From mark.reinhold at oracle.com Thu Feb 3 08:56:08 2011 From: mark.reinhold at oracle.com (mark.reinhold at oracle.com) Date: Thu, 03 Feb 2011 17:56:08 +0100 Subject: OpenJDK Community Bylaws and Governing Board Message-ID: <20110203165608.D9D362904@eggemoggin.niobe.net> A draft of the OpenJDK Community Bylaws is now available [1]. After IBM and then Apple joined the OpenJDK Community last fall it became clear that it was time to revive the effort to create a written set of rules by which the Community will operate. This document was written in collaboration with John Duimovich and Jason Gartner of IBM, Prof. Doug Lea of SUNY Oswego, Mike Milinkovich of Eclipse, and Adam Messinger of Oracle. We didn't start this effort from scratch. The draft draws upon the earlier work of the Sun-chartered OpenJDK Interim Governance Board and has exactly the same goal: To be a written set of rules that will foster the long-term health and growth of the OpenJDK Community by enabling and encouraging its members to act in an open, transparent, and meritocratic manner. We've attempted to define efficient, lightweight rules that are aligned with how the Community operates today, under the interim guidelines for Groups and Projects [2,3] established in 2007. This is a starting point, not a done deal. There are numerous bugs and missed corner cases in this draft and, no doubt, even more numerous ways in which it can be improved. Please send questions, comments, and suggestions to the gb-discuss list [4]. (If you haven't already subscribed to that list then please do so first [5], otherwise your message will be discarded as spam.) I'll revise the draft in response to comments over the next four weeks, until 3 March 2011. After the draft is finished it will be offered to OpenJDK Community members for ratification via an appropriate democratic process. The Bylaws define a Governing Board which "oversees the structure, operation, and overall health of the Community. It upholds and maintains these Bylaws, resolves procedural disputes, and ensures that sufficient infrastructure is available to Community members. The Governing Board has no direct authority over technical or release decisions." The members of the initial Governing Board are: - Adam Messinger (Chair, Oracle), - Jason Gartner (Vice Chair, IBM), - Prof. Doug Lea (At-Large, SUNY Oswego), - Mike Milinkovich (At-Large, Eclipse), and - Yours truly (OpenJDK Lead, Oracle). At-Large members serve for a term of one calendar year. After the first year they'll be nominated and elected by a democratic process open to all key members of the Community. - Mark [1] http://cr.openjdk.java.net/~mr/bylaws/draft-openjdk-bylaws-07.html [2] http://openjdk.java.net/groups [3] http://openjdk.java.net/projects [4] gb DASH discuss AT openjdk DOT java DOT net [5] http://mail.openjdk.java.net/mailman/listinfo/gb-discuss From mark at klomp.org Thu Feb 3 12:21:42 2011 From: mark at klomp.org (Mark Wielaard) Date: Thu, 03 Feb 2011 21:21:42 +0100 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <20110203165944.6B528AC4@eggemoggin.niobe.net> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> Message-ID: <1296764502.3341.87.camel@springer.wildebeest.org> Hi Mark, On Thu, 2011-02-03 at 17:59 +0100, mark.reinhold at Oracle.com wrote: > A draft of the OpenJDK Community Bylaws is now available and an initial > Governing Board has been named. For further information please see: > > http://mail.openjdk.java.net/pipermail/gb-discuss/2011-February/000078.html Thanks for publishing those. Here are some comments and concerns. > The Bylaws define a Governing Board which "oversees the structure, > operation, and overall health of the Community. It upholds and maintains > these Bylaws, resolves procedural disputes, and ensures that sufficient > infrastructure is available to Community members. There is no definition what infrastructure. It also doesn't state where this infrastructure comes from or who maintains it. > The Governing Board > has no direct authority over technical or release decisions." But there is a technical Appeals Process. > The members of the initial Governing Board are: > > - Adam Messinger (Chair, Oracle), > - Jason Gartner (Vice Chair, IBM), > - Prof. Doug Lea (At-Large, SUNY Oswego), > - Mike Milinkovich (At-Large, Eclipse), and > - Yours truly (OpenJDK Lead, Oracle). What happened to the previous board members? Fabiane, Simon, Dalibor, Andrew and Martin? > The OpenJDK Community is an association of developers who collaborate > upon open-source implementations of present and future versions of the > Java Platform, Standard Edition, as de?ned by the Java Community > Process As you know the current platform JSRs have licenses for the spec, ri and tck published by the JCP under terms that are in conflict with the GPL. It would be wise to resolve this before it gets people in trouble. I published a bit more analysis about this on my blog: http://gnu.wildebeest.org/blog/mjw/2010/11/28/moving-java-forward-through-the-jcp/ This is a bit of a blocker for progress if that issue isn't resolved first. But I am happy it does explicitly mention that it is a collaboration upon open-source implementations. It should also make sure that the minimum (GPL) copyright, trademark and patent grants that the project participants will collaborate under are explicitly defined, so people know they can rely on them. > A Participant is an individual who has subscribed to one or more > OpenJDK mailing lists. A Participant may post messages to a list, > submit simple patches, and make other kinds of small contributions. > > A Contributor is a Participant who has signed the Oracle Contributor > Agreement, or who works for an organization that has signed that > agreement or its equivalent and makes contributions within the scope > of that work and subject to that agreement. Only a Contributor may > submit anything larger than a simple patch. > > An OpenJDK Member is a Contributor [...] This effectively says that only those who assign all rights on their contributions to Oracle can be Members. That doesn't seem fair. Oracle is just one of the Participants in the community. Aggregating legal rights isn't a bad thing perse, but it is if it is done by an entity that then gets to assert more rights than the rest of the community on non-reciprocal terms. Again I have written why this is troubling about this a bit more in my blog: http://gnu.wildebeest.org/blog/mjw/2009/11/14/trusting-companies-with-your-code/ You should at least fix things so that either people can be members without having to assign all their rights on non-reciprocal terms to Oracle. > Projects vs Groups Personally I think it would be better to match Projects and Groups one on one. Projects have a clear "lifetime", at a certain point they don't produce anything new anymore (commits stop, nobody maintains things anymore, no users). Then they can be archived and garbage collected. That seems a healthy thing to have. Groups are a little too "vague" in my view. Although this is not a very strong argument I admit. This mainly comes from how a project seems defined by whether or not a group keeps it running. That doesn't feel right to me. A project should be able to control its own "destiny". Also because the Group Lead and the Project Lead seem to mostly overlap, so why not merge? But maybe this is the intention and I am just utterly confused about the Groups/Projects splitup. > New releases of Java SE Platform implementations are Projects, though > particularly large ones. Such JDK Release Projects may only be > proposed by the OpenJDK Lead and may only be Sponsored by the > Governing Board. Why this distinction and what defines "particularly large"? > The Governing Board consists of ?ve individuals: > * The Chair, appointed by Oracle; > * The Vice-Chair, appointed by IBM; > * The OpenJDK Lead, appointed by Oracle; and > * Two At-Large Members, nominated and elected as described below. It doesn't seem right that the first three say they are appointments. If you want to involve corporations as directly involved then at least make them only nominate someone that is actually involved in the project. Why Oracle and IBM? The previous board had engineers from Red Hat and Google who are the biggest contributors to OpenJDK after Oracle. And there is only a minority of independent candidates if you let companies nominate/appoint seats. > The Governing Board is, in part, a legislative body: It is empowered > to revise these Bylaws to re?ne existing processes, to de?ne new > processes, and to dispose of processes that are no longer required. > Any revision of these Bylaws must be approved by an Absolute > Two-Thirds Majority of the Governing Board. Shouldn't such major changes also be voted on by all Members? > During a two-week nomination period any OpenJDK Member may nominate an > individual who does not currently hold an appointed Governing Board > seat to ?ll one of the At-Large seats. That individual need not > already be an OpenJDK Member. This gets around the issue of Members having to assign all rights to Oracle to participate in the community, so that is good. But it is a little strange to have the board consist of non-members. > Once per calendar quarter, and one week prior to that quarter?s > scheduled meeting of the Governing Board, the OpenJDK Lead shall > publish a written report summarizing recent activities in the > Community. This report should include: [...] > > * A list of Projects that have made major state changes such as > publishing a release, integrating into a JDK Release Project, > or submitting or completing a JSR; > > * A list