Session Start: Mon Jul 07 20:55:04 2008 Session Ident: #twiki_release [20:55] * Now talking in #twiki_release [20:55] * Topic is 'http://twiki.org/cgi-bin/view/Codev/GeorgetownReleaseMeeting2008x05x12' [20:55] * Set by OliverKrueger on Mon May 12 21:02:18 [21:01] * uebera|| has joined #twiki_release [21:03] Good evening/afternoon everyone [21:03] Hey there, welcome to the show. [21:03] hi koen, kwang, kenneth, michael, oliver, sven (here?), tom, markus [21:03] Hi, Peter. [21:04] hi peter [21:04] good participation today [21:04] :) [21:04] * LarsEik has joined #twiki_release [21:04] i may leave early, am a bit sick today :( [21:04] arthur sent an e-mail that he can't join, no internet in his temp apartment in germany [21:04] hi lars! [21:05] Nabend [21:05] OliverKrueger, lol [21:05] hi all. [21:05] oh, kwang, drink lot of hot tea to sweat out the bugs! [21:06] shall we give it two more minutes? [21:06] yup, i do :) [21:06] who is taking notes, who is facilitating? [21:06] I'll take the notes as always ;-) [21:07] thanks kenneth! [21:07] * SopanShewale has joined #twiki_release [21:07] shall i facilitate again? [21:07] hi sopan! [21:07] Hi Peter - Good morning [21:07] most be middle of night for you [21:07] thanks for joining! [21:07] yup. .. 12:40 midnight [21:08] wow! effort! [21:08] thx [21:08] 5:06 for Sven, so he is deep asleep [21:08] dreaming [21:08] wow..thats gr8 Sven [21:08] * Carlo has joined #twiki_release [21:09] Hi Carlo [21:09] can someone please set the intro topic to http://twiki.org/cgi-bin/view/Codev/GeorgetownReleaseMeeting2008x07x07 [21:09] Hello everyone [21:09] ... dreaming of me sitting on his hunchback [21:09] :) [21:09] hi carlo, nice to see you here! [21:09] * OliverKrueger changes topic to 'http://twiki.org/cgi-bin/view/Codev/GeorgetownReleaseMeeting2008x07x07' [21:09] tks! [21:10] proposed agenda: [21:10] # 1. Introduce new TWiki Governance [21:10] # 2. Review Urgent Bugs with 4.2.1 scope [21:10] # 3. Feature requests for Georgetown Release [21:10] anything to add? [21:10] that's already a lot [21:10] yep! [21:10] especially the first one :-) [21:11] without further delay, let's start [21:11] ---++ 1. Introduce new TWiki Governance [21:11] i have been working for quite some time on this [21:11] finally i have something to present [21:11] thanks a lot four your proposal, peter [21:11] worked whole sunday (jet lagged) [21:11] thanks [21:12] http://twiki.org/cgi-bin/view/Codev/IntroducingTWikiGovernance is the main entry point [21:12] i assume everybody here has read it? [21:12] actual governance topic is http://twiki.org/cgi-bin/view/Codev/TWikiGovernance [21:12] not in every detail [21:12] sure but it takes some time to digest and understand it, not sure if we get what you are trying to say. [21:13] the proposal is not yet fully baked [21:13] so for now feedback is more of a first impression than a more thorough feedback as it deserves it. [21:13] i need your support [21:13] Peter I understand that what you want from the community is feedback and proposals for improvements. [21:14] yes [21:14] many details need to be defined [21:14] and more to add [21:14] for example, the initial focus teams [21:14] and the members of the community council and technical board [21:15] and the mission statement [21:15] some things might even be better on another discussion topic as it is not governance/organization related [21:15] the structure copies^H^H^H^H^H^borrows heavily from the ubuntu project [21:15] hm i see [21:15] mark s introduced me to jono bacon, ubuntu community lead [21:15] peter, did you read CDot's comparison of ubuntu vs twiki? [21:16] yes [21:16] what do you think about it? [21:16] i will address his comments, i do not fully agree [21:17] most importantly, mark forked from debian to build a new community [21:17] On the other hand I do not see the new proposal as exactly the same as Ubuntu, so instead of discussing if TWiki is like Ubuntu it is probably better to focus on the actual proposal [21:17] in which areas should we 'borrow' hte ideologies of