the "Be collaborative" and "When you are unsure, ask for help" items from the regular code against members, e.g., lock someone out of an IRC channel for failing to be collaborative? But to answer your original question - if someone's long-term failure to be collaborative or ask for help results in them adversely affecting other people's attempts to contribute, even after requests for them to modify their behaviour, then yes, they'll probably be expelled from the project. The slightly handwavy attitude of the code of conduct is deliberate. We don't want to attempt to strictly codify behaviour, because doing so just encourages people to try to find loopholes - see the constant gamesmanship applied to the OSD, for instace ("Oh, but this clearly actually means this, so it's fine if we require you to urinate on your chair every time you want to run the software!") In an ideal world, the code of conduct would be something along the lines of "Behave sensibly". It ended up containing some broad guidelines to give people a better idea of what we think "sensible" is. However, the process is not (per above) particularly honest. Me, I prefer honest. While I know that I said I wasn't trying to convince you that Ubuntu isn't evil or anything, I think it's worth me mentioning this. I know most of the people involved in the Ubuntu governance structure fairly well, both at a professional and personal level. I trust them to make correct decisions. However, if anyone sees anything that could be construed as misappropriate happening in the decision making process, feel free to get in touch with me. If I'm not able to provide a decent answer about what's happened, then I'll happily resign - at that point, it'll be fairly clear that my standards don't match those of the rest of the distribution.
Ubuntu and governance Posted Oct 10, 2007 21:43 UTC (Wed) by rickmoen (subscriber, #6943) [Link] mjg59 wrote:http://people.ubuntu.com/~fabbione/irclogs/ Good for Fabio M. Di Nitto. However, (1) people looking for logs at the obvious places aren't going to find them there, (2) the appear (per URL) to be an individuals, rather than an official offering of the committee's, (3) other people have no idea whether they're complete and accurately represent what occurred, and (4) even if Fabio does manage to log everything on an ongoing basis, for other reasons described (private channel and mailing list), they might well not capture the discussions generally of interest, anyway. I'm not aware of this ever happening Irrelevant to the point. It doesn't. When I said "most", I of course meant "most significant". The provision for as-needed resort to explicitly private and explicitly confidential discussion tends to speak for itself. I'm not in any way criticising the Ubuntu Project for having such mechanisms. I'm merely saying that these and the other things described are the real mechanisms for implementing conduct discipline, and that the Code(s) of Conduct is (are) not substantially relevant to process. It's true - Ubuntu isn't a democracy.[...] Again, I voiced no criticism. Ubuntu Project does not need to justify its organisational form. Again, I merely detailed what is real, in the area of conduct enforcement, and what is not. I'm certainly not going to try to convince you that you're wrong about Ubuntu. Actually, if my description is inaccurate, please do feel welcome to clarify. However, you should probably start by re-reading what I wrote, since obviously you inferred in error that I was criticising the Ubuntu Project's core nature, in which reading you erred, badly. You basically ignored the substance of my post completely, and responded with a completely irrelevant non-response about Ubuntu Project's reasons for its organisational form. This is mildly vexing, because I was, I think, rather clear in my earlier post, so I infer that you must not have tried very hard to read before posting. But to answer your original question - if someone's long-term failure to be collaborative or ask for help results in them adversely affecting other people's attempts to contribute, even after requests for them to modify their behaviour, then yes, they'll probably be expelled from the project. I'm somewhat embarrassed, and saddened, to have to report this news: The question was 100% rhetorical. It is an evident absurdity to seriously maintain that an open-source project's conduct rules actually require that participants "Be collaborative" and "When you are unsure, ask for help", and that disciplinary proceedings leading to sanctions up to and including expulsion can result from detected violations. If you are honestly saying that you cannot figure out why, then I fear we have no basis for discussion. While I know that I said I wasn't trying to convince you that Ubuntu isn't evil or anything, Again, you have very badly misread, if you inferred that I was saying that Ubuntu Project or anyone else is "evil" or any other such characterisation. ...free to get in touch with me... That might be more likely if you would let people know who you are, and give some clue concerning your relationship, if any, to the Ubuntu Project. Rick Moen
Ubuntu and governance Posted Oct 10, 2007 22:46 UTC (Wed) by mjg59 (subscriber, #23239) [Link] Google has little difficulty in finding a link to the IRC logs, but you're right that it wasn't obviously linked from any of the official documentation. I've rectified that.I'm merely saying that these and the other things described are the real mechanisms for implementing conduct discipline, and that the Code(s) of Conduct is (are) not substantially relevant to process Which isn't the case at all. When any issues have arisen they've been discussed publically. We don't hold court on private mailing lists or in IRC backchannels. Except in cases where the participants have explicitly requested that discussions take place in private, every decision of any note whatsoever has taken place in public and been (where relevant) judged against the code of conduct. Actually, if my description is inaccurate, please do feel welcome to clarify You were wrong about the lack of IRC logs, the forums where decisions actually end up getting made, the lack of transparency, the lack of a reliable record and the lack of honesty. Personally, I'd tend to interpret a claim that a process lacks honesty as a criticism of said process - accusations of dishonesty are a fairly common source of libel suits. evident absurdity Whether you feel it to be absurd or not, it remains true. People who do not act in a manner conducive to the development of a collaborative project are not welcome in the Ubuntu project. if you would let people know who you are Google gives a fairly good overview, but relevantly I've been a member of the Ubuntu technical board for just under two years now. For most of that time I've had no financial relationship with Canonical, though in the interests of disclosure I'm currently halfway through a three month contract with them.