of Projects that should, in the OpenJDK Lead?s opinion, > be considered for inclusion in a future JDK Release Project > and its corresponding Umbrella JSR; As mentioned above it seems bad to tie Projects to JSRs until it is clear that the JCP will have rules to make sure that JSR spec, RI and TCK licenses will be published under GPL-compatible terms so they can be used by OpenJDK participants. > Technical Appeals Process Why is this included when the bylaws also state that the Governing Board has no direct authority over technical or release decisions? Hope that helps see where the blockers are and how to resolve them. Cheers, Mark From fabiane at tridedalo.com.br Thu Feb 3 14:05:43 2011 From: fabiane at tridedalo.com.br (Fabiane Bizinella Nardon) Date: Thu, 03 Feb 2011 20:05:43 -0200 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <1296764502.3341.87.camel@springer.wildebeest.org> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> Message-ID: <4D4B26B7.2080704@tridedalo.com.br> On 2/3/11 6:21 PM, Mark Wielaard wrote: > What happened to the previous board members? > Fabiane, Simon, Dalibor, Andrew and Martin? > The interim governing board, from which I was part of, was supposed to be dissolved in May of 2008. There was later an amendment [1] stating that the interim governing board should be dissolved in May of 2009. As expected, the transition from Sun/Oracle stalled the work we were doing previously on the board. Although until now there was no official board to replace the interim board, I, for the last months, believed that my term at the board was effectively finished. So, it's good to see that a new board will move things on. Fabiane [1] http://openjdk.java.net/legal/charter/#a1 From gnu_andrew at member.fsf.org Thu Feb 3 14:38:31 2011 From: gnu_andrew at member.fsf.org (Dr Andrew John Hughes) Date: Thu, 3 Feb 2011 23:38:31 +0100 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <20110203165944.6B528AC4@eggemoggin.niobe.net> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> Message-ID: On 3 February 2011 17:59, wrote: > A draft of the OpenJDK Community Bylaws is now available and an initial > Governing Board has been named. ?For further information please see: > > ?http://mail.openjdk.java.net/pipermail/gb-discuss/2011-February/000078.html > > - Mark > I pretty much agree with everything Mark Wielaard just said, but here are my initial impressions from reading this draft. I'd be happy to be proved wrong on the assumptions herein. * 'The OpenJDK Lead is responsible for the overall technical direction and activities of the major efforts within the Community, and for the openness and transparency of the development process. The OpenJDK Lead sits on the Governing Board.' and 'The Governing Board is not an executive body: It has no direct authority over technical or release decisions; that authority is held by the OpenJDK Lead. ' So the project's direction is dictated not by the community, but by one person appointed not by consensus but by Oracle. Not good. * 'A Participant is an individual who has subscribed to one or more OpenJDK mailing lists. A Participant may post messages to a list, submit simple patches, and make other kinds of small contributions.' So we can finally have trivial patches go in without an OCA? Nice to see some progress. On the flipside, I think Mark already outlined pretty clearly the whole issue with having higher positions that this dominated by copyright assignment to one corporate entity. * 'An OpenJDK member is a Contributor who has demonstrated a history of signi?cant contributions to the Community' Who determines 'significant'? This seems essential, given effectively these members get control over most of OpenJDK as a whole (voting rights, etc.) *' * The Chair, appointed by Oracle; * The Vice-Chair, appointed by IBM; * The OpenJDK Lead, appointed by Oracle; and * Two At-Large Members, nominated and elected as described below. A board dominated by Oracle, with a position for IBM who have so far contributed next to nothing, and two members chosen by a vote from 'OpenJDK members', presumably most of which will be Oracle employees. * 'The members of the initial Governing Board are: - Adam Messinger (Chair, Oracle), - Jason Gartner (Vice Chair, IBM), - Prof. Doug Lea (At-Large, SUNY Oswego), - Mike Milinkovich (At-Large, Eclipse), and - Yours truly (OpenJDK Lead, Oracle).' Correct me if I'm wrong, but I've only seen contributions from two of these five to OpenJDK (you and Doug). Why have they been given these positions? Why should they get to make decisions over people who have actually done work on OpenJDK? What happens if we don't want to abide by these rules? Do we not get to join the OpenJDK club? This seems less like progress than a step backwards into proprietary JDK land. -- Andrew :-) Support Free Java! Contribute to GNU Classpath and the OpenJDK http://www.gnu.org/software/classpath http://openjdk.java.net PGP Key: F5862A37 (https://keys.indymedia.org/) Fingerprint = EA30 D855 D50F 90CD F54D ?0698 0713 C3ED F586 2A37 From simon at webmink.com Thu Feb 3 16:38:29 2011 From: simon at webmink.com (Simon Phipps) Date: Fri, 4 Feb 2011 00:38:29 +0000 Subject: Governance Evaluation Message-ID: As an outgoing member of the v1 IGB I've posted comments on the new governance draft on my blog, at http://webmink.com/2011/02/04/rating-openjdk-governance/ and will be pleased to discuss them either on-list or at FOSDEM this weekend. Let me know any other way I can help. S. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.openjdk.java.net/pipermail/gb-discuss/attachments/20110204/7d8c88e9/attachment.html From dl at cs.oswego.edu Fri Feb 4 10:12:50 2011 From: dl at cs.oswego.edu (Doug Lea) Date: Fri, 04 Feb 2011 13:12:50 -0500 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <1296764502.3341.87.camel@springer.wildebeest.org> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> Message-ID: <4D4C41A2.9070207@cs.oswego.edu> On 02/03/11 15:21, Mark Wielaard wrote: > Here are some comments and concerns. Here are a few followups. My main hope is that, for present purposes, people will focus on the bylaws themselves, and not on related but distinct issues such as revising contributor agreements. These also need attention, but first things first. > >> The Bylaws define a Governing Board which "oversees the structure, >> operation, and overall health of the Community. It upholds and maintains >> these Bylaws, resolves procedural disputes, and ensures that sufficient >> infrastructure is available to Community members. > > There is no definition what infrastructure. I recommended that we not enshrine a particular infrastructure in the bylaws. However perhaps this could be minimally specified by replacing "sufficient infrastructure" with "code repositories"? (See also last item below). >> The Governing Board >> has no direct authority over technical or release decisions." > > But there is a technical Appeals Process. I agree that things would be clearer with the sentence removed, because the preceding sentence (quoted above) pretty much covers everything. The GB deals with *procedural* disputes. The considerations entailed in resolving them aren't limited in any way. > >> The OpenJDK Community is an association of developers who collaborate >> upon open-source implementations of present and future versions of the >> Java Platform, Standard Edition, as de?ned by the Java Community >> Process As you might expect from me or anyone else who understands that the JCP is an irrelevant body, I was OK with its reference here only because I don't know of another way to say what JavaSE is without wasting pages of boring description. > As you know the current platform JSRs have licenses for the spec... I don't think this directly impacts bylaws (but again is an issue in need of attention). Different Projects may have different rules for review, commit, test and release. OpenJDK is a project -- the only one with a hard-wired lead. > >> The Governing Board consists of ?ve individuals: >> * The Chair, appointed by Oracle; >> * The Vice-Chair, appointed by IBM; >> * The OpenJDK Lead, appointed by Oracle; and >> * Two At-Large Members, nominated and elected as described below. > > It doesn't seem right that the first three say they are appointments. I didn't like this much at first. But notice from the various Voting Rules clauses, that having more than one dissent is generally enough to kill "important" votes. And starting with a small GB to oversee the OpenJDK reboot seems to be the only way to get it done any time soon. So I'm not convinced that it should be changed for initial bylaws. > > ... rules to make sure that JSR spec, RI and > TCK licenses will be published under GPL-compatible terms so they can be > used by OpenJDK participants. > Note that you could equally have said that repositories must make available the code, javadocs (aka specs) and tests necessary for Contributors to contribute, in which case I agree. -Doug From tinjon at gmail.com Fri Feb 4 17:06:02 2011 From: tinjon at gmail.com (Jonas Tingeborn) Date: Sat, 5 Feb 2011 02:06:02 +0100 Subject: OpenJDK Community Bylaws Message-ID: Feedback on the DRAFT 7 version as posted at http://cr.openjdk.java.net/~mr/bylaws/draft-openjdk-bylaws-07.html Section 2: Clarification needed "Only a Contributor may submit anything larger than a simple patch." This is too vague. If making a reference to a term, then that term needs to be self evident or defined and the meaning of a "simple patch" is unfortunately not self evident. Section 7: Missing mechanism or authority to revoke commit access. If the Project Lead lacks the