Session Start: Mon Apr 23 22:00:54 2007 Session Ident: #twiki_release [22:00] * Now talking in #twiki_release [22:00] * Topic is 'http://twiki.org/cgi-bin/view/Codev/FreetownReleaseMeeting2007x04x23' [22:00] * Set by sven__ on Mon Apr 23 11:21:56 [22:01] Good evening [22:03] dito. :) [22:03] Sven are you here real live or just lurking? [22:05] Trying to see if people arrive. Just issued a reminder in #twiki [22:05] * ArthurClemens has joined #twiki_release [22:05] Hi Arthur - so now we are 3 [22:05] hi kenneth, oliver and sven [22:06] Hi Arthur, hi Kenneth, hi Sven [22:07] * PeterThoeny has joined #twiki_release [22:07] Hi Peter [22:07] Does TWikiFeature04x02 have a complicated search or is t.o. slow? [22:07] hi oliver, sven, kenneth. arthur! [22:07] just rushed to a public library in downtown san jose [22:08] A search has just been added to TWikiFeature04x02 so I could list the items ready for discussion [22:08] t.o. is slow at the moment [22:09] By tomorrow it should be the end of hacking TWikiFeature04x02. The new form is ready and I have updated all the open proposal items that have a committed owner. [22:09] * PeterThoeny has quit IRC (Read error: 104 (Connection reset by peer)) [22:09] * peterthoeny_ has joined #twiki_release [22:09] i would like to add AddspaceOutWikiWordtoFunc [22:09] because it is such a nobrainer [22:10] can I just change the status of that topic? [22:10] where are we in the meeting? [22:10] (sorry for the delay) [22:10] at the very beginning [22:10] We are just getting the last people to join. [22:10] we are waiting for momentum [22:10] or even before that [22:10] ok [22:10] who is doing the minutes? [22:10] SteveRJones seemed to try to join but is fighting the client. [22:11] I am already on the minutes [22:11] great [22:11] i asked a few folks if they are interested in the customer advocacy role [22:11] hopefully steve can join [22:11] the meeting [22:11] * SteveRJones has joined #twiki_release [22:11] hi Steve [22:11] speaking of the devil ;-) [22:11] hi steve [22:12] hey there! [22:12] hi steve [22:12] :-) [22:12] 6 people now [22:12] Welcome Steve. [22:12] i just said that i hope you can join, and voila, here you are :-) [22:12] I think we can start [22:12] who is facilitating? [22:12] I propose YOU Peter [22:13] well, I know I'm not! [22:13] second [22:13] ok [22:13] it's my pleasure to do that [22:13] Dynamic duo as always [22:13] so lets start, we have critical mass [22:13] let me start by saying its great we have much activity on twiki [22:14] proposed agenda: [22:14] # 1. Action Item Review [22:14] # 2. Next Release 4.2 or 5.0 [22:14] # 3. Review Proposed Features [22:14] anything to add? [22:14] Not from me. [22:14] yes, we are building up momentum recently [22:14] lots of activity in the last 2 days [22:14] thanks to kenneth work on the feature tracker [22:15] also on Bugs, thanks to CDot mostly [22:15] ok, lets start with first [22:15] ---+ 1. Action Item Review [22:15] well peter, your point [22:15] Yes. CDot has done a great job catching up with old bugs. [22:15] - Peter: Create a TWikiBetaTesting topic in Codev, with info for beta testers. Need to recruit beta testers [22:15] this is pending, needs to be carried over to next meeting [22:16] - Kenneth: Looking to add at least two more people - preferable non-developers to share the role as customers advocate [22:16] i invited 4 people yesterday [22:16] if it is created it should be on twiki.org homepage as well [22:16] steve is interested, thanks steve!!! [22:16] yep! [22:17] Thanks Steve. Steffen and I need help so we can share the load on more people. [22:17] I'll do what I can [22:17] And with the new TWiki app I made yesterday it should go much easier. [22:17] steve, do you want to introduce yourself quickly? [22:17] thanks kenneth for that [22:17] Sure. [22:17] yes, excellent work, kenneth [22:18] I work at Freescale (spun from Motorola) and inherited the old Twiki site from CDot [22:18] aka - Crawford [22:18] how many emploees at freescale? [22:19] I do this is my sparetime, we are running Cairo, and we have approx 1500 registered users (we require registration) [22:19] how big is the wiki? [22:19] 20,000 employees [22:19] a BIG spinoff [22:19] We have approx 30 separate webs and growing daily. [22:19] any other wiki champions at your job? [22:20] We need to go to 4.x -- people are clamoring for "Web 2.0" stuff [22:20] Freescale is the old Motorola Semiconductors in case someone do not know :-) [22:20] (psst that is TWiki 5) [22:20] with sexy twiki apps :-) [22:20] Just got a requesst to add another advocate [22:20] Twiki 5 is Web2.)