Ubuntu and governance Posted Oct 11, 2007 5:01 UTC (Thu) by rickmoen (subscriber, #6943) [Link] I'm afraid I really must demur, again.mjg59 wrote: Google has little difficulty in finding a link to the IRC logs Irrelevant to the point as stated. Which isn't the case at all. But it is. Once again, your denial ignores the substance of what I said, and professes to refute some assertion I did not, in fact, make. You can, if you wish, read what I actually wrote -- or you're perfectly welcome to continue to ignore it. Fine with me, either way. You were wrong about the lack of IRC logs Excuse me, but here I am having an increasingly difficult time extending the spirit of charity, and continuing to assume you are merely having reading-comprehension problems, rather than deliberately misstating. I nowhere claimed IRC logs didn't exist at all. What I said -- and you have had ample opportunity to see it in plain ASCII -- is that "The IRC sessions are not logged and made available in the general case." The meaning was perfectly clear: People would not find them in any of the expected place, nor if they found them would have any reasonable assurance of them being accurate and complete. If by some rather improbable chance the point wasn't obvious from my initial post, it certainly should have been in the detailed followup that spelled it out explicitly. In any event, the hyperlink you've now supplied to http://people.ubuntu.com/~fabbione/irclogs/ (for which, thanks, by the way) purports to cover meetings of the Technical Board -- but the ones that would be relevant to member-conduct discipline would be those of the Community Council. So, your wiki edit to http://www.ubuntulinux.org/community/processes/techboard, though appreciated, is mostly irrelevant to the preceding discussion. (By the way, as a point of interest, which files inside your irclogs directory are those of the Technical Committee? Looking for, e.g., the string "tech" turned up nothing.) the forums where decisions actually end up getting made Nothing you said has claimed that decisions don't get made based on those discussions -- and, frankly, it would be very strange for them not to. Why else have private and confidential discussions, other than to help reach conclusions on delicate matters? the lack of transparency To the contrary, I showed that lack of transparency; you have merely talked around it. (Again, the gesture concerning Technical Committee IRC logs -- which I assume must be somewhere in there, is an appreciated exception, but came after my post, which, I hope you realise, illustrates the point.) the lack of a reliable record The precise wording was "Anyone trying to find a reliable record of what they've done and why will be stymied by the surrounding information void." You seem to have conceded this point, in allowing that, yes, there was nowhere on the Technical Committee pages, let alone the much more relevant Community Council ones, that lets one find relevant records -- and that, actually, there are no such records within those groups' pages at all. I don't consider this worth harping on, merely worth noting once and moving on. I would appreciate doing that now, and not be obliged to restate that simple point yet again, just because you don't get it. Google gives a fairly good overview, A more prosaic and less convoluted reply might have been "Oh, good point. I'm Matthew Garrett, member of the Ubuntu Technical Board and Ubuntu Kernel Team" [and whatever else might be relevant]. Your world timeline may differ, but in mine, LWN nicks just aren't really useful (in the general case) as identifiers of one's person and affiliations. Personally, I'd tend to interpret a claim that a process lacks honesty as a criticism of said process - accusations of dishonesty are a fairly common source of libel suits. Are you actually threatening me with a libel lawsuit, Mr. Matthew J. Garrett? If you wish to confirm that, I will be glad to provide you with contact information for my attorney, for service of process -- but our conversation will necessarily conclude at that point. Before you do, however, I'll point out -- once again -- what I actually said. Quoting: "Now, it's entirely possible if not likely that the end-results are mostly desirable, e.g., incorrigible personal flamers and posters of large amounts of offtopic drivel being exiled temporarily or permanently. However, the process is not (per above) particularly honest. Me, I prefer honest." I did not impute personal dishonesty to you, or to Mark Shuttleworth, or to Canonical, Ltd. I said it appeared to me that the process was dishonest, e.g., in asserting that project governance is run by straightforward application of the Code(s) of Conduct, when self-evidently that cannot be the case, because those two Codes are nothing at all like functional governance documents. For what it's worth, I am glad to say that I very much doubt you're in any way dishonest -- a hothead who doesn't bother to read attentively, absolutely, and who really should edit his posts a lot more carefully before sounding off in public, but I have no reason whatsoever to cast aspersions on your honesty. If and when you speak to your own attorney, by the way, please make sure to ask him whether a "process" (of, e.g., a Linux community project) has standing to bring libel litigation in court. The question will absolutely make his day. Rick Moen