authority to revoke commit access, then who has that authority? Section 7: Unclear community contribution process What mechanism or procedure exists for developers not affiliated with the cathedral outlined in the document, to provide patches and retain recognition for those contributions? A framing example should the question as stated not be recognizable to you follows. A common procedure employed by Committers of projects which manage their artifacts in the Git Version Control System (VCS) is to leverage the concept of pulling changeset contributions from developers not affiliated with the project. Through this simple concept, to not only push to a master repository, but also to pull from unofficial project branches created by non-project members, Committers are awarded a very efficient and straight forward process for incorporating outside changes they deem worthy and which meet their project's specific criteria for inclusion. Recognition is in the given example both awarded and also retained through VCS metadata, associated with each changeset and is incorporated when and if a Committer deems the contribution fitting, into the annals of project history upon his or her decision to push the changeset onto the official project artifact repository (VCS repository), Section 13: Clarification or rephrasing needed "Only three unsuccessful appeals by any particular Governing Board member are permitted in any twelve-month period." The sentence needs at best rephrasing, at worst be expanded. Was the word "unsuccessful" included by mistake? If not then the sentence makes no sense. An appeal raised has an indeterminate qualification prior to judgement and only after it has been judged can its qualification status be determined as successful or unsuccessful. If you meant to convey that each board member may object at most three times per year to a technical or release decision made by the OpenJDK Leader, then I believe the rationale or at least the logic behind it needs to be expanded because it is not intuitive nor understandable to a reader such as I, in contrast with the rest of the document's edicts. From mark at klomp.org Sat Feb 5 04:56:01 2011 From: mark at klomp.org (Mark Wielaard) Date: Sat, 5 Feb 2011 13:56:01 +0100 (CET) Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <4D4C41A2.9070207@cs.oswego.edu> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> <4D4C41A2.9070207@cs.oswego.edu> Message-ID: <58912.193.191.38.62.1296910561.squirrel@gnu.wildebeest.org> On Fri, February 4, 2011 19:12, Doug Lea wrote: > Here are a few followups. My main hope is that, for > present purposes, people will focus on the bylaws > themselves, and not on related but distinct issues > such as revising contributor agreements. These also > need attention, but first things first. Some of these things are crucial for effective governance. Ignoring those issues means you are crippling the community and board by design. I really like Simon's score card, you really need to make sure you address at least every item on that list to get a real community going. Or as someone said less politely, without addressing all issues, you are just setting up some nice rules for working in the sweat-shop. Seriously, if you are not addressing these things, please at least document the rational for not addressing these issue. And who is responsible for them if not the community of participants nor the governance board. >>> The Bylaws define a Governing Board which "oversees the structure, >>> operation, and overall health of the Community. It upholds and >>> maintains >>> these Bylaws, resolves procedural disputes, and ensures that sufficient >>> infrastructure is available to Community members. >> >> There is no definition what infrastructure. > > I recommended that we not enshrine a particular infrastructure > in the bylaws. However perhaps this could be minimally > specified by replacing "sufficient infrastructure" with > "code repositories"? (See also last item below). The reason I brought it up, is that it should be clear whether or not it is ultimately the governance board and by extension the community who controls what happens with the resources or not. If you are not in control, then just document that. >>> The OpenJDK Community is an association of developers who collaborate >>> upon open-source implementations of present and future versions of the >>> Java Platform, Standard Edition, as de???ned by the Java Community >>> Process > > As you might expect from me or anyone else who understands that > the JCP is an irrelevant body, I was OK with its reference here > only because I don't know of another way to say what JavaSE is > without wasting pages of boring description. :) But either it is an irrelevant body and then OpenJDK shouldn't be linked to it, or it is relevant, but then we have the problem that the chaos or legal issues around the JCP leak into OpenJDK. So, does OpenJDK implement/define JavaSE or not? It might be more constructive to admit the reality that OpenJDK is just the code base that is done in the open by the community and then others might or might not use that code base for defining JavaSE under their own separate rules. Just don't link them if the rules are in conflict. >>> The Governing Board consists of ???ve individuals: >>> * The Chair, appointed by Oracle; >>> * The Vice-Chair, appointed by IBM; >>> * The OpenJDK Lead, appointed by Oracle; and >>> * Two At-Large Members, nominated and elected as described below. >> >> It doesn't seem right that the first three say they are appointments. > > I didn't like this much at first. But notice from the various > Voting Rules clauses, that having more than one dissent is > generally enough to kill "important" votes. And starting with > a small GB to oversee the OpenJDK reboot seems to be the only > way to get it done any time soon. So I'm not convinced that > it should be changed for initial bylaws. Killing votes goes both ways. I just object to the appointed part. As others have said, it is perfectly clear why you and Mark are part of the initial board. The other appointed people however have no known interaction with the community. So why are the even on the board? There are better candidates to choose from from the current community and/or previous government boards. >> ... rules to make sure that JSR spec, RI and >> TCK licenses will be published under GPL-compatible terms so they can be >> used by OpenJDK participants. >> > > Note that you could equally have said that repositories must > make available the code, javadocs (aka specs) and tests necessary > for Contributors to contribute, in which case I agree. That would be a very good addition. Make sure that projects and/or contributions consist of everything necessary under free licenses, code, specifications and conformance tests. Thanks for the feedback, Mark From mike.milinkovich at eclipse.org Sat Feb 5 19:50:32 2011 From: mike.milinkovich at eclipse.org (Mike Milinkovich) Date: Sat, 5 Feb 2011 22:50:32 -0500 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <4D4C41A2.9070207@cs.oswego.edu> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> <4D4C41A2.9070207@cs.oswego.edu> Message-ID: <00a801cbc5b1$025550a0$06fff1e0$@milinkovich@eclipse.org> Doug, Mark, et al, A few comments below. > > There is no definition what infrastructure. > > I recommended that we not enshrine a particular infrastructure > in the bylaws. However perhaps this could be minimally > specified by replacing "sufficient infrastructure" with > "code repositories"? (See also last item below). My opinion is that "code repositories" would be worse than "infrastructure". You need a lot more than code repositories to run a project and community. If people are really worried about the lack of definition around "infrastructure", perhaps something like "...infrastructure, including such services as code repositories, issue databases, mail list servers and the like..." Personally I think that trying to define infrastructure is worse than just leaving it undefined, but that's just my opinion. Mark also raised the point that " It also doesn't state where this infrastructure comes from or who maintains it." I expect that the answer is Oracle. But I think that putting this level of detail into the Bylaws would be a mistake, as it is an execution matter not a governance rule. > >> The Governing Board > >> has no direct authority over technical or release decisions." > > > > But there is a technical Appeals Process. I think that the statement is actually correct as written, as it says "...no _direct_ authority...". The concept that is being captured is that the Board isn't going to be meddling in technical and release decisions being made by the projects and the OpenJDK Lead. Which I suspect we all believe is a good thing. Perhaps a solution to the confusion would be to say "Except as noted in 'Technical Appeals Process', the Governing Board has no direct authority over technical or release decisions." From mike.milinkovich at eclipse.org Sat Feb 5 20:24:34 2011 From: mike.milinkovich at eclipse.org (Mike Milinkovich) Date: Sat, 5 Feb 2011 23:24:34 -0500 Subject: Governance Evaluation In-Reply-To: References: Message-ID: <00aa01cbc5b5$c32e3950$498aabf0$@milinkovich@eclipse.org> Simon, I think your analysis is well and fairly done. Unsurprisingly, you have a good grasp of the constraints at play. There are a few areas that you have pointed out that I think we may be able to improve. Areas which come to mind are: - "Roadmap": A you've pointed out, capturing some baseline principles on how releases will be conducted would be helpful. There