? [22:20] 2.0? [22:21] twiki IS web 2.0 [22:21] Depends on how you define Web2.0... ;) [22:21] twiki 4.1 is already 4.0 if you add blogs, headlinesplugin etc [22:21] just needs nicer packaging [22:21] but not what people think is web 2.0 (gradients, big fonts :-) [22:21] s/4.0/ web 2.0/ [22:21] Ahhhh, well, people here are ready for something more graphical (ack) than Cairo. [22:21] in fact any wiki is 2.0 [22:22] any textarea is 2.0 >;-> [22:22] that will be a huge backup [22:22] steve, what is your personal interest in twiki, e.g. where would you like to have it evolved? [22:22] people look at wikimedia (?) and compare to Cairo. Really, we just need to upgrade. [22:23] Hmmm, evolve is a good question. From what I see the big impediment is learning the syntax of the markup language. People like bringing up websites very quickly. [22:24] so, a good wysiwyg editor [22:24] Also, wikis are generally not good at providing access controls that easily scale [22:24] yes, wysiwyg is looked for [22:24] Well Steve. We can take a chat maybe tomorrow about the advocate role and I am also happy to help if you need advice/help about the upgrading from Cairo to 4.1.2. [22:25] But corporations these daze are big into Intell Property protection. [22:25] * McAldo has joined #twiki_release [22:25] hi [22:25] Hi Torsten [22:25] hi torsten! [22:25] welcome to the meeting! [22:25] thanks [22:25] Hi Torsten [22:25] ok, lets continue with the action item review [22:26] - Arthur/Kenneth/Peter: Need to update the TWiki application so the process becomes less manual [22:26] The status is.... [22:26] good progress kenneth! [22:26] 1. the form is done and all the supporing form topics done. [22:26] what can we do further? [22:26] 2. All the proposals with a known driver are updated [22:27] The proposals without a driver are not yet updated. THe ones that are parked. I can do those tomorrow (I have taken a vacation day) [22:27] 3. Only remaining is to create a few topics with relevant searches. [22:27] hope you enjoy the weather [22:27] over time we can create some meaningful reports [22:28] can we list them somewhere? [22:28] I just panic created a quick search for the items ready for release meeting. [22:28] :-) [22:28] I mean create a list of lists we think will be useful? [22:28] at wind river i once created a tracker that did some spreadsheet time calc based on dates [22:29] embedded in a search [22:29] What I will so it to change the TWikiFeature04x02 topic so it contains a list of links to searches AND a submit form for submitting new proposals. [22:29] may we can do somthing to automate the dispaly of the 14 day rule [22:29] Peter: don't push yourself as volunteer [22:29] :-) [22:29] :-) [22:29] Yes. If you have a good search mechanism to highlight the 14-day rule topics that would be nice. [22:30] i will look into it, but can't promise [22:30] do we have a topic on the 14 day rule? [22:30] it started as 2-week rule [22:30] Yes - same thing. [22:30] and its a bit hard to find [22:31] TWikiRelease04x01Process [22:31] ok, then I will lift something from that [22:31] The link is on the TWikiFeature04x02 topic. [22:31] The rule is.... [22:31] The 14-day rule applies when... [22:32] 1. The proposal has a developers committed to drive the implementation. [22:32] 2. Noone has raised any concern against the feature or its spec. [22:32] Anyone can raise ANY concern. [22:32] The minute there is concern the clock stops. [22:33] The rule is there so that proposals that are nobrainers that noone will argue against or do not care about do not end up rotting. [22:33] AND is also creates an incentive for the community to react [22:34] actually, i would not be too religious about stopping the clock on the 14 day rule [22:34] yes that is useful [22:34] unfortunately i must going offline. i wish you a succesful meeting and a good night after finishing. cu [22:34] if there is a healty dialog and all agree it still can be accepted, right? [22:34] * McAldo has quit IRC ("using sirc version 2.211+KSIRC/1.3.11") [22:35] If there is concern then the discussion needs its time to finish. That is why the clock stops. [22:35] If people then agree - and this happens often - then there is