I was just wondering how long it takes to get a project approved? I sent an e-mail to Sun and faxed the signed contributors agreement to Sun also. I have created the Xith3D project page and I would like it placed into the middleware area of the game project.
So I cannot join that project now…
Also, I’ve been downgraded from ‘developer’ to ‘observer’ without notice…
What about sourceforge?
I would prefer to do it in the Java.net area, but if that does not work I would consider SF.
No one can join the project right now because it is marked “Approval Pending”. Also, your Observer/Developer status is per project.
I hate to repeatedly ask for patience (I’m sure as much as everyone hates being asked.) We have spent the last couple weeks getting the approval process in place, details of which may be found at: http://games.dev.java.net/govern.html You’ve probably read this as you sent your proposal to the dev mailing list.
Unfortunately, this group has historically been centered around these discussion forums, so there have been few subscribers to the mailing lists. And so, in trying to keep this a community-driven process, we’re still waiting for the necesssary feedback from the community before approving the project. I will bring up the possibility of moving the project approval process to the forums (instead of the mailing list) during our weekly meeting tomorrow.
In the meantime, you may want to try advertising the project here on the forums and encouraging people to send their votes to dev@games.dev.java.net
You will have to forgive me, I misunderstood the voting process.
I will ask people to vote there.
Dave
i voted for it, I did not know we were supposed to respond to project requests.
nathan
I do not understand few things.
-
There seems to be only 1 message at dev mailing list - it is xith3d announcement. Maybe sending a message there from maintainer, telling the subscribers that they are supposed to vote at all, would help ? In fact, for last month, I was not even sure if I’m subscribed to this list and if it exists at all.
-
Is Xith3d first project which requires such approval ? There seems to be quite a few games projects already, from people less known that David, but I have not seen any voting/etc about them.
My personal IMHO is that this approval process will kill java.net. On one side, you are requring at least week for approving, with possibility of one person blocking it for longer period of time - on the other hand, project is target for deletion after 30 days of inactivity. Former can scare away some people, but let’s suppose it is acceptable (as we can assume that people scared away by such ‘trivialties’ are also most likely to abandon their projects non finished soon). Second part is a lot more dangerous - does it mean that working, finished library will be deleted after month or two ? Will this also apply to jogl/joal/jinput ? I can easily imagine them freezing in few months - does it mean that they are ‘inactive’ ? IMHO, at least sum of downloads and cvs should be used (cvs shows only development, not popularity). In other case, you will just force people to make 1 empty commit each month…
I’m afraid that java.net goes into ‘community-driven commitee’ trap. I understand possible reason (so it does not look like Sun rules the site) - but I don’t think it will really work. I think it would be a lot better to just have 1-2 people at Sun with ‘god’ permissions, which would arbitrary accept/reject projects. After all, rejected project can try it’s luck at sourceforge - no reason to cry. And for accepted projects, 24 hours turnaround is a huge gain compared to comitee-decided, veto-blockable, weeks-long process.
As for the vetos, look at history. In Poland (my country), liberum veto was given to senators[1], where single voice could block entire project. Thanks to this, Poland has fallen from one of 5 most influencial countries in Europe to place where it was invaded by 3 neigbour countries and ceased to exist for over 100 years.
Discussions kill projects. Action saves them. You can be almost sure that anybody who will have to fight their way through complicated approval system and fight with vetos trying to defend himself (maybe not having anything to show at the moment), will either move to other site or abandon open-source idea for some time.
I guess next time we can try to remember to add a “Please Vote Now” to the bottom of the request until we get familiar with the process.
Hmmm yeah, when java.net first appeared, I submitted a request for my project (jME) which is not nearly as refined as David’s stuff. It was approved in about one day without any requirements. They must have recently implemented the requirements for submission.
[quote]I do not understand few things.
- There seems to be only 1 message at dev mailing list - it is xith3d announcement. Maybe sending a message there from maintainer, telling the subscribers that they are supposed to vote at all, would help ? In fact, for last month, I was not even sure if I’m subscribed to this list and if it exists at all.