could be some controversy on this that I'm unaware of, but at face value it seems plausible that we could add something useful there. - Under "Modern License", you point out that the Bylaws are silent on the use of the GPLv2. However, the referenced Contributor Agreement does require the use of "...under a suitable FSF (Free Software Foundation) or OSI (Open Source Initiative) approved license". Does that address your concern, or do you think that needs to be included in the Bylaws? - On "Transparency", your points are valid. All I can say that that now that we have the draft in place that we can demonstrate to the community over time that transparency is a value that we all share. No matter how (im)perfect a governance document may be, in the end it is deeds, not words, which will shape the success of a community. One quibble is that in the first section labelled "Open, Meritocratic Oligarchy", I would point out that the small board in its current form actually has one advantage, which is that because a 2/3's majority is required for many major decisions it only takes two votes to block a mistake. That doesn't change your larger point, but it does imply that a useful check is in place against capricious or inhospitable decisions. From: gb-discuss-bounces at openjdk.java.net [mailto:gb-discuss-bounces at openjdk.java.net] On Behalf Of Simon Phipps Sent: February-03-11 7:38 PM To: gb-discuss at openjdk.java.net Cc: discuss at openjdk.java.net Subject: Governance Evaluation As an outgoing member of the v1 IGB I've posted comments on the new governance draft on my blog, at http://webmink.com/2011/02/04/rating-openjdk-governance/ and will be pleased to discuss them either on-list or at FOSDEM this weekend. Let me know any other way I can help. S. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.openjdk.java.net/pipermail/gb-discuss/attachments/20110205/1657c848/attachment.html From mike.milinkovich at eclipse.org Sat Feb 5 20:24:34 2011 From: mike.milinkovich at eclipse.org (Mike Milinkovich) Date: Sat, 5 Feb 2011 23:24:34 -0500 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: References: <20110203165944.6B528AC4@eggemoggin.niobe.net> Message-ID: <00a901cbc5b5$c2dee0d0$489ca270$@milinkovich@eclipse.org> Andrew, A few comments below: > * 'An OpenJDK member is a Contributor who has demonstrated a history > of signi?cant contributions to the Community' > > Who determines 'significant'? This seems essential, given effectively > these members get control over most of OpenJDK as a whole (voting > rights, etc.) My belief is that the existing OpenJDK Members will themselves define what 'significant' means. The only way to become an OpenJDK Member (after the initial start up phase) is to be nominated by an existing OpenJDK Member, followed by a Three-Vote Consensus vote of the existing OpenJDK Members. Which is a roundabout way of saying that the OpenJDK membership is intended to be a meritocracy in the normal sense of the word. Or at least that's what we were trying to say :-) > Correct me if I'm wrong, but I've only seen contributions from two of > these five to OpenJDK (you and Doug). Why have they been given these > positions? Why should they get to make decisions over people who have > actually done work on OpenJDK? I believe that I was asked to participate because I have a lot of practice governing a community where there are lots of companies involved, as well as many individual contributors. Time will tell whether this community believes that my contributions were helpful. I can tell you that I think that the Governing Board should have relatively little impact on the activities of individual projects. Under the draft Bylaws, I hope you agree that the projects themselves are self-governing projects in the normal free software or open source way. From tinjon at gmail.com Sun Feb 6 02:24:06 2011 From: tinjon at gmail.com (Jonas Tingeborn) Date: Sun, 06 Feb 2011 11:24:06 +0100 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <58912.193.191.38.62.1296910561.squirrel@gnu.wildebeest.org> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> <4D4C41A2.9070207@cs.oswego.edu> <58912.193.191.38.62.1296910561.squirrel@gnu.wildebeest.org> Message-ID: <4D4E76C6.3060800@gmail.com> On 2011-02-05 13:56, Mark Wielaard wrote: > Or as someone said less politely, without addressing all issues, you are > just setting up some nice rules for working in the sweat-shop. > > Seriously, if you are not addressing these things, please > at least document the rational for not addressing these issue. > And who is responsible for them if not the community of participants > nor the governance board. I specifically refrained from commenting on this one in my initial feedback since I wanted to get a clarification first on the rules surrounding objections made by the board members. My hope was that this objection power might act as a last resort should the OpenJDK Lead make harmful decisions regarding the product. However since I haven't gotten a response on that yet, I'll echo your concern that the document currently reads as a lot of flowery words for a community with no say what so ever over OpenJDK. You might as well draw a razor over the document and divide it in two parts. One part for people who like to talk about OpenJDK or execute the orders from the OpenJDK Lead ("the community") and another part for the OpenJDK Lead which has a single sentence "This role decides everything concerning OpenJDK such as technical direction, features and release schedules. Oh and by the way, the role is always to be appointed by Oracle.". I don't see much openness about this JDK under the proposed construct. >>>> The Governing Board consists of ???ve individuals: >>>> * The Chair, appointed by Oracle; >>>> * The Vice-Chair, appointed by IBM; >>>> * The OpenJDK Lead, appointed by Oracle; and >>>> * Two At-Large Members, nominated and elected as described below. >>> > Killing votes goes both ways. > I just object to the appointed part. > Another thing I refrained from commenting on since I figured that if either Oracle or IBM started pulling in their own direction, then they would have to sell (through logic or "other means") the At-Large members on the idea, which hopefully would serve as a containment mechanism for some of the company agendas. Though, I am concerned that specific companies are written into the bylaws. Wouldn't it be better to simply prop the roles with Oracle and IBM people when the bylaws take effect and then let people on the board earn their right to be there like the other groups? (ahm.. meritocracy someone said) Looking elsewhere, the Eclipse foundation has so far managed to keep their product alive without having IBM, Nokia or other specific companies written into their bylaws. From simon at webmink.com Sun Feb 6 05:49:34 2011 From: simon at webmink.com (Simon Phipps) Date: Sun, 6 Feb 2011 14:49:34 +0100 Subject: Governance Evaluation In-Reply-To: <00aa01cbc5b5$c32e3950$498aabf0$@milinkovich@eclipse.org> References: <00aa01cbc5b5$c32e3950$498aabf0$@milinkovich@eclipse.org> Message-ID: <7940C8AE-A8DE-4177-8BC7-0C78EB13498B@webmink.com> Hi Mike, thanks for your reply. On 6 Feb 2011, at 05:24, Mike Milinkovich wrote: > Simon, > > I think your analysis is well and fairly done. Unsurprisingly, you have a good grasp of the constraints at play. Thanks. As I said at FOSDEM, I think one of the key positive steps would be an FAQ where the reasons for the various compromises embodied in this document can be explained. For example, there's a reason why the main non-Sun contributor to OpenJDK, Red Hat, is not included in the draft at any point whereas IBM, who have historically never contributed to OpenJDK (despite being a Java powerhouse), have a permanent seat on the governing Board. It should be documented so that everyone understands where they stand and there's no need for back-channel explanation. Another positive step would be to track all the suggestions on this list in the bug-trackers so they are out in the open and none get overlooked. This has the additional advantage that when people want to discuss a point in depth they can do so in the bug tracker rather than on the mailing list. > > There are a few areas that you have pointed out that I think we may be able to improve. Areas which come to mind are: > > - ?Roadmap?: A you?ve pointed out, capturing some baseline principles on how releases will be conducted would be helpful. There could be some controversy on this that I?m unaware of, but at face value it seems plausible that we could add something useful there. I had some questions about this. To be clear, what I am suggesting is a release schedule for OpenJDK builds, together with a list of the contributions that will go into each future build. Having this allows every community member to see and participate in the process. I am not referring to the JCP-related roadmap for new language features. > - Under ?Modern License?, you point out that the Bylaws are silent on the use of the GPLv2. However, the referenced Contributor Agreement does require the use of ?...under a suitable FSF (Free Software Foundation) or OSI (Open Source Initiative) approved license?. Does that address your concern, or do you think that needs to be included in the Bylaws? Not really, as the contributor agreement is an external document under unilateral control. I'd suggest that the governance assert that "OpenJDK is a community based around the use of software licensed under the GPL supplemented with necessary exceptions" - avoiding mentioning the version and leaving room for the move to GPLv3 that the Oracle guys here at FOSDEM foreshadowed. > > - On ?Transparency?, your points are valid. All I can say that that now that we have the draft in place that we can demonstrate to the community over time that transparency is a value that we all share. No matter how (im)perfect a governance document may be, in the end it is deeds, not words, which will shape the success of a community. > > One quibble is that in the first section labelled ?Open, Meritocratic Oligarchy?, I would point out that the small board in its current form actually has one advantage, which is that because a 2/3?s majority is required for many major decisions it only takes two votes to block a mistake. That doesn?t change your larger point, but it does imply that a useful check is in place against capricious or inhospitable decisions. While that's a potential feature, it cuts both ways and ensures that a non-preferred competitor like Azul or Red Hat can also have their proposals indefinitely blocked. At least it would be public (unlike the apparent long-term block we heard about at FOSDEM on Azul's OpenJDK TCK license) but it would still be a problem. Regards S. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.openjdk.java.net/pipermail/gb-discuss/attachments/20110206/18f7e60b/attachment-0001.html From simon at webmink.com Sun Feb 6 07:32:20 2011 From: simon at webmink.com (Simon Phipps) Date: Sun, 6 Feb 2011 16:32:20 +0100 Subject: Problems acquiring an OpenJDK TCK license? In-Reply-To: References: <00aa01cbc5b5$c32e3950$498aabf0$@milinkovich@eclipse.org> <7940C8AE-A8DE-4177-8BC7-0C78EB13498B@webmink.com> Message-ID: <6BCD92F7-F42C-453D-A784-6D0AC4081B2B@webmink.com> [Switching to gb-discuss only] On 6 Feb 2011, at 15:33, Geir Magnusson Jr. wrote: > If true, this seems like an excellent thing for the new OpenJDK governing board to address immediately. That's actually an excellent test case. I can see that some might argue it is purely a contractual matter between Oracle and a "customer", but since maintaining a level playing field requires that the TCK is indeed offered in a fair, transparent and timely way there should probably be an explicit statement in the governance concerning the Board's authority to at least ask questions and preferably intervene in TCK cases. S. From mike.milinkovich at eclipse.org Sun Feb 6 07:59:14 2011 From: mike.milinkovich at eclipse.org (Mike Milinkovich) Date: Sun, 6 Feb 2011 10:59:14 -0500 Subject: Problems acquiring an OpenJDK TCK license? In-Reply-To: <6BCD92F7-F42C-453D-A784-6D0AC4081B2B@webmink.com> References: <00aa01cbc5b5$c32e3950$498aabf0$@milinkovich@eclipse.org> <7940C8AE-A8DE-4177-8BC7-0C78EB13498B@webmink.com> <6BCD92F7-F42C-453D-A784-6D0AC4081B2B@webmink.com> Message-ID: <002e01cbc616$cdeafa20$69c0ee60$@milinkovich@eclipse.org> At the risk of picking a barely-healed scab, isn't TCK licensing outside of the scope of OpenJDK governance? I understand how --- if you are so inclined --- you can start pulling at the threads to the point where every issue related to Java can be conflated with OpenJDK governance. But I don't see how that is going to be a particularly useful or constructive exercise. > [Switching to gb-discuss only] > > On 6 Feb 2011, at 15:33, Geir Magnusson Jr. wrote: > > > If true, this seems like an excellent thing for the new OpenJDK governing > board to address immediately. > > That's actually an excellent test case. I can see that some might argue it > is purely a contractual matter between Oracle and a "customer", but since > maintaining a level playing field requires that the TCK is indeed offered > in a fair, transparent and timely way there should probably be an explicit > statement in the governance concerning the Board's authority to at least > ask questions and preferably intervene in TCK cases. From simon at webmink.com Sun Feb 6 08:16:58 2011 From: simon at webmink.com (Simon Phipps) Date: Sun, 6 Feb 2011 17:16:58 +0100 Subject: Problems acquiring an OpenJDK TCK license? In-Reply-To: <002e01cbc616$cdeafa20$69c0ee60$@milinkovich@eclipse.org> References: <00aa01cbc5b5$c32e3950$498aabf0$@milinkovich@eclipse.org> <7940C8AE-A8DE-4177-8BC7-0C78EB13498B@webmink.com> <6BCD92F7-F42C-453D-A784-6D0AC4081B2B@webmink.com> <002e01cbc616$cdeafa20$69c0ee60$@milinkovich@eclipse.org> Message-ID: <9F718B8A-6553-45FA-9210-DDBE7B20A561@webmink.com> On 6 Feb 2011, at 16:59, Mike Milinkovich wrote: > At the risk of picking a barely-healed scab, isn't TCK licensing outside of > the scope of OpenJDK governance? > > I understand how --- if you are so inclined --- you can start pulling at the > threads to the point where every issue related to Java can be conflated with > OpenJDK governance. But I don't see how that is going to be a particularly > useful or constructive exercise. TCK licensing as a general commercial topic is undoubtedly beyond the scope of the OpenJDK governance and indeed none of the OpenJDK community's business. However, the availability of a no-charge TCK license to members of the OpenJDK community to ensure their contributions and private builds do not diverge from the JSRs they implement is a key part of the dynamic of the OpenJDK community. Without it, the ability of community members who do not have a private agreement with Oracle to contribute in a meaningful way is severely restricted. As such surely falls well within the oversight of the governance board. Clearly it's a tough subject and must not become an excuse to pry into commercially sensitive contracts, but it's entirely reasonable for the community to be assured that the TCK is in fact available as promised. S. From mike.milinkovich at eclipse.org Sun Feb 6 09:22:52 2011 From: mike.milinkovich at eclipse.org (Mike Milinkovich) Date: Sun, 6 Feb 2011 12:22:52 -0500 Subject: Problems acquiring an OpenJDK TCK license? In-Reply-To: <9F718B8A-6553-45FA-9210-DDBE7B20A561@webmink.com> References: <00aa01cbc5b5$c32e3950$498aabf0$@milinkovich@eclipse.org> <7940C8AE-A8DE-4177-8BC7-0C78EB13498B@webmink.com> <6BCD92F7-F42C-453D-A784-6D0AC4081B2B@webmink.com> <002e01cbc616$cdeafa20$69c0ee60$@milinkovich@eclipse.org> <9F718B8A-6553-45FA-9210-DDBE7B20A561@webmink.com> Message-ID: <003501cbc622$7cf3eee0$76dbcca0$@milinkovich@eclipse.org> Thanks for the clarification. I now understand the issue you and Geir are raising. > On 6 Feb 2011, at 16:59, Mike Milinkovich wrote: > > > At the risk of picking a barely-healed scab, isn't TCK licensing outside > of > > the scope of OpenJDK governance? > > > > I understand how --- if you are so inclined --- you can start pulling at > the > > threads to the point where every issue related to Java can be conflated > with > > OpenJDK governance. But I don't see how that is going to be a > particularly > > useful or constructive exercise. > > TCK licensing as a general commercial topic is undoubtedly beyond the scope > of the OpenJDK governance and indeed none of the OpenJDK community's > business. > > However, the availability of a no-charge TCK license to members of the > OpenJDK community to ensure their contributions and private builds do not > diverge from the JSRs they implement is a key part of the dynamic of the > OpenJDK community. Without it, the ability of community members who do not > have a private agreement with Oracle to contribute in a meaningful way is > severely restricted. > > As such surely falls well within the oversight of the governance board. > Clearly it's a tough subject and must not become an excuse to pry into > commercially sensitive contracts, but it's entirely reasonable for the > community to be assured that the TCK is in fact available as promised. From dl at cs.oswego.edu Sun Feb 6 15:08:48 2011 From: dl at cs.oswego.edu (Doug Lea) Date: Sun, 06 Feb 2011 18:08:48 -0500 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <58912.193.191.38.62.1296910561.squirrel@gnu.wildebeest.org> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> <4D4C41A2.9070207@cs.oswego.edu> <58912.193.191.38.62.1296910561.squirrel@gnu.wildebeest.org> Message-ID: <4D4F2A00.8020807@cs.oswego.edu> On 02/05/11 07:56, Mark Wielaard wrote: > if you are not addressing these things, please > at least document the rational for not addressing these issue. I think there are only two such things: I hope a revised Contributor agreement is dealt with soon. We also need some per-project process documents describing rules and conventions about repository usage, commit rules, reviews, testing, and releases. I still agree that these need attention by the GB and others, but I still don't see how they directly impact bylaws. > > So, does OpenJDK implement/define JavaSE or not? No; OpenJDK will produce stable releases of projects such as "jdk8" that might just so happen to implement some other bodies' definition of "Java SE 8". I continue to think that one can and should omit reference to any such other bodies or corporations in these parts of the bylaws. So for example, no reference TCK licenses or Oracle branding agreements pertaining to products. However, the GB is required to ensure that contributors have access to "infrastructure": code, docs, tests, reviewers, etc necessary to successfully contribute. Which is likely to include those tests that comprise the TCK, as well as participation in a (hopefully revamped) "compatibility review" process that occasionally rejects (occasionally for good reason :-) bug fixes on the grounds of backward compatibility. And so on. -Doug From dl at cs.oswego.edu Sun Feb 6 15:33:18 2011 From: dl at cs.oswego.edu (Doug Lea) Date: Sun, 06 Feb 2011 18:33:18 -0500 