a consensus decision. [22:35] If people end up not agreeing - then this meeting takes a vote. [22:35] and if the release meeting agrees, and someone still raise a concern? [22:36] I sometimes raise a principle concern when I feel a decision is so important that it should go to this meeting. [22:36] (within the 14 days) [22:36] The release meeting should not decide unless people have had a chance to evaluate the proposal. Unless it is a total nobrainer. [22:37] i'd day the release meeting is the place to make the decision [22:37] It is not a religion we are part of. Just some basic rules that enables the community to make decisions fast and effectively [22:37] this is an incentive to participate :-) [22:37] yes, common sense after rules [22:38] s/after/over/ [22:38] so far the process is working pretty good [22:38] There have been several cases where all says HURRAY in the Codev topic and then I chose not to put it forward here to save time. It has worked pretty well until now [22:39] lets move on [22:39] - Crawford/Sven - A few features sponsered by customers are urgent for a 4.2 release. Sven and Crawford will draft a realistic release schedule for a 4.2 for next release meeting. [22:39] is sven here in person? [22:39] SvenDowideit: ping [22:39] They added a very near term schedule. Feature freeze already in one week. [22:40] i think we can move to the next action item, since it is related [22:40] ---+ 2. Next Release 4.2 or 5.0 [22:40] it was my understanding that we did not decide on 4.2. or 5.0 [22:40] I am on vacation from the 29th till the 5th of May - in Portugal. So if I am to create release work then I will need another week. [22:41] that crawford/sven should bring a schedule with scope [22:41] for a potential 4.2 release [22:41] or sven should do some building work [22:41] (or a lot of it) [22:41] The actual building is not an issue at all. The build script does it all. [22:42] on last meeting i did not see a clear list of features that warrant a 4.2 release [22:42] 4.2-5.0 is a marketing issue, isnt it? [22:42] no, it is a scope issue as well [22:42] no, it is a strategic question [22:42] focus on long term big features or on short term incremental ones [22:42] ok, understood [22:43] do we go for the big scheme first, or do we solve midterm problems. It looks we are going for 4.2 full steam nonetheless [22:43] All the bug fixing Crawford has done makes the 4.2 more interesting than it was (at least for me) two weeks ago where 4.2 was mainly features my users will not need. [22:44] can we list the noteworthy 4.2 features? [22:44] my users will! [22:44] What I was thinking about was the integration of TWiki with OpenID and Joomla. [22:45] THAT I will not need. [22:45] * ktwilight_ has joined #twiki_release [22:45] But there are other enhancements. An important one is EditTable improvements. [22:45] yo [22:45] it won't hurt either [22:45] mmm, is anything interesting happening? [22:46] hi kwang! [22:46] totally forgot, my bad :) [22:46] i'm not actually here, but 2 pings drew my attention [22:46] hi peter [22:46] hmm, I got lost in all edittable versions [22:46] what are noteworthy features of 4.2? [22:47] There are TWO. There is the "official" that you was part of that allows moving and deleting rows. THAT is a great features. [22:47] bug fixes, stability, scalability and plugability [22:47] The plugin Crawford did is not compatible YET. [22:47] some things look small, but EasyUrlQueryHandling will make building of dynamic apps a lot easier [22:47] ok [22:48] Yes. There are enough features now to call it a 4.2.0 [22:48] and ProcessAddToHeadAdds will make adding plugins less expensive [22:48] i am trying to figure out what we could list as twiki 4.2 features [22:48] btw Lavr wrt DatabaseStore, no, you're wrong. those are all implementation details I won't be determining before i undertake the work [22:48] ok [22:48] oh, scalable & pluggable user system [22:48] a press release with two minor enhancements does not look great (lots of internal refactoring does not count for marketing folks) [22:49] Sven. Let is take that afterwards. I think we agree more than you think [22:49] atm twiki spends alot of time doing nothing [22:49] when there are lots of users [22:49] or lots of groups [22:49] as CDot mentioned last, we can spend a release on clearer docs [22:49] um, that was ME [22:50] but back to user stuff [22:50] if you read back