ubuntu? [21:17] i am the founder of twiki.org who introduces a commercial aspect to the open source twiki project [21:17] that is the main difference [21:18] yes, kenneth [21:18] lets spend some time now on the actual proposal [21:18] what is your initial impression? [21:18] this is a question to all [21:18] I really don't like the proposed mission statement and that it was changed without any discussion. That is certainly not helpful given all the governance issues we had already. [21:18] First I am happy that it is here now. Thanks for that [21:19] good point Lavr [21:19] Carlo - it is a proposal! [21:19] and that is my comment to the proposal [21:19] it was not marked as a proposal [21:19] here is the link: http://twiki.org/cgi-bin/view/Codev/TWikiMission [21:19] hm, I had the impression that this is too much a mixture of different issues. most of them are not related to governance aspects from a point where we need th shape twiki as an organization. [21:20] true [21:20] Sounds interesting Peter-some changes are happening, we will surely move to some new direction [21:20] yes, the changes are more than just governance [21:21] how would you define web2.0 in twiki's context? [21:21] see goal statement in http://twiki.org/cgi-bin/view/Codev/IntroducingTWikiGovernance [21:21] in some sense it fails to analyze why twiki as an organization failed [21:21] web2.0 means alot in different areas and industry. and i fail to see its definition in twiki's framework [21:21] and why we need to rethink governance [21:22] the commonly understood way is the label web 2.0 as "user generated content" [21:22] twiki is that, and on top of that adds "user generated application logic" with the twiki app feature [21:22] a lotus notes for users, not IT [21:23] Blogs, forums, wikis etc are commonly known as Web 2.0. Web 1.0 was the read only web. [21:23] sorry but to me "Enterprise Web 2.0 platform" means nothing [21:23] I am afraid web2.0 is already "out", same as e-commerce is out, or people talking about e-commerce in presentations are not up to the next hype ;) [21:23] yep [21:23] Enterprise2.0 [21:23] i guess it's more of http://connections.thepodcastnetwork.com/2008/01/19/connections-018-improving-performance-profitability and http://www.slideshare.net/slgavin/meet-charlie-what-is-enterprise20/ [21:23] please no 2.0 or 3.0 or whatever [21:23] * AndreU has joined #twiki_release [21:24] ya, Enterprise2.0 is very hype. but not much people have to say something substantial about it. [21:24] the enterprise web 2.0 vision is best depicted at http://twiki.org/cgi-bin/view/Codev/TWikiPresentation2007x10x22#Slide_15_Enterprise_Collaboratio [21:24] here's my 2 cents as the relative newcomer [21:24] Mission statements should last the next hype, so shouldnt contain such hype terms in my oppinion. [21:24] I don't think that TWiki has "failed" [21:25] However I think that TWiki, outside the community of core contibutors, is perceived as engineering centric and wiki centric [21:25] That has allowed oher companies and oher projects to steal TWiki's thunder [21:25] I do think that it is extremely important to the project to define an expanded vision [21:25] I agree [21:26] whether we call it Enterprise Web 2.0 is somewhat irrelevant (although we must clearly communicate to be understood properly) [21:26] but Enterprise Web 2.0 platform is bullsh*t bingo [21:26] true [21:26] to me, what Web 2.0 means is that there is a seamless user experience across a range of tools [21:26] hi andre, welcome! we are in the middle of the new governance and mission statement discussion [21:26] hm [21:26] for example, RSS feeds, discussion groups, social networking context, the wiki itself, the structured application platform [21:27] all of this needs to have a well through through data model, security model, presentation model, and in fact I like the idea of publicizing a broader API [21:27] in the consumer world, people talk about the OpenSocial API (I am sure you are familiar) [21:27] something similar is required in the enterprise