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <4D4E76C6.3060800@gmail.com> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> <4D4C41A2.9070207@cs.oswego.edu> <58912.193.191.38.62.1296910561.squirrel@gnu.wildebeest.org> <4D4E76C6.3060800@gmail.com> Message-ID: <4D4F2FBE.5040009@cs.oswego.edu> On 02/06/11 05:24, Jonas Tingeborn wrote: >>>>> The Governing Board consists of five individuals: >>>>> * The Chair, appointed by Oracle; >>>>> * The Vice-Chair, appointed by IBM; >>>>> * The OpenJDK Lead, appointed by Oracle; and >>>>> * Two At-Large Members, nominated and elected as described below. >>>> > Though, I am concerned that specific companies are written into the bylaws. > Wouldn't it be better to simply prop the roles with Oracle and IBM people when > the bylaws take effect and then let people on the board earn their right to be > there like the other groups? FWIW: My take is that corporations these days are too cowardly. Here, they insist on appointment to avoid some (for them) worst-case scenarios. My sense is that if Oracle and IBM ever do anything that would make the community not want them on the GB, then we will have much bigger problems than just GB membership rules. So I tolerate their paranoia. I do agree that it is not very nice not to have members from Red Hat, Azul, Google, or others. But as I mentioned, limiting the initial GB to what seems to be minimum possible size has the advantage of being less likely to stall for years (as did some previous efforts). -Doug From aph at redhat.com Tue Feb 8 02:52:45 2011 From: aph at redhat.com (Andrew Haley) Date: Tue, 08 Feb 2011 10:52:45 +0000 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <4D4F2FBE.5040009@cs.oswego.edu> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> <4D4C41A2.9070207@cs.oswego.edu> <58912.193.191.38.62.1296910561.squirrel@gnu.wildebeest.org> <4D4E76C6.3060800@gmail.com> <4D4F2FBE.5040009@cs.oswego.edu> Message-ID: <4D51207D.2050309@redhat.com> On 02/06/2011 11:33 PM, Doug Lea wrote: > I do agree that it is not very nice not to have members > from Red Hat, Azul, Google, or others. But as I mentioned, > limiting the initial GB to what seems to be minimum > possible size has the advantage of being less likely to stall > for years (as did some previous efforts). Ahem, point of fact: the previous effort at a governance board (there has only been one such AFAIAA) did not stall because of a lack of consensus. Andrew. From mark at klomp.org Tue Feb 8 09:53:24 2011 From: mark at klomp.org (Mark Wielaard) Date: Tue, 8 Feb 2011 18:53:24 +0100 (CET) Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <00a801cbc5b1$025550a0$06fff1e0$@milinkovich@eclipse.org> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> <4D4C41A2.9070207@cs.oswego.edu> <00a801cbc5b1$025550a0$06fff1e0$@milinkovich@eclipse.org> Message-ID: <45829.80.101.103.228.1297187604.squirrel@gnu.wildebeest.org> Hi Mike, Thanks for participating in the discussion. On Sun, February 6, 2011 04:50, Mike Milinkovich wrote: > Mark also raised the point that " It also doesn't state where this > infrastructure comes from or who maintains it." I expect that the answer > is Oracle. But I think that putting this level of detail into the Bylaws > would be a mistake, as it is an execution matter not a governance rule. I guess I wasn't very clear explaining. My point is mainly that the governance bylaws should either explain how the community "owns" the processes, or if the community isn't really in control over the processes described, then it shouldn't be in the bylaws (since it will only lead to frustration that it might be something we think we control, but ultimately will be vetoed outside the normal processes.) Lets give a concrete example. We are in desperate need of a better bugtracker. What would the process be if someone volunteered to set one up? How would "control" be delegated to this volunteer, who gets to appoint the bug-masters, and how would the governance board accept or reject donations for hosting and/or machines for this purpose? If this is this a process that the community really controls, and we can come up with a process for doing that. Then infrastructure definitely needs to be mentioned in the bylaws. But if this isn't something controlled by the participants, but something dictated/controlled outside of the democratic process, then it is better to just leave it out (or codify that this isn't up for community participation). Cheers, Mark From mark at klomp.org Tue Feb 8 10:07:18 2011 From: mark at klomp.org (Mark Wielaard) Date: Tue, 8 Feb 2011 19:07:18 +0100 (CET) Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <4D4F2A00.8020807@cs.oswego.edu> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> <4D4C41A2.9070207@cs.oswego.edu> <58912.193.191.38.62.1296910561.squirrel@gnu.wildebeest.org> <4D4F2A00.8020807@cs.oswego.edu> Message-ID: <57551.80.101.103.228.1297188438.squirrel@gnu.wildebeest.org> On Mon, February 7, 2011 00:08, Doug Lea wrote: > On 02/05/11 07:56, Mark Wielaard wrote: >> So, does OpenJDK implement/define JavaSE or not? > > No; OpenJDK will produce stable releases of projects such as > "jdk8" that might just so happen to implement some other > bodies' definition of "Java SE 8". > > I continue to think that one can and should omit reference > to any such other bodies or corporations in these parts > of the bylaws. So for example, no reference TCK licenses > or Oracle branding agreements pertaining to products. Agreed. > However, the GB is required to ensure that contributors have > access to "infrastructure": code, docs, tests, reviewers, etc > necessary to successfully contribute. Yes, this is very important. All participants should have equal access to code, specs, tests, etc. And all contributors should make sure their contributions are complete and provide all these. > Which is likely to > include those tests that comprise the TCK But here I disagree. There is no way that the current TCK setup can be integrated with an open community approach. As long as the TCK is proprietary, only available to an arbitrary subset of participants, based on the whims of a commercial corporation, requires signing an NDA that prevents effective exchange of information between all participants, the TCK should be avoided. (If only to shield the governance board members who undoubtedly will become very frustrated like has happened with the JCP.) > as well as > participation in a (hopefully revamped) "compatibility > review" process that occasionally rejects (occasionally for > good reason :-) bug fixes on the grounds of backward > compatibility. And so on. Here I do agree again. There should be clear rules for how anyone can participate in this process and make sure every participant does provide tests for such backward compatibility guarantees we decide to make through our codebase. Cheers, Mark From mark at klomp.org Tue Feb 8 10:17:29 2011 From: mark at klomp.org (Mark Wielaard) Date: Tue, 8 Feb 2011 19:17:29 +0100 (CET) Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <4D4F2FBE.5040009@cs.oswego.edu> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> <4D4C41A2.9070207@cs.oswego.edu> <58912.193.191.38.62.1296910561.squirrel@gnu.wildebeest.org> <4D4E76C6.3060800@gmail.com> <4D4F2FBE.5040009@cs.oswego.edu> Message-ID: <51722.80.101.103.228.1297189049.squirrel@gnu.wildebeest.org> Hi Doug, On Mon, February 7, 2011 00:33, Doug Lea wrote: > FWIW: My take is that corporations these days are too cowardly. > Here, they insist on appointment to avoid some (for them) > worst-case scenarios. My sense is that if Oracle and IBM ever > do anything that would make the community not want them on > the GB, then we will have much bigger problems than just > GB membership rules. So I tolerate their paranoia. I don't think anybody thinks Oracle and/or IBM should be denied representation on the board. But the choice of nominees is certainly questionable. As is the fact that some of these seats seem to be by appointment only. If you want to have a IBM representative on the board, why not nominate someone like Steve Poole who actively communicates with the rest of the participants in the community? Just like Mark is a good choice for a nomination that will make Oracle feel like being represented. However nominating (or even just appointing without discussion) people from these companies who have not even once participated on the mailinglists means we do have a big problem. IMHO. > I do agree that it is not very nice not to have members > from Red Hat, Azul, Google, or others. But as I mentioned, > limiting the initial GB to what seems to be minimum > possible size has the advantage of being less likely to stall > for years (as did some previous efforts). Can you explain what was wrong with the previous board of seven members (Martin, Andrew, Fabiane, Simon, Mark, Dalibor and you) that made it so that things stalled? That board felt much more representative of the participants of this community than this proposed board. Thanks, Mark From mike.milinkovich at eclipse.org Tue Feb 8 10:49:53 2011 From: mike.milinkovich at eclipse.org (Mike Milinkovich) Date: Tue, 8 Feb 2011 13:49:53 -0500 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <45829.80.101.103.228.1297187604.squirrel@gnu.wildebeest.org> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> <4D4C41A2.9070207@cs.oswego.edu> <00a801cbc5b1$025550a0$06fff1e0$@milinkovich@eclipse.org> <45829.80.101.103.228.1297187604.squirrel@gnu.wildebeest.org> Message-ID: <015101cbc7c0$f9c50cc0$ed4f2640$@milinkovich@eclipse.org> Mark, Thanks for the clarification. At the moment, I don't actually know if Oracle thinks this infrastructure is something they control, or if it's something the community controls. The Bylaws are