Re: [VOTE] PPMCs for Incubating Projects
On Mon, Dec 15, 2003 at 02:47:43PM -0500, Noel J. Bergman wrote: > A PMC policy to be determined is whether the PPMC mailing list is optional > for PMC members. The one mailing list created to date (geronimo-ppmc) is > opt-in, although I did pre-subscribe you (along with myself, Geir and > James). What do you mean by this? Are you asking, for example, if Incubator PMC members should be forced to join all [EMAIL PROTECTED] lists? Or, are you asking, for example, if Geronimo PMC members should be forced to join their own [EMAIL PROTECTED] list? If it is the first, I completely disagree. The ASF is a volunteer organization. If someone wishes to help mentor a project under Incubation, then they should simply join the ppmc list and participate. -aaron - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: Exiting Incubation - Status Check
Jochen Wiedmann asked: > * No non ASL or ASL compatbile dependencies in the code base > In the case of JaxMe, there are several unit tests, which > depend on hsqldb. I don't know off-hand, so I'll ask you: isn't hsqldb's license BSD-style? What makes it incompatible? > hsqldb is required to build the sources. Why is hsqldb required to compile the code? Is the code not using JDBC? > * Check of project name for trademark issues > How do I do that? I'd suggest google and trademark databases. > In particular, how do I record that I have checked and found no issues? The project's STATUS file (http://cvs.apache.org/viewcvs.cgi/incubator/site/projects/jaxme.cwiki), which needs updating. > *When* do I check? As soon as you can. > * Demonstrate an active and diverse development community > Who can give me a clue whether the community is supposed > to be sufficiently "active and diverse"? Does any one organization control the project? How diverse is the community? > * Demonstrate ability to tolerate and resolve conflict within the community. Conflict != tantrum. :-) Has there never been a disagreement, a debate over something? > * Developers tied into ASF PGP web of trust Have the developers generated PGP keys and started to get signed into the web of trust? > I do not know what is required to conform to this part. Don't feel badly. :-) This is something we're trying to work out throughout the ASF, and we recognize that not everyone is conveniently located to each other. > For the record, I would like to note, that IMO the other requirements > are fulfilled by JaxMe. Speaking of the record, the status file is the place to record such things. :-) Sounds good all around, from what you are saying. Dims, who has participated in your project, will have the best idea. --- Noel - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [VOTE] PPMCs for Incubating Projects
Aaron Bannert wrote: On Mon, Dec 15, 2003 at 02:47:43PM -0500, Noel J. Bergman wrote: A PMC policy to be determined is whether the PPMC mailing list is optional for PMC members. The one mailing list created to date (geronimo-ppmc) is opt-in, although I did pre-subscribe you (along with myself, Geir and James). What do you mean by this? Are you asking, for example, if Incubator PMC members should be forced to join all [EMAIL PROTECTED] lists? Or, are you asking, for example, if Geronimo PMC members should be forced to join their own [EMAIL PROTECTED] list? If it is the first, I completely disagree. The ASF is a volunteer organization. If someone wishes to help mentor a project under Incubation, then they should simply join the ppmc list and participate. This is not exactly how it should work. What happens ATM: if a problem has to be solved on the private PMC list, we have to use [EMAIL PROTECTED] This is a *big* problem, as the future PMCers of the project in question cannot participate in the discussion. I think you can remember the "pass it on" that has been done by Mentors between the pmc and the developers. If we make all developers come on the PMC list it may become at times a really confusing thing, and all projects will know of all others. Besides, the project is still not formed, and so the should not be on a real PMC. Furthermore, we know that more eyes are there, the better it is. If we partition the incubator PMC list in many PPMC lists, we have the *same* workload we had before, but are able to make developers participate, and in practice we may even have to do less work, as we will pass to become observers. For example, it's no secret that the incubator PMC has been discussing some action items about Geronimo on the private pmc list. If we already had a geronimo-ppmc, the discussion would have simply be done there, but with the important participation of geronimo developers. Note that this is *not* generic mentoring, but the same identical issue solving that we did till now, only that we get more help and get away with useless indirections. Finally, the private pmc list should be a last resort mechanism, so we will not be doing normal project mentoring there. I think that the below proposal explains the PPMC in sufficient detail: http://wiki.apache.org/incubator/PpmcProposal -- Nicola Ken Barozzi [EMAIL PROTECTED] - verba volant, scripta manent - (discussions get forgotten, just code remains) - - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: [VOTE] PPMCs for Incubating Projects