world in order to attract more developers to the project [21:28] sounds reasonable [21:28] there are some common questionair about an organisational (in)effectiveness: (1) the ability to set collective, realostic and achievable goals, (2) the ability to make detailed plans, organise and manage resources and coordinate actions (3) the achievement, partial or otherwise of those goals (4) a degree of consistency in extent to which goals are achieved (5) the ability to make appropriate changes to plans and actions (6) the [21:28] ability to identify and respond appropriately to opportunities and risks in the market [21:28] the expanded mission will hopefully attract more developers since it offers some sexy new projects [21:28] is the proposal then, to have a new mission statement that correlates with current hype woudl bring in more developers? [21:28] s/woudl/to/ [21:29] IMHO, twiki has enough in its basket to fulfil before adding new projects that stacks ontop of codes that seriuosly requires alot of work [21:29] (hm, apparently typing too slow) The term "next generation Enterprise platform"--or, using British understating, "modern Enterprise platform"--seems motr reasonable to me, too. What's important is that we can back this description with good examples/explanations. [21:30] that said, twiki doesn't lack any devs, it just lacks dev participation at large. [21:30] note that there are only two lines changed in the mission statement page: the one liner mission statement (expanded vision), and expanding "Kick ass" to "Kick ass Web 2.0 extensions" [21:30] twiki has a lot in its basket, but if we get more developers we get more done [21:30] so core mission is the same, just expanded [21:31] peterthoeny, can we collect these ideas offline, please and move on. we get stuck. [21:31] also, important to note that the web 2.0 vision happens mostly on the extension/app level [21:31] looking at Codev/ConsultantsForHire, twiki has a very large resource available. [21:31] i think it's better to focus in how we harness that resource rather than not. [21:31] hi, ja I have to catch up with reading now [21:31] e.g. i do not see a need to blow up our open source distro with a full web 2.0 stack [21:32] lets not waste more time on a new mission statement. thats offtopic anyway, IMHO. [21:32] I personally would like to see more developers from the user community. There is an inbalance of independent consultants vs people like me. And I would not want to have less consultants so increasing user participation is what I would like to encourage [21:32] MichaelDaum++ [21:32] but it make sense to create add-on packages, one with web 2.0 tools, one for crm, etc [21:32] MichaelDaum++ Lavr++ [21:33] let us focus on point (1) of our agenda again, please [21:33] that's where we are, right? [21:33] * ktwilight nods [21:33] sorry, my bad :) [21:33] ok, let's go bck to the governance question [21:34] time check: +30 min [21:34] peter, I'd really like to see this document to be worked out properly! [21:34] what question do you guys have on your mind? [21:35] Peter share some of your thoughts about the TWiki Community Council. [21:35] yes, me too, i had very crappy internet connectivity in the swiss mountains, i could not achieve as much as i liked [21:35] over the next few days i will work on more details [21:35] The driving question imho is: Is twiki as an organisation effective enough and how can a change in governance improve the situation! [21:35] for example, the TWikiCommunityCouncil page, the CodeOfConduct page etc [21:36] how can we restructure twiki as an organisation to be most effective in the sense I've outlined above? [21:36] frankly, we as a community are pretty much stalled at this time [21:36] this would be a good opener for a governance proposal. [21:36] I understand the TCC as a small board that rarely need to meet. [21:36] the new governance structure brings new wind and new opportunities [21:37] well, the tcc needs to be active initially to define the processes [21:37] peter, do you think your