Ubuntu and governance Posted Oct 11, 2007 6:39 UTC (Thu) by mjg59 (subscriber, #23239) [Link] Now, to a large extent I really don't want to get involved in this. It exemplifies precisely the style of discussion that I think is responsible for driving people away from open source communities. It's entirely needlessly hostile.Are you actually threatening me with a libel lawsuit No, of course not. I was attempting to illustrate my point that dishonesty would generally be regarded as a criticism, in response to your suggestion that you hadn't criticised the form of Ubuntu's governance. As you point out, suggesting that a process could be libelled is laughable. My apologies for any lack of clarity in that respect. "The IRC sessions are not logged and made available in the general case." They are logged. They are made available. You made an entirely justifiable criticism that it was not trivially obvious that they were made available, and I've attempted to rectify that (you're right about the Community Council logs being more interesting. The failure to link to those was an oversight on my part, and I've fixed it), but that doesn't alter the fact that anyone putting in any moderate amount of effort could find them - especially since they're linked from the wiki page entitled Meeting Logs, along with links to cleaned up versions (I'd assumed that you'd prefer the entirely unedited forms). If by "not made available in the general case" you meant "are slightly difficult to find" then I'd be forced to agree, but I think that's a slightly odd definition. Why else have private and confidential discussions, other than to help reach conclusions on delicate matters? As briefly outlined before:
those two Codes are nothing at all like functional governance documents Now, I have to say that I'm confused. We're clearly using different definitions of either "functional" or "governance" here, but it's not clear to me whether this is a cultural or technical misunderstanding. When comparing the Ubuntu Code of Conduct to, say, the Apache Corporation Code of Business Conduct, the differences appear to be in areas of coverage (hardly surprising) and the somewhat more general tone of the Ubuntu document. Both cover sets of acceptable behaviour. Both specify the body with responsibility for interpretation and enforcement. Both are designed to cover situations other than those explicitly enumerated within the document. With the exception of the reduced level of precision in the Ubuntu document, I can't determine any significant functional difference. The Ubuntu CoC is not intended to be a legal document or something that can be mechanically interpreted - it's a statement of what classes of behaviour will and will not be tolerated within the project community, and it has (and, I expect, will) been used to determine whether to exclude people from parts of the project where they fail to meet those expected levels of behaviour. In that respect, it's functional and it defines a certain level of governance.
Ubuntu and governance Posted Oct 18, 2007 5:48 UTC (Thu) by einhverfr (guest, #44407) [Link] I am afraid that if my experience of Rick is typical that he is both needlessly hostile and actively seeks to drive people he disagrees with away. I would point out one additional point that should be made. I don't generally find the "benevolant dictatorship" model to work for open soruce projects. Everyone wants to be that benevolant dictatorship but very few succeed...... At the same time, one needs only to look at the politics involve in Debian to see that democracy as such is not a great option either. The best system (which I believe that PostgreSQL, LedgerSMB, and Ubuntu share) is that of the non-representative (or meritocratic) republic. Who cares what everyone says, the core team does what is right for the project, and is structured to minimize the impact of conflicts of interest. You get the benefits of a dictatorship (freedom to do what is best for the project with a minimum of unnecessary politics) with some of the safeguards of a democracy.
Ubuntu and governance Posted Oct 11, 2007 21:57 UTC (Thu) by nix (subscriber, #2304) [Link] Good grief. Stop drinking the coffee, *really*. You're acting as if Mattis some sort of emissary of Satan.
People might treat you more nicely if you stopped reading everything they
Ubuntu and governance Posted Oct 13, 2007 2:18 UTC (Sat) by Duncan (guest, #6647) [Link] > The slightly handwavy attitude of> the code of conduct is deliberate.
I'm not sure if this is what Rick is/was after, but as I've often put it,
An example of routinely broken written rules is unionized work slowdowns.
It doesn't matter the organization. I've seen it all over, government,
On the logging, I'm inclined to agree with you. They could have been
OTOH, the existence of private channels, while convenient and as you
As for why lawyers were even mentioned, ugh. An (I believe) intended
Duncan
Ubuntu and governance Posted Oct 18, 2007 5:39 UTC (Thu) by einhverfr (guest, #44407) [Link] I am not an Ubuntu user, but LedgerSMB adopted the Ubuntu Code of Conduct early in its life (over a year ago). WHile I don't know how Ubuntu governance works, or how well it works, I will say that these codes of conduct have been helpful primarily because it provides a generic reference to the request to behave. In general, our core team has not been as transparent as it could be, but that has changed. There are times and places where transparency is not appropriate (for example, in responding to security reports), and it has taken some time to sort out what works and what doesn't. But in general, this is a good thing. BTW, in our first year, we only had to remind one person to follow the code of conduct once. I think what you are missing, Rick, is that you want a code of conduct to flexibly embody a set of principles and ask people who are outside any specific conflicts to decide whether to issue a public warning based on moving away from those principles. (This distinction is important. If someone personaly attacks me, I won't bring up the code of conduct-- I will let others on the core team do that. This avoids the perception of a conflict of interest or hiding behind the rules when it is convenient.)
|
|||||||||||||