the irc logs of the last month [22:50] you will see somewhere between 5-10 people asking q's [22:50] because they _are_ implementing their own user mappings and auths [22:50] and at the moment its very painful, hard and bug prone [22:51] and thats ignoring the number of users that ask how to stop registration [22:51] ok, scalable & pluggable user system is a noteworthy feature [22:51] which is also hard atm [22:51] what else? [22:51] its one of _THE_ enterprise things we havn't done well enought [22:52] anyone have the list of commits handy? [22:52] hm, i would like to work on the docs (said that long time ago too). but not at the moment, maybe next stable release? once i get my stuff out [22:52] I'm in in this mental space atm, its 11pm, and i'm about in bed [22:53] in fact, does no-one here know the features tha have been commited since 4.1 ? [22:53] Sven as I understand from CDots schedule - the code for 4.2 user related features will all be complete by end of this month? [22:53] that was our hope and intent [22:53] they are here: http://twiki.org/cgi-bin/view/Codev/TWikiFeature04x02 [22:53] i have two questions: [22:53] 1. is the existing feature set big enough for a 4.2 release in near future, or work on more feature/better docs to warrant a 4.2 release? [22:53] 2. are we going for a 4.2 or a 5.0 release? [22:54] * SteveRJones has quit IRC (Remote closed the connection) [22:54] I say we go for a near term 4.2 [22:54] i may have jus thad my stay here extended - i was going to be travelling for the first weeks of May, but things chage [22:54] lemme put it this way [22:54] ignoring my work [22:54] cdot has done rather alot of work, and has requested there be a release of his contributions soon [22:55] but we should also have a list of fixed bugs since 4.1 [22:55] personally, i'm in the same boat, missing FuncuserContrib was an issue [22:55] List of bugs and enhancements is auto generated from Bugs web. But someone should summarize it on the release note. [22:56] so, back to bed for me, this time i'll remember to turn off the speaker [22:56] When I was against a 4.2 last time it was because I felt it took focus away from the long term things like database store and Wysiwyg [22:56] But with a quick release of 4.2 then I support it 100% [22:56] my need to feed myself takes away time from those too [22:57] i am for a 4.2 release, but am wondering if we have enough features that warrant it to be called 4.2 [22:57] oh, did anyone that wants easier twik installs try the 173 debian packages i released yesterday? [22:58] It is too large to be called 4.1.3 because upgrading will require more than just copy over. [22:58] i have not looked into it sven, but anything that makes it easier to get started with twiki is good [22:58] thank you! [22:58] * SteveRJones has joined #twiki_release [22:59] Sven. I have no Debian so I cannot provide feedback but it is great what you are doing. [22:59] :) install debian, run =apt-get install twiki twiki-tagmeplugin twiki-moveabletypeskin [22:59] Lavr, yes, you can [22:59] please put on the minutes [22:59] Lavr, to use Alien to convert the debian package to rpm [22:59] and to provide info on what happen [22:59] You build a deb for every plugin? :) [22:59] only 172 of them [22:59] i'm lazy [23:00] wow [23:00] see http://blog.distributedinformation.com for updates [23:00] if you use it on your production system, please docco the result so i can have a giggle [23:01] it seems we have consensus to go for 4.2, not 5.0 [23:01] question is timing [23:01] To follow up on the 4.2 and 5.0. I would like people to consider my two proposals for RELEASE THEMES for 5.0. [23:01] and scope [23:02] i feel we should have at least 5 noteworthy features for 4.2 [23:02] And with a release theme I mean a larger feature family that the community agrees will be the main focus for a major release. [23:02] you better get coding then [23:02] yeah, i'd love to know what the community is going to commit to doing, pushing and testing for 5.0 too [23:03] I had TWO proposals. ONE is an alternative storage model that allows scalability. THAT is your DB storage Sven among others. [23:03] And TWO is Wysiwyg. [23:03] ooo, list 2 others [23:04] you can't tease me with other stores without telling us something! [23:04] db storage is an 5.0 thingy [23:05] what's the timeframe of 5.0? [23:05] better wysiwyg is an extension that can be rolled out at