proposal is build on a solid understanding of the situation? [21:37] then they can meet less frequently [21:37] i see the tcc as the replacement of the core team [21:37] Is this a 5 person board or so? [21:39] this size sounds reasonable, yes [21:39] is it safe to assume the current core team to be transfered as tcc? [21:39] Sounds good. [21:39] it would be nice to have crawford here, he as the key architect has a say in the technical board [21:40] let's not forget about sven ;) [21:40] The current core team is a mix of technical and org type people. It may be better to divide the current team members out on the other boards. [21:40] as well as arthur [21:40] so i guess there won't be any overlaps of a team to another? including the council? [21:41] The TCC people should be interested in organisation more than technology. [21:42] there might be overlap, folks who are strong in community aspects and technical side might be in both [21:42] what kind of calibers are we looking for in TCC? [21:42] and the team size for each Team? is there a minimum/maximum head count? [21:42] someone like kenneth is a candidate for the tcc, he is a clear people leader [21:43] these are good discussions that help me refine the governance structure [21:43] keep going :-) [21:43] i think it's important to define the roles of each team/council (and its members) [21:44] absolutely! [21:44] I will be happy to accept. And I would propose Michael Daum for TCC as he has shown interest in organisational matters. [21:44] even if it's not possible, at least a list of tasks of each Team/Council [21:44] I would be happy to join. [21:44] how much do the members have to say/decide? [21:45] the tcc and tech board members are appointed initially [21:45] i think they should be elected when the time is right [21:46] What would typical decisions look like? [21:46] criteria for election is size of community and trust in community [21:46] both need to be built up [21:47] time check: +45 min [21:47] shall we spend 5 more min on governance? [21:47] How do you measure "trust"? [21:47] we need a lot more than 5 more minutes, but not today. [21:47] there are additional codev pages i changed this morning [21:47] any feedback on those? [21:48] lets find some good closing for this new starter [21:49] overall I think we will all do what we can to help you, peter, in refining, extending and sharpening the new TWikiGovernance initiative [21:50] thanks michael! [21:50] your support is very much appreciated! [21:50] with the goal to get the thing rolling with lots of fun and much of succecs for all of us. [21:50] amen to that [21:50] I think we should make a plan for some follow up meetings the next few weeks - only on Governance. And make them voice calls. [21:50] :) [21:50] that sounds like a plan [21:51] how about using next monday marketing meeting slot? [21:51] yep [21:51] yep [21:51] okay [21:51] yes, conf call where we hear voices sounds good for that meeting [21:52] kenneth, could you sponsor the bridge? [21:52] sounds good [21:52] skype? [21:52] our bridge is crappy for intl calls [21:52] I would also suggest a call maybe on wednesday ONLY where the scope is Q & A only. [21:52] I can sponser the call on our Moto bridge [21:52] cool, thanks kenneth! [21:53] no, lets do the call in one week, instead of the mkt meeting [21:53] OK. Let us do that [21:53] so we should be expecting some more changes within this week? [21:53] cool [21:53] yes [21:53] great [21:53] well, s/changes/refinements/ [21:53] well, yes. :) [21:53] so ... we are nearly at the end of #1 of our agenda, finishing up in good spirit. now lets enter #2. [21:53] shall we move on to next agenda item? [21:54] OK [21:54] ---++ 2. Review Urgent Bugs with 4.2.1 scope [21:54] * TomBarton has left #twiki_release [21:54] ... [21:55] emergency call [21:55] Michael the work is yours [21:55] word [21:55] sorry [21:55] I would have liked to talk to CDot & Sven about it [21:56] we need a decision and a commitment [21:56] the situation is: [21:56] the current user code is broken [21:56] soo, what busg need to be fixed for 4.2.1? [21:56] "bugs" [21:56] have a look at Item5118. [21:57] this is the most central bug. most others depend on it [21:57] http://develop.twiki.org/~twiki4/cgi-bin/view/Bugs/Item5118 [21:57] there are two possibilities to resolve the situation [21:57] either we go on fixing the current code [21:57] or we revert it to 4.1.2 [21:57] * gmc has left #twiki_release [21:58] CDot has signalled good faith in the code [21:58] thinking it is still reasonable to go forward instead of reverting [21:58] my personal threshhold for fixing vs rewriting is lower [21:58] There is a problem with reverting - even though that would be the most fair action.... [21:58] * PeterThoeny_ has joined #twiki_release [21:59] the user name in the topics is not CUID!. So it is not backwards compatible to people already on 4.2.0 to revert [21:59] I try that again [21:59] the user name in the topics is NOW CUID!. So it is not backwards compatible to people already on 4.2.0 to revert [21:59] that was better [22:00] So we are forced to fix. I have actually made some progress this weekend and have some not yet checked in code changes. [22:00] I am not done yet and I will not check in a half cooked solution. [22:01] we changed it from login name --> wikiname --> cuid [22:01] no matter what we do with 4.2.1 we cause upgrade pains for customers [22:01] I have made some improvement of the code Michael did. Could not have done it without your changes first Michael ;-) [22:02] I still need to fix the Group case. If a login name is kenneth.lavrsen (this is common format many places) the access rights do not work. [22:02] i would not oppose a change back to wikiname as in 4.1.2 [22:02] The code strips off anything before the . [22:02] well, there might be a chance to fix things. however, the code will still conserve the mess. [22:03] wouldn't 3rd party logins depend on cuid? [22:03] Peter. Problem is I now have tons of topics with login name. And I do not want to upgrade to 4.2.1 and then see login names instead of wikiname everywhere [22:03] ah, that stripping is probably done to cut off web prefix from wikiname [22:03] Yes. But the stripping can be changed to strip off {usersweb}. instead [22:03] actually having login names somewhere else despite ... during login ... seems to be wrong. [22:04] still, a Main.John login name clashes with login name John, even without CUIDs. [22:04] because login names get stripped even though they are no wikinames. [22:04] what are CUIDs? [22:04] only wikinames should be treated that way. [22:05] canonical user ids [22:05] I think we can live with the limitation that you cannot have Main. as prefix of a login name. [22:05] cheres [22:05] * peterthoeny has quit IRC (Read error: 60 (Operation timed out)) [22:05] I do not know anyone called Main ;-) [22:05] i'd prefer that login names would not go through the strip-off regex [22:06] They don't! [22:06] oh they do [22:06] It is the access rights that are stripped. [22:06] The login names are converted. [22:06] kenneth.lavrsen becomes kenneth_46lavrsen [22:07] the getWikiName strips the login name if you are not registered but that function should not at all be used for access rights. [22:07] yea right. but problem is that the internal api tries to be polymorph, as it is called with a $user string which is a wikiname, a login name or a cuid, sometimes [22:08] Yes the API is totally unclear which is why it is so difficult to fix the problem. You fix one thing and you break something else. [22:08] hm [22:09] time check: +66 min [22:09] we need CDots view here. so let's continue the analysis on a separate user code meeting. [22:09] Maybe we can try and IRC chat about 5118 in a near future - in early evening where I am home from work and CDot is not yet punched out from IRC. [22:09] at a certain point I think: let's do both (1) hotfix the current code (2) take one step back and rewrite it. [22:10] can kenneth, sven, michael and crawford take this offline? [22:10] Yes. We need to hotfix - and then rewrite for