[/quote]
We are currently discussing moving the project proposal process to the forums, as this is clearly the preferred method of communication within this community. It may take a few days to implement, but if people are in favor of it, I think we can make it happen.
Xith3D is the first proposal to be introduced since the new guidelines have been in place. There’s been a unanimously positive response to the proposal so far, so I anticipate that it will be approved. Hopefully most of these issues will be ironed out in time for the next proposal.
We’ve tried to make the guidelines as lightweight and inclusive as possible. The principal motivation for the voting process is to enable the community to express its interest, request clarification, voice its concerns. The principal reason for rejecting a project should be the existance of another project with similar goals, in which case it is preferable for the proect owners to join forces rather than have a host of similar, understaffed projects. As for project deletion, the intent is that an inactive project may be deleted after a month, but only if the project has had no CVS or download activity and the projects owners fail to respond to requests for project status. It is designed to reduce dead or stagnant projects, not stable projects actively in use.
The process is modelled on another Sun-sponsored open source project, JXTA.org, where it has been employed successfully for the last two and a half years. In practice the only time the voting process gets prolonged is on projects whose goals or definitions are initially unclear, and the process has enabled the proposer to refine their project descriptions so that the commmunity understands and appreciates their value. This occurs very rarely, and in most cases results in an approved project.
All I can ask is that you give the process a little time, and continue to provide feedback. Your concerns are certainly understandable, but I think in practice they are less significant than they appear to be in theory. As you’ve pointed out, the process may not be for everyone, and SF is certainly an option for those who prefer less formal controls. Our hope is that by employing a (hopefully) minimally intrusive process we can avoid a lot of fragmentation and redundancy, thereby maintaining a higher signal to noise ratio that SF permits.
[quote]Hmmm yeah, when java.net first appeared, I submitted a request for my project (jME) which is not nearly as refined as David’s stuff. It was approved in about one day without any requirements. They must have recently implemented the requirements for submission.
[/quote]
We were initially reluctant to enforce any sort of process, but we began to see proposals for projects with little or no description, questionable value to the community or unclear relevance to the community. It was decided that as this site is intended to serve the needs of the community, that we would incoporate a process that would enable the community to drive the development of the platform while increasing the signal to noise ratio.
As for existing projects, we are looking to the community to help determine the fate of them. In the very least we are asking project owners to provide detailed descriptions of their projects. If people have any other strong opinions about how these “grandfathered” projects should be handled, please let us know.
I’m not sure that I want it that way… given the problems that some people have been having with getting email notification of new messages I think the dev list is more reliable.
If you could browse the archives of the dev list easily from this site it might be less of an issue. I think things should be left as is until java-gaming.org and games.dev.java.net are more integrated. (javadesktop.org has done a pretty good job so far… you guys need to talk 
)
[quote]I’m not sure that I want it that way… given the problems that some people have been having with getting email notification of new messages I think the dev list is more reliable.
[/quote]
At the same time, we’ve been having difficulty getting people to use the dev list, as it is currently a few steps away from the forums, which seems to be a point of entry for most people. Even project specific issues seem to get more play here on the forums than they do in the project-specific mailing lists. Personally, I prefer the mailing lists, but pragmatically speaking, that doesn’t appear to be the consensus of the community. Any other opinions?
[quote]If you could browse the archives of the dev list easily from this site it might be less of an issue. I think things should be left as is until java-gaming.org and games.dev.java.net are more integrated. (javadesktop.org has done a pretty good job so far… you guys need to talk )
[/quote]
I’ll definitely take another look at javadesktop.org. In the meantime any other suggestions for simplifying the process would be greatly appreciated.
I think that for voting process, forum would be just great. One thread per project, with poll at top with comments below (one time in which these forum-based polls could be really useful ). Same rules can apply - negative votes would have to include reasons, discussion for given project would be contained in single thread which could be later closed/archived/etc. Only problem is that it would be a lot easier to vote there - I’m especially afraid of certain Sun-VRML-loader supporter, who could try to kill projects which do not help deaf children to learn finger spelling… Maybe only ‘approved’ forum members could vote ? (something like collecting n dukes without getting banned could be proof of not being a freak).