any time [23:05] really good wysiwyg requries good strong TOM [23:05] which we don't have [23:05] TOM? [23:06] and DBStore just happenes to also need TOM [23:06] TopicObjectModel [23:06] :/ [23:06] tom & jerry ;-) [23:06] Ah. [23:06] Can I request something? [23:06] a not good name, for well addressable topic sections [23:06] with get&set [23:06] ah [23:06] to cell's set's paragraphs and more [23:06] steve, ask. don't wait [23:06] SteveRJones, you can ask _anything_ [23:07] BTW it is in http://twiki.org/cgi-bin/view/Codev/TopicObjectModel [23:07] We have slow adoption of TWiki due to lack of a robust security model [23:07] ooo, yep, we'd love a well defined ACL proposal [23:07] with tests and validation etc [23:08] I doubt it is your users that asks for less freedom and less empowerment. [23:08] steve, i'd like to hear more details on this [23:08] It forces us to put protected data into a separate repos. [23:08] yup [23:08] probably best to document your needs in codev [23:08] back to agenda [23:08] I can do that [23:09] on release timing [23:09] is this doable/desirable? [23:09] # 4.2 feature freeze - 1 May 2007 01:00GMT [23:09] # Pre-release testing until 6 may 2007 [23:09] # Build and release beta package, 6 May 2007 [23:09] # beta test until release team is satisfied with quality (not less than 2 weeks) [23:09] cc obviously thinks so [23:09] that is what crawford proposed [23:09] I'm happy to dance to his tune until we know more / less [23:10] I would like to see a week added. Freeze date is OK with me, but I am in Portugal the first week of May. [23:10] I'm currently porting the joomla work to it, for testing [23:11] but there is still changes todo to the core user code for it, twikiadminuser and the other related stuffs [23:11] atm i think my added docco is outweighing the code change, but >shrug< [23:11] my concern is that we do not seem to have enough noteworthy features, but i might be wrong [23:11] But all in all we are aiming for a release end of May. [23:11] peterthoeny_, 2 options [23:12] go look, and decide [23:12] or trust that other people think its so [23:12] i trust cc [23:12] Lavr, yup [23:12] Peter. When we list the features added I think it will look pretty OK. Especially the features related to better plugin support. [23:12] In API. [23:12] my concern is from a marketing perspective, not a dev perspective [23:12] yes, [23:12] a release is a lot of work on marketing side [23:13] last week, did n't you agree that we could market a massive docco improvement release? [23:13] we will have zero impact if we can't show something flashy [23:13] make that happen, and you will have more to market [23:13] it's not always features sometimes [23:13] WHAT I would like to see in 4.2 is Crawfords new proposal for a proper META search feature. The reason.... [23:14] but if it's end of May, i might be able to squeeze sometime to improve on the overall look of the docs? especially README or INSTALL [23:14] y, Lavr i concur [23:14] if we do not have enough feature in core we potentially could describe some cool plugin or apps in the press release [23:14] that was a fantastic discovery of his [23:14] the .html, 'cuz it looks intimidating [23:14] If rel 5. is with a good SB storage it will be a HUGE advantage if this search syntax has been fielded for a year before. [23:14] DB storage not SB [23:14] ya, concur [23:14] peterthoeny_, true, but reiterating existing features and boasting about its uses can also be good. [23:15] its the begining of teaching users TOM, so that they can tell us what they actually _use_ [23:15] spaced out wiki words is almost a reality [23:16] i feel we should give a few more weeks before feature freeze [23:16] so that we have some more noteworthy features [23:16] ok, tell me when, and i'll start work later [23:16] the meta search is a good one [23:16] i've got enough stuff on my plate atm [23:16] which can also be applied to search [23:16] If we can reach consensus quickly on the meta search proposal and CDot will implements it then I would also propose delaying 4.2 by the 1-2 weeks it takes. [23:17] >:} [23:17] on meta search enh, i feel enh the %SEARCH would be even more benefitial [23:18] where's the doc on meta search? [23:18] put on topic [23:18] i will [23:18] get contributor feedback [23:18] oh nm [23:18] found. [23:18] If we can drive http://twiki.org/cgi-bin/view/Codev/SimpleFieldQueriesInMETASEARCH