a 4.3 or 5.0 [22:10] I will create a separate call, following the emergency call email to organize a user code meeting. [22:10] 5.0 :-) [22:10] 'And next time design first. You cannot design with unit tests. We need to design the API. [22:11] so true [22:11] amen! [22:11] We also need to address the security part for 4.2.1 [22:12] Even if it means some limitation to exotic user IDs [22:12] there's even no proposal how to fix that, i.e. the function forceCUID() [22:12] anyway lets stop here [22:13] what other bug items should we cover here? [22:13] One moment. I have a list [22:13] of 2 [22:14] There are around 13 urgent bugs open and they are in 3 groups. [22:14] Misc which are [22:14] * TWikibug:Item5382 - TWiki.pm sets wrong urlHost if https protocol and ShorterUrlCookbook is used [22:14] * TWikibug:Item5555 - Signature in the copy/paste field when editing is suddenly in square bracket form. (trunk only) [22:14] * TWikibug:Item5729 - a redirect quits cgi process before finishing the twiki object [22:14] 5555 is closed [22:14] http://develop.twiki.org/~twiki4/cgi-bin/view/Bugs/Item5382 [22:14] http://develop.twiki.org/~twiki4/cgi-bin/view/Bugs/Item5555 [22:14] http://develop.twiki.org/~twiki4/cgi-bin/view/Bugs/Item5729 [22:15] 5382 is one Sven wanted to fix but did not. There is a fix proposed and it does not look too complicated so if someone want to spend an hour bug fixing.. [22:15] I will comment on Item5729, maybe more people are able to check. [22:16] 5279 that is your Michael - yes [22:16] Is there a release blocker bug in that? [22:16] this nearly drove me crazy [22:17] You have a fix ready? [22:17] if CGI.pm quits as soon as the redirect gets printed to STDOUT, then yes. [22:17] nope. the fix would be to delay writing out the redirect until the twiki object(s) are finalized [22:17] i think i should start contributing in bugs [22:17] and I don't know how to do that without restructuring the code massively [22:18] This problem is mod_perl related right? [22:18] i will have a look at 5382 [22:18] Thanks Sopan. [22:18] not only mod_perl releated [22:18] cool, thank you sopan! [22:18] it shows up as memory leaks under mod_perl, but also as unflushed session files. [22:19] SopanShewale, :) [22:19] I see. [22:19] SopanShewale, it'd be great if you can! [22:19] Any dirty workaround Michael for 4.2.1? [22:19] would a sleep work before the redirect? waiting 0.2 sec does not make a difference to users on a redirect [22:19] not by me [22:20] Item 5279 still needs a "is this really so" check [22:20] OK. Maybe I should continue the list if "2" [22:21] Then there are 2 more groups. Next is the "User code related bugs". [22:21] 5 of them [22:21] * TWikibug:Item5118 - Difference from 4.1.2 - 4.2: Apache loginname no longer works with access control lists [22:21] * TWikibug:Item5704 - Performance tuning of HtPasswdUser for large number of users [22:21] * TWikibug:Item4824 - Security bug: TWiki's user management code needs rework again [22:21] * TWikibug:Item5451 - MailInContrib reveals error in Password.pm, that has been carried over to LdapContrib [22:21] * TWikibug:Item5443 - Searches related to TOPICINFO don't work as expected [22:21] in any case getting 4.2.1 out with a fixed Item5118 is much more important than getting Item5279 fixed. [22:22] Peter 5704 is waiting for feedback from you. Not sure why Sven did that. [22:22] Can it be closed? [22:22] http://develop.twiki.org/~twiki4/cgi-bin/view/Bugs/Item5118 [22:22] http://develop.twiki.org/~twiki4/cgi-bin/view/Bugs/Item5704 [22:22] http://develop.twiki.org/~twiki4/cgi-bin/view/Bugs/Item4824 [22:22] http://develop.twiki.org/~twiki4/cgi-bin/view/Bugs/Item5451 [22:22] http://develop.twiki.org/~twiki4/cgi-bin/view/Bugs/Item5443 [22:23] Item5704 is not even addressed correctly. that's basically because group memership is checked using a list operator instead of a hash lookup [22:23] again not fixable without rewriting larger parts of the user code :( [22:24] So can we close it in a 4.2.1 scope downgrade it to Normal for the 5.0 rewrite of user code? [22:24] yes, a lot of performance tuning can be done in user code [22:24] yes, i think that is the pragmatic approach [22:24] ... related to getting away with isInList() [22:26] * Lynnwood has joined #twiki_release [22:26] another inefficiency is induced by the feature to mix two user mappings, one of witch is the BaseMapper all the time, providing TWikiGuest and TWikiAdminGroup etc [22:26] michael, your fix for 5704: [22:26] - if( $line =~ /^(.