actually ambiguous on that point. I promise to find an answer. Given that the starting point is that the majority of the committers work for them, Oracle obviously has some skin in that game. I hope that it is something the community controls. At least to the point where a consensus for change can force a response. That said, having gone through several similar infrastructure transitions at Eclipse, I think this example could be a very ticklish case. A lot of development infrastructure becomes engrained in work flows and even team culture. Change is not easy when that's true. Especially when there is time pressure to ship releases. If the answer to the question is that the community controls the infrastructure, I expect that an Infrastructure Group which sponsored a collection of projects working in the space would be how the topic would be handled under these Bylaws. Or alternatively the OpenJDK Members Group or the Board Group could sponsor such projects. > Thanks for participating in the discussion. > > On Sun, February 6, 2011 04:50, Mike Milinkovich wrote: > > Mark also raised the point that " It also doesn't state where this > > infrastructure comes from or who maintains it." I expect that the answer > > is Oracle. But I think that putting this level of detail into the Bylaws > > would be a mistake, as it is an execution matter not a governance rule. > > I guess I wasn't very clear explaining. My point is mainly that the > governance bylaws should either explain how the community "owns" the > processes, or if the community isn't really in control over the > processes described, then it shouldn't be in the bylaws (since it > will only lead to frustration that it might be something we think > we control, but ultimately will be vetoed outside the normal > processes.) > > Lets give a concrete example. We are in desperate need of a better > bugtracker. What would the process be if someone volunteered to > set one up? How would "control" be delegated to this volunteer, who > gets to appoint the bug-masters, and how would the governance board > accept or reject donations for hosting and/or machines for this purpose? > > If this is this a process that the community really controls, and we > can come up with a process for doing that. Then infrastructure > definitely needs to be mentioned in the bylaws. But if this isn't > something controlled by the participants, but something > dictated/controlled outside of the democratic process, then it is > better to just leave it out (or codify that this isn't up for > community participation). > > Cheers, > > Mark From mike.milinkovich at eclipse.org Tue Feb 8 13:36:09 2011 From: mike.milinkovich at eclipse.org (Mike Milinkovich) Date: Tue, 8 Feb 2011 16:36:09 -0500 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <51722.80.101.103.228.1297189049.squirrel@gnu.wildebeest.org> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> <4D4C41A2.9070207@cs.oswego.edu> <58912.193.191.38.62.1296910561.squirrel@gnu.wildebeest.org> <4D4E76C6.3060800@gmail.com> <4D4F2FBE.5040009@cs.oswego.edu> <51722.80.101.103.228.1297189049.squirrel@gnu.wildebeest.org> Message-ID: <007201cbc7d8$333d7480$99b85d80$@milinkovich@eclipse.org> > I don't think anybody thinks Oracle and/or IBM should be > denied representation on the board. But the choice of nominees > is certainly questionable. As is the fact that some of these > seats seem to be by appointment only. If you want to have a > IBM representative on the board, why not nominate someone like > Steve Poole who actively communicates with the rest of the > participants in the community? Just like Mark is a good choice > for a nomination that will make Oracle feel like being represented. > However nominating (or even just appointing without discussion) > people from these companies who have not even once participated > on the mailinglists means we do have a big problem. IMHO. For good or bad, I think I can help explain the corporate logic that goes into the appointees you are referring to. Simply put, they are the executives who control the resources and business deliverables. Adam is the SVP in charge of Java Platform at Oracle. Jason is the Director of the Java Technology Centre at IBM. They have the responsibility to their respective businesses to ensure that Java and OpenJDK are successful. And it is, to a certain extent, their jobs on the line if it screws up. A more positive interpretation of their involvement on the board is that it is a reflection of the seriousness with which their respective companies view the OpenJDK community. The Eclipse Board explicitly has a mix of business-centric and community-centric representatives on it. In practice, it has actually worked well because the diversity of views have generally speaking resulted in better decisions. Diversity takes many forms, but it is almost always a force for good. Note that I am not making a value judgement on any of this. I'm just explaining how it works. Please don't shoot the messenger :-) A theme that I've noticed in several comments is one which could be summarized as "the board needs to be bigger, with a more diverse representation". That is perfectly valid and actionable feedback and I've noted it. But I would also point out that it's not really a criticism of the draft Bylaws per se, as it is certainly possible to grow the board within the proposed governance framework. /mike From simon at webmink.com Tue Feb 8 13:59:36 2011 From: simon at webmink.com (Simon Phipps) Date: Tue, 8 Feb 2011 21:59:36 +0000 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <007201cbc7d8$333d7480$99b85d80$@milinkovich@eclipse.org> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> <4D4C41A2.9070207@cs.oswego.edu> <58912.193.191.38.62.1296910561.squirrel@gnu.wildebeest.org> <4D4E76C6.3060800@gmail.com> <4D4F2FBE.5040009@cs.oswego.edu> <51722.80.101.103.228.1297189049.squirrel@gnu.wildebeest.org> <007201cbc7d8$333d7480$99b85d80$@milinkovich@eclipse.org> Message-ID: <5D84BD24-D529-47F1-A783-7219E4FF2692@webmink.com> On 8 Feb 2011, at 21:36, Mike Milinkovich wrote: > A theme that I've noticed in several comments is one which could be > summarized as "the board needs to be bigger, with a more diverse > representation". That is perfectly valid and actionable feedback and I've > noted it. But I would also point out that it's not really a criticism of the > draft Bylaws per se, as it is certainly possible to grow the board within > the proposed governance framework. This was feedback that MR and Joe received strongly in FOSDEM from the DevRoom. The previous interim Board ended up with seven members, and always had a majority of non-Sun employees on it - as in fact dictated by the OpenJDK Charter[1], in which Sun made a public commitment to the nature of the governance and set the absolute minimum baseline requirements for governance. I'd suggest the feedback does in fact fall within scope as the current draft does not fulfil the Charter requirements. The Charter currently says: > The Governance Board (?GB?) shall be comprised of seven natural persons. Three members shall be employees of Sun, and four shall be elected from the OpenJDK Community (?the Community?). I suggest the draft be updated to comply with the commitment Oracle inherited from Sun. In particular, I strongly suggest the interim Board respect clauses 1.3 and the spirit of clause 7.1, modify the draft to reflect the Charter and have the non-Charter-compliant appointees occupy their seats only for the purposes of conducting an election. Simon [1] http://openjdk.java.net/legal/charter/ - note especially the amendment. From mike.milinkovich at eclipse.org Tue Feb 8 14:10:22 2011 From: mike.milinkovich at eclipse.org (Mike Milinkovich) Date: Tue, 8 Feb 2011 17:10:22 -0500 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <5D84BD24-D529-47F1-A783-7219E4FF2692@webmink.com> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> <4D4C41A2.9070207@cs.oswego.edu> <58912.193.191.38.62.1296910561.squirrel@gnu.wildebeest.org> <4D4E76C6.3060800@gmail.com> <4D4F2FBE.5040009@cs.oswego.edu> <51722.80.101.103.228.1297189049.squirrel@gnu.wildebeest.org> <007201cbc7d8$333d7480$99b85d80$@milinkovich@eclipse.org> <5D84BD24-D529-47F1-A783-7219E4FF2692@webmink.com> Message-ID: <009001cbc7dc$fb58bb60$f20a3220$@milinkovich@eclipse.org> Well, you've stumped me. I don't whether the scenario we are operating under is a reboot or a continuation of the previous charter. MR is probably the best person to answer that question. > On 8 Feb 2011, at 21:36, Mike Milinkovich wrote: > > > A theme that I've noticed in several comments is one which could be > > summarized as "the board needs to be bigger, with a more diverse > > representation". That is perfectly valid and actionable feedback and I've > > noted it. But I would also point out that it's not really a criticism of > the > > draft Bylaws per se, as it is certainly possible to grow the board within > > the proposed governance framework. > > This was feedback that MR and Joe received strongly in FOSDEM from the > DevRoom. The previous interim Board ended up with seven members, and always > had a majority of non-Sun employees on it - as in fact dictated by the > OpenJDK Charter[1], in which Sun made a public commitment to the nature of > the governance and set the absolute minimum baseline requirements for > governance. I'd suggest the feedback does in fact fall within scope as the > current draft does not fulfil the Charter requirements. > > The Charter currently says: > > The Governance Board ("GB") shall be comprised of seven natural persons. > Three members shall be employees of Sun, and four shall be elected from the > OpenJDK Community ("the Community"). > > I suggest the draft be updated to comply with the commitment Oracle > inherited from Sun. In particular, I strongly suggest the interim Board > respect clauses 1.3 and the spirit of clause 7.1, modify the draft to > reflect the Charter and have the non-Charter-compliant appointees occupy > their seats only for the purposes of conducting an election. > > Simon > > > [1] http://openjdk.java.net/legal/charter/ - note especially the > amendment.