Aaron Bannert wrote: > Noel J. Bergman wrote: > > A PMC policy to be determined is whether the PPMC mailing list > > is optional for PMC members. The one mailing list created to > > date (geronimo-ppmc) is opt-in, although I did pre-subscribe > > you (along with myself, Geir and James). > What do you mean by this? I mean that I pre-subscribed Nicola Ken, myself, Geir and James to the PPMC, as well as all of the current Geronimo Committers. Anyone else from the PMC who wanted to be on the mailing list needs to subscribe. > Are you asking [if] Incubator PMC members should be forced to join > all [EMAIL PROTECTED] lists? > If [so], I completely disagree. Opinions seem to be mixed on the subject, so I wanted to raise the question. You can agree or disagree, but I've got to know what I'm doing when I create these lists, which is why I am asking for a policy. --- Noel - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Exiting Incubation - Status Check
Hi, Noel, first of all, thanks for your reply. Noel J. Bergman wrote: I don't know off-hand, so I'll ask you: isn't hsqldb's license BSD-style? What makes it incompatible? Quoting one Noel J. Bergman: ;-) http://marc.theaimsgroup.com/?l=incubator-general&m=106330819625235&w=2 Btw, using Axion is not an option, as long as Axion doesn't support composed primary keys. Why is hsqldb required to compile the code? Is the code not using JDBC? Some part of JaxMe is able to generate code by querying for JDBC metadata. The test suite uses hsqldb to verify that part. And building the distribution includes building the test suite. We can discuss whether this needs to be like it is, but I definitely like it for reasons of quality. A better solution would be to resolve the license issue. The project's STATUS file (http://cvs.apache.org/viewcvs.cgi/incubator/site/projects/jaxme.cwiki), which needs updating. How do I update that file? See http://marc.theaimsgroup.com/?l=incubator-general&m=106741007220764&w=2 which was never replied. Does any one organization control the project? How diverse is the community? No. Made of individuals. Conflict != tantrum. :-) Has there never been a disagreement, a debate over something? There have been discussions and votes, but so far no true disagreements. Have the developers generated PGP keys and started to get signed into the web of trust? That second part is the interesting thing: How do we add our keys to the "web of trust"? (Not that I know what it is. :-) Speaking of the record, the status file is the place to record such things. :-) See above. :-) Jochen - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: Exiting Incubation - Status Check
Jochen Wiedmann wrote: > Noel J. Bergman wrote: > > I don't know off-hand, so I'll ask you: isn't hsqldb's license BSD-style? > > What makes it incompatible? > http://marc.theaimsgroup.com/?l=incubator-general&m=106330819625235&w=2 That makes the project unavailable for changing the license (to become an ASF project), but I don't see why the license is incompatible. > using Axion is not an option, as long as Axion doesn't support composed > primary keys. Are they aware of it? Not that it matters for your status. > > Why is hsqldb required to compile the code? Is the code not using JDBC? > Some part of JaxMe is able to generate code by querying for JDBC metadata. > The test suite uses hsqldb to verify that part. And building the > distribution includes building the test suite. So make the process conditional upon finding what it needs, and generating a warning that certain test cannot be performed because something is missing. I think that would be fine, too. > The project's STATUS file > (http://cvs.apache.org/viewcvs.cgi/incubator/site/projects/jaxme.cwiki), > which needs updating. > How do I update that file? The PPMC concept makes this much easier, IMO. I think that you could make with either review-then-commit or commit-then-review, but either way the PPMC would vote on the changes. At the moment, I think that the committers, along with dims, could vote on the status changes as the project's statement to the PMC. > > Does any one organization control the project? How diverse is the > > community? > No. Made of individuals. And since it will be part of the WS project, that helps. What would really help is for projects to adopt the philosophy that: (1) Seeks to get all active Committers onto the PMC in a manner based upon