*):(.*)(::(.*))?/ ) { [22:26] + if( $line =~ /^(.*?):(.*?)(?::(.*))?$/ ) { [22:26] did you measure the performance impact of greedy to non-greedy match? [22:27] nope. I only reverted the regex so that I was able to continue to work on my dev engine. so my fix was rather ... a pragmatic. [22:27] i think the greedy match might have been the main goal for the sun guy on performance tuning (not sure) [22:28] Next is 4824 which MUST be addressed. An installation where people can register with new login names is insecure. If you can only authenticate via a company LDAP that you cannot create new login in then it is safe abough I guess [22:28] hi lynnwood [22:28] Greetings all [22:28] Hi Lynnwood [22:28] THe greedy match failed as I understand it [22:29] didn't we discuss 4824 as part of 5118? [22:30] Yes. I have nothing else to add. We bring this up when we talk to Crawford about 5118 [22:30] okay [22:30] he did a good summary on twiki-dev! [22:30] Yes. And I confirmed that it is possible to authenticate with different logins against the same access right :-( [22:31] let me say, that I am very sorry for not having talked to twiki-security first. [22:32] lets move on it is getting late [22:32] 5451 it this one you look at Michael being LDAP [22:32] ? [22:33] * MichaelDaum clicks [22:33] yep, will check in the fix. [22:33] Thanks [22:33] time check: +90 min [22:33] * Carlo has quit IRC ("ChatZilla 0.9.83 [Firefox 2.0.0.15/2008062306]") [22:34] And then the last group. There are 5 wysiwyg bugs [22:34] They are either setup problems most likely. [22:34] Or UTF8 related. [22:35] I think - based on the input from Richard - that we should be happy with the UTF8 we have now for 4.2.1. And scope rest of utf8 for 5.0 [22:35] gr8! [22:36] Do we agree on that assessment? [22:36] i have not looked at the details, but i trust richard [22:37] I am sure now that Crawford is back from vacation that he will address the Wysiwyg items he can. But other than that it is mainly closing the user stuff that gates a 4.2.1 release. [22:38] I have no more. [22:38] ok, good [22:39] that means next agenda item? [22:39] ---++ 3. Feature requests for Georgetown Release [22:40] I do not think anyone has prepared any new proposal for decision. And I personally do not want to drive other peoples proposals if they are not here. [22:40] this is 5.o, right? [22:40] Yes. [22:40] can we collect a _detailed_ roadmap of the accepted proposals? [22:40] we do have a high level roadmap but no detailed one as far as I know [22:40] There is a search that shows all the accepted proposals [22:41] are you talking about the detailed spec on the proposal topic? [22:41] last time I looked it was more of a collection of personal roadmaps, could be wrong. [22:41] The roadmap topic only shows THE roadmap with the headlines and a search of personal roadmaps that I personally think are not related [22:42] right [22:42] open feature proposals are listed at http://twiki.org/cgi-bin/view/Codev/TWikiFeatureProposals [22:42] But the ACCEPTED proposals are here [22:42] http://twiki.org/cgi-bin/view/Codev/TWikiFeatureProposals#Accepted_Feature_Proposals [22:42] 24 of them so it is not scope we miss for 5.0 :-) [22:43] I'd prefer something that you can attract sponsors for 5.0 instead [22:43] people can't say in which state we are with our plans for 5.0 [22:44] That would be the larger scope work which is better covered by the roadmap and which needs detailed spec and design. [22:44] I was hoping to make this work the key work at the September summit in Berlin. [22:44] kind of upcoming features [22:44] yes kenneth, that is a good use of time in the berlin meeting [22:45] last but not least this is also marketing related, working in both directions: outwards + inwards [22:45] TheRoadmap should be our central point to get back in line, when we tend to get lost on the way. [22:46] some things on a roadmap can be "looking for contributors" so that people that like to join 5.0 could get attracted [22:46] like: where can I help [22:47] We created the first in Rome. And in Sunnyvale noone wanted to change the headlines. I think it is good to use the Summit to maintain the roadmap and to do some actual work getting the headlines turned into tasks. [22:47] agreed [22:47] seems as if that's our way. [22:48] however [22:48] in general I'd prefer to get detailed roadmaps out sooner [22:48] time check: +106 min [22:49] We are less than 2 months from the next Summit. [22:49] some projects come up with a detailed roadmap as soon as the last one has been finished and the release is out. [22:49] By the way: the hotel prereservation expired on July 1st. We have to book soon. [22:50] So we should focus on getting the Governance issue and 4.2.1 done before so we meet in Berlin - with conflicts resolved - and 4.2.1 shipping so we can focus on 5.0 [22:50] Good point Oliver. [22:50] argh [22:51] What names was the booking done in? [22:51] I used my name. [22:51] sorry. [22:51] OK. I'd better call and book myself in then :-) [22:51] Didnt get ur question. :) [22:52] I did not mention Moto. [22:52] oh, i need to book the hotel too! [22:52] I meant - the block booking was that in your name - in case they still hold the rooms. [22:52] Yes, I used my name. [22:52] OK. [22:52] yes, goal is to have governance and 4.2.1 settled by the berlin summit so that we can focus on 5.0 [22:53] we should also settle the twiki.org redesign asap [22:53] I had an email conversation with Mrs Sommer. [22:54] redesigning twiki.org is nearly as important than all technical developments for 5.0 together [22:55] yes, this is important, it drives more users to the twiki project [22:55] PeterThoeny_, can you ask Arthur what his plans are on materializing the new design? [22:55] yes i will [22:56] ok, anything else? shall we close the meeting? [22:56] I have nothing more [22:56] one last question [22:57] any updates on the twiki.org servers? I get more and more internal server errors on twiki.org. [22:57] me, too [22:57] we currently procure a new x86 server [22:57] server should be ready in a week or so [22:57] Will be Centos based? [22:57] yes [22:57] this is rpm, right? [22:58] so, familiar env for us geeks [22:58] Good ;-) Then I will feel home [22:58] rpm, yes [22:58] * SopanShewale has quit IRC (Read error: 110 (Connection timed out)) [22:58] :( [22:58] server is quad core x46 with 8gb ram and 500gb raid 1 disk [22:58] Centos is RedHat Enterprice - just without RedHat logos and branding - and no support naturally [22:59] why did you choose Centos? [22:59] centos is very solid [22:59] that's no distinguishing feature, is it? [23:00] RedHat is know to be stable. But since RedHat became subscription only and they created Fedora then Centos became a good alternative [23:00] i believe centos is 99% redhat, without some proprietary stuff (monitoring tools?) [23:00] It isn't? ;) [23:00] Fedora is not security supported more than 12 months so it is bad choice for servers [23:01] most rpm systems are bad in coping with time, so to say [23:02] time check: +120 min [23:02] In practical for a webserver - it works pretty cool with the rpms also. [23:02] ok, lets close the meeting (and carry on the discussion in #twiki) [23:02] I need some sleep [23:03] Other alternative would be a debian family distro. But the advantage of Centos is that t.o then runs same as most of our large user bases [23:03] thank you all, this was a productive meeting [23:03] Corporations live RedHat for some reason [23:03] i think we can make this work [23:03] Yes. Thanks. It was so encouraging to see so many today. Thank you. And thank Peter for the proposal.