In general, I also prefer mailing lists - but this is I think one of very few boards where noise to signal ratio is very good. No point (and no way in fact) to force people to switch to mailing list now.
Flavin comes here?
looks around
Little me agrees with abies here! ;D
Maybe something like this (just thinking out loud here, ehrm, writing it down too 
How to vote on forums (sorted by rank and veto ability):
Highest rank, 1 (veto): Project owner + members with “god”/5 dukes, can vote and veto.
2 (veto): members with 3-4 dukes, can vote and veto, veto can be overruled by a majority decision of rank 1 members (e.g. atleast 3 of them or one more rank 1 than 2?)
3: members with 1-2 dukes, can vote and try to lobby higher ranks to veto by providing insightful comments on why said project shouldn’t be approved.
4: members without any duke: “casteless bastards - you’re just slaves” ;). Can’t reply to forum poll, but can view, can reply to “project notice” thread if they have something important to say.
5: non-members, can’t view.
Should be simple to implement and, in my own opinion, easy to understand system. Users who participate much in the forums are given more power, which I believe will make people more willing to answer questions/have interesting discussions (hopefully we won’t get spammers - these should be easy to notice and ban though
Any comments on this?
EDIT: some extra comments to clarify bits and pieces
Since the current ‘duke rank’ is based only on the number of messages posted, and not the content or quality of the posts I don’t think your proposed policy changes are fair.
As for how and where we do the voting… I don’t think it matters too much as long as this site and the space at java.net share a bit more info. The banner at the top of these forums has all sorts of links (home, help, search, members, profile, notification, logout) but it is not integrated with java.net…
I think the HOME link above should take me to the main Java Games home on java.net for instance.
With more/better integration info such as new projects to vote on and buttons to help get you there will be hard to miss.
There is a lot of wasted space at the top of this page. and no links to the gaming blogs, wiki, projects etc.
Hey SW,
I hear you. keep in mind that the original plan was to can this site totally and just use java.net. You guys quickly convinced us that this was a baaad idea, but now we’re in catch up mode.
Keep on at us and once the amazing Space Ghost is up and around again I’m sure these things will start getting addressed
JK
[quote]Since the current ‘duke rank’ is based only on the number of messages posted, and not the content or quality of the posts I don’t think your proposed policy changes are fair.
[/quote]
-
First important distinction is to sieve out random people just wanting to mess. I think that requiring umpteen posts to other forums is a reasonable rule - just to make it not so easy to just reregister after banning and messing again.
-
As for the more power for more frequent posters… of course number of posts has no direct connection with how ‘good’ given person is. But if you will look at ‘Gods’ of this board, I think that you will agree they are not a bunch of random people. Plus, number of posts is a kind of useful metric - amount of experience with this board (how many posts you have read and replied to). You do not have to be super brilliant to just reply - but in the process of producing few hundred replies, you will read lot enough to know which projects should get accepted and which should not. We are not proposing giving some uber power for multi-dukes people - just a bit more important voice in voting for projects which are in some way based on this board.
Maybe strict layering proposed by Anders is an overkill, but IMHO requirement of few dukes before being able to veto is a good idea.
Just to add another voice: I would prefer the mailing list.
I know some of you will object, but I actually find this forum a little cumbersome. The high amount of mouse clicking required is an inherent problem of web based forums. I just like my Gnus mail & news client better, it gets less in the way than the browser interface. Especially while composing posts Am I totally lost with this opinion or is there someone feeling the same way?
The best of both worlds would be if you could read this forum with a news client via nntp. Is this possible with YaBB?
Hmm, Gnus has a module for reading Ultimate BB, perhaps I can hack it up. Wouldn’t solve the post editing problem, though.
BTW. Flavin has been here already… just before Java One.