their being trusted to participate in decision making. (2) Accept that only PMC member votes are binding. (3) Every Committer has karma for every module. It is just one project with multiple, related, codebases. > > Conflict != tantrum. :-) Has there never been a disagreement, a debate > > over something? > There have been discussions and votes, but so far no true disagreements. Sound good so far. :-) > > Have the developers generated PGP keys and started to get signed into the > > web of trust? > That second part is the interesting thing: How do we add our keys to the > "web of trust"? (Not that I know what it is. :-) In fairness, that is something we (the ASF) need to help you with. We cannot, in my opinion, expect you to know. There are some proposals, but I do not believe that anything is final. Are any of you geographically co-located? --- Noel - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: Exiting Incubation - Status Check
i'm interested in the web of trust as well. i've just started following the incubator/ infrastructure lists. is there a summary of the proposals some where? i imagine there are a number of apache committers in the SF bay area, making it possible to physically verify identity.. cheers, andy > -Original Message- > From: Noel J. Bergman [mailto:[EMAIL PROTECTED] > Sent: Tuesday, December 16, 2003 9:25 AM > To: [EMAIL PROTECTED] > Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED] > Subject: RE: Exiting Incubation - Status Check > > > Jochen Wiedmann wrote: > > > Noel J. Bergman wrote: > > > I don't know off-hand, so I'll ask you: isn't hsqldb's license > BSD-style? > > > What makes it incompatible? > > > > http://marc.theaimsgroup.com/?l=incubator-general&m=1063308196 > 25235&w=2 > > That makes the project unavailable for changing the license > (to become an > ASF project), but I don't see why the license is incompatible. > > > using Axion is not an option, as long as Axion doesn't > support composed > > primary keys. > > Are they aware of it? Not that it matters for your status. > > > > Why is hsqldb required to compile the code? Is the code > not using JDBC? > > > Some part of JaxMe is able to generate code by querying for > JDBC metadata. > > The test suite uses hsqldb to verify that part. And building the > > distribution includes building the test suite. > > So make the process conditional upon finding what it needs, > and generating a > warning that certain test cannot be performed because > something is missing. > I think that would be fine, too. > > > The project's STATUS file > > > (http://cvs.apache.org/viewcvs.cgi/incubator/site/projects/jax > me.cwiki), > > which needs updating. > > > How do I update that file? > > The PPMC concept makes this much easier, IMO. I think that > you could make > with either review-then-commit or commit-then-review, but > either way the > PPMC would vote on the changes. At the moment, I think that > the committers, > along with dims, could vote on the status changes as the > project's statement > to the PMC. > > > > Does any one organization control the project? How diverse is the > > > community? > > > No. Made of individuals. > > And since it will be part of the WS project, that helps. > > What would really help is for projects to adopt the philosophy that: > > (1) Seeks to get all active Committers onto the PMC in a > manner based upon their being trusted to participate > in decision making. > > (2) Accept that only PMC member votes are binding. > > (3) Every Committer has karma for every module. > > It is just one project with multiple, related, codebases. > > > > Conflict != tantrum. :-) Has there never been a > disagreement, a debate > > > over something? > > > There have been discussions and votes, but so far no true > disagreements. > > Sound good so far. :-) > > > > Have the developers generated PGP keys and started to get > signed into > the > > > web of trust? > > > That second part is the interesting thing: How do we add > our keys to the > > "web of trust"? (Not that I know what it is. :-) > > In fairness, that is something we (the ASF) need to help you with. We > cannot, in my opinion, expect you to know. There are some > proposals, but I > do not believe that anything is final. > > Are any of you geographically co-located? > > --- Noel > > > - > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Joining the ASF Web of Trust
Andy Cutright wrote: > i'm interested in the web of trust as well. i've just started following > the incubator/ infrastructure lists. is there a summary of the proposals > some where? Not that I can give you, unfortunately. Best place for you to participate in that would be infrastructure@, I guess. > i imagine there are a number of apache committers in the SF bay area, > making it possible to physically verify identity. Absolutely. :-) A question is what (else) should be trusted other than a face-to-face meeting with one or more ASF Members. --- Noel - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [VOTE] PPMCs for Incubating Projects
On Mon, Dec 15, 2003 at 02:16:49PM +0100, Nicola Ken Barozzi wrote: [snip everything above that I agree with] > Incubator PMC members not engaged in active discussion and development > on a project are on the project PPMC in quality of observers. They > should refrain from voting on PPMC decisions unless really necessary, > thus acting as vetoers of last resort. [...] > Development and discussions go on the dev lists, where the Mentors are > the ones doing active oversight. These two sub-sections are contradictory. Since technical decisions should not be occuring on any PMC list*, it is not necessary that PMCers be technical contributors. It is completely possible that non-technical contributions earn someone an invitation to a particular PMC (eg. document contributors). I don't agree that Incubator PMC members should only be second-class PPMCers. If an Incubator PMC member wishes to volunteer their time to participate as a seed PMC member on the new PPMC, then they should be a first-class member. (*in some rare cases technical discussions might happen on some PMC list in order to avoid public disclosure of a sensitive topic -- eg. security vulnerabilities. In general, however, technical discussions should always stay on the development list.) > The status update occurs on the PPMC list. Thus, the notion of reporting > to the "main Incubator PMC" is a non-issue, as all Incubator PMC members > are also on the PPMC. I also disagree with this. The purpose of the report is to present a condensed view of the happenings within a project so that others can see how things are going. If they are only posted to the PPMC list, then who will be the audience? These reports should be going to [EMAIL PROTECTED] -aaron - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: [VOTE] PPMCs for Incubating Projects
Aaron Bannert wrote: > Nicola Ken Barozzi wrote: > > Incubator PMC members not engaged in active discussion and development > > on a project are on the project PPMC in quality of observers. They > > should refrain from voting on PPMC decisions unless really necessary, > I don't agree that Incubator PMC members should only be second-class > PPMCers. If an Incubator PMC member wishes to volunteer their time > to participate as a seed PMC member on the new PPMC, then they should > be a first-class member. I concur. The "old ways" of legislating behavior die hard. We don't need them. Pardon me for quoting myself from another list: "You are a PMC Member, which means that you have been elected to provide oversight. Can't I trust you to ask for advice/ consent when you feel it is needed? How many rules must we create to legislate trust, respect and collaboration? [...] Those few rules that we need are driven by what we require for oversight and believe are best for Community Building. Do you know any community of developers that thrives on lots of rules?" We are creating the PPMC structure. I think that it is necessary, and quite possibly sufficient. Now let us let the structure do its job. If we have unforeseen some issue, we can address it when it happens. > > The status update occurs on the PPMC list. Thus, the notion of > > reporting to the "main Incubator PMC" is a non-issue, as all > > Incubator PMC members are also on the PPMC. > I also disagree with this. The purpose of the report is to present > a condensed view of the happenings within a project so that others > can see how things are going. If they are only posted to the PPMC > list, then who will be the audience? These reports should be going > to [EMAIL PROTECTED] The status report is a public document, unless there are private items, of which you can imagine some few. Discussion and approval of the status report can be on the PPMC list. Let's see how it plays out. --- Noel - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Exiting Incubation - Status Check
Noel J. Bergman wrote: That makes the project unavailable for changing the license (to become an ASF project), but I don't see why the license is incompatible. Does that mean we may even add the hsqldb.jar to the CVS repository and to the source distribution? The PPMC concept makes this much easier, IMO. I think that you could make with either review-then-commit or commit-then-review, but either way the PPMC would vote on the changes. At the moment, I think that the committers, along with dims, could vote on the status changes as the project's statement to the PMC. In other words: We, the committers, (which includes Dims) vote on a change of the status file. In the case of a positive vote Dims is supposed to record the changes? Regards, Jochen - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