= From mark at klomp.org Tue Feb 8 14:18:50 2011 From: mark at klomp.org (Mark Wielaard) Date: Tue, 08 Feb 2011 23:18:50 +0100 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <009001cbc7dc$fb58bb60$f20a3220$@milinkovich@eclipse.org> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> <4D4C41A2.9070207@cs.oswego.edu> <58912.193.191.38.62.1296910561.squirrel@gnu.wildebeest.org> <4D4E76C6.3060800@gmail.com> <4D4F2FBE.5040009@cs.oswego.edu> <51722.80.101.103.228.1297189049.squirrel@gnu.wildebeest.org> <007201cbc7d8$333d7480$99b85d80$@milinkovich@eclipse.org> <5D84BD24-D529-47F1-A783-7219E4FF2692@webmink.com> <009001cbc7dc$fb58bb60$f20a3220$@milinkovich@eclipse.org> Message-ID: <1297203530.3956.94.camel@springer.wildebeest.org> On Tue, 2011-02-08 at 17:10 -0500, Mike Milinkovich wrote: > Well, you've stumped me. I don't whether the scenario we are operating under > is a reboot or a continuation of the previous charter. BTW. I think this is why a lot of people feel the newly proposed board is a step backwards. It is as if the knowledge accumulated by the old board which operated inside the community for the last couple of years is just thrown away. Please do bring back some of the old board members to get this process on track faster. From mike.milinkovich at eclipse.org Tue Feb 8 14:44:46 2011 From: mike.milinkovich at eclipse.org (mike.milinkovich at eclipse.org) Date: Tue, 8 Feb 2011 22:44:46 +0000 Subject: OpenJDK Community Bylaws and Governing Board Message-ID: <470310663-1297205087-cardhu_decombobulator_blackberry.rim.net-518591141-@bda055.bisx.prod.on.blackberry> Mark, I understand. But I would point out that we should not confuse the issue of the board with providing feedback on the bylaws. Additional feedback on the document itself would be appreciated. ------Original Message------ From: Mark Wielaard To: Mike Milinkovich Cc: 'Simon Phipps' Cc: gb-discuss at openjdk.java.net Subject: RE: OpenJDK Community Bylaws and Governing Board Sent: Feb 8, 2011 5:18 PM On Tue, 2011-02-08 at 17:10 -0500, Mike Milinkovich wrote: > Well, you've stumped me. I don't whether the scenario we are operating under > is a reboot or a continuation of the previous charter. BTW. I think this is why a lot of people feel the newly proposed board is a step backwards. It is as if the knowledge accumulated by the old board which operated inside the community for the last couple of years is just thrown away. Please do bring back some of the old board members to get this process on track faster. Mike Milinkovich mike.milinkovich at eclipse.org +1.613.220.3223 (mobile) From mark at klomp.org Tue Feb 8 15:00:12 2011 From: mark at klomp.org (Mark Wielaard) Date: Wed, 09 Feb 2011 00:00:12 +0100 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <470310663-1297205087-cardhu_decombobulator_blackberry.rim.net-518591141-@bda055.bisx.prod.on.blackberry> References: <470310663-1297205087-cardhu_decombobulator_blackberry.rim.net-518591141-@bda055.bisx.prod.on.blackberry> Message-ID: <1297206012.3956.99.camel@springer.wildebeest.org> On Tue, 2011-02-08 at 22:44 +0000, mike.milinkovich at eclipse.org wrote: > Mark, > > I understand. But I would point out that we should not confuse the issue of the board with providing feedback on the bylaws. Additional feedback on the document itself would be appreciated. Even more? Have you resolved all earlier reported issues already? :) Seriously, I was under the impression that the makeup of the newly proposed board was part of the bylaws. I have given feedback on the text of that, which I hope you will consider for revision. In particular, the bylaws currently suggest that the board should have three appointed members with a minority number of seats for independent candidates. I propose you change that to something more in line with the previous bylaws which at least made sure there was a majority of independent seats. And ideally it doesn't have any non-democratic seats. Thanks, Mark From dl at cs.oswego.edu Tue Feb 8 17:11:27 2011 From: dl at cs.oswego.edu (Doug Lea) Date: Tue, 08 Feb 2011 20:11:27 -0500 Subject: OpenJDK Community Bylaws and Governing Board In-Reply-To: <1297203530.3956.94.camel@springer.wildebeest.org> References: <20110203165944.6B528AC4@eggemoggin.niobe.net> <1296764502.3341.87.camel@springer.wildebeest.org> <4D4C41A2.9070207@cs.oswego.edu> <58912.193.191.38.62.1296910561.squirrel@gnu.wildebeest.org> <4D4E76C6.3060800@gmail.com> <4D4F2FBE.5040009@cs.oswego.edu> <51722.80.101.103.228.1297189049.squirrel@gnu.wildebeest.org> <007201cbc7d8$333d7480$99b85d80$@milinkovich@eclipse.org> <5D84BD24-D529-47F1-A783-7219E4FF2692@webmink.com> <009001cbc7dc$fb58bb60$f20a3220$@milinkovich@eclipse.org> <1297203530.3956.94.camel@springer.wildebeest.org> Message-ID: <4D51E9BF.8050400@cs.oswego.edu> On 02/08/11 17:18, Mark Wielaard wrote: > I think this is why a lot of people feel the newly proposed board > is a step backwards. It is as if the knowledge accumulated by the old > board which operated inside the community for the last couple of years > is just thrown away. This is my very least favorite topic -- I think that accepting the proposed initial GB is a tolerable tradeoff for the sake of actually having bylaws. I don't have any other defense of it. But there is no "old board". There was one interim GB meeting ever, attended by me, Dalibor Topic, Mark Reinhold, Simon Phipps, and Fabiane Nardon, resulting in plans to write the constitution (as it was then called, not bylaws). Which never actually got written. (I think it is safe to say that it didn't get done because doing so was a low priority at Sun at the time.) When Dalibor joined, and Simon left, Sun, some members were added, but this group never met or did anything before being sunsetted out of existence. My only GB-related activity for the past two years has been occasional prods to Mark Reinhold to someday restart this. -Doug From jgartner at ca.ibm.com Wed Feb 9 16:14:18 2011 From: jgartner at ca.ibm.com (Jason Gartner) Date: Wed, 9 Feb 2011 19:14:18 -0500 Subject: IBM participation in OpenJDK Message-ID: IBM joined OpenJDK so our first order of business is figuring out how to participate. We're committed and we're working there now and without community bylaws, or board or anything. We do have active members today and as we get more and more involved in OpenJDK, I expect you to see more of us within the community. I have been digesting (not fast enough) the feedback that has been coming in through this mechanism and I'm enthusiastic that we are getting this level of involvement and passion. We need to continue to progress on the charter of OpenJDK. We all want more people to join and we need to understand how best we can all work together. As IBM ramps its participation, we are learning as we go and we this as an evolution, not a final step. IBM is committed to seeing OpenJDK as an open community that fosters innovation and vitality of Java. There are many challenges that we all face in light of reality, many of which have been raised here in the mailing list. Active participation is how this community and the future of Java will ultimately be judged and we can not wait until the rules are perfect. We have members contributing today in spite of a lack of bylaws because we only see it as a single step of a long journey. Progress may be interpreted as slow or fast depending on your vantage point, but we are talking. That is the main point. IBM has built open source projects before with significant rules and regulations to drive clarity - all of that is good , but in the end, it's the intent that matters, and our intent is to work here on this to move Java forward. We took a position that if we need to rely on the rule of law, and process at openJDK, we're finished before we even started. A little about myself... No, I don't write code anymore, but do consider myself technical. I do however, represent and lead a large Java investment from IBM. An investment that IBM has made since 1995. Individuals such as Steve Poole (referenced) take business and technical direction from me. I am here as the bridge between the pure technology and to balance that with IBM's investment in Java from both a business and resource point of view. IBM is committed to Java and the healthier the ecosystem is, the better it is for IBM as a business. Jason 2670 Queensview Gartner Drive Director, Ottawa, ON K2B 8K1 Java Technologies IBM Software Canada Group Phone: +1-613-356-6242 e-mail: jgartner at ca.ibm.com -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.openjdk.java.net/pipermail/gb-discuss/attachments/20110209/f32ad722/attachment.html -------------- next part -------------- A non-text attachment was scrubbed... Name: ecblank.gif Type: image/gif Size: 45 bytes Desc: not available Url : http://mail.openjdk.java.net/pipermail/gb-discuss/attachments/20110209/f32ad722/attachment.gif -------------- next part -------------- A non-text attachment was scrubbed... Name: 0C221731.gif Type: image/gif Size: 360 bytes Desc: not available Url : http://mail.openjdk.java.net/pipermail/gb-discuss/attachments/20110209/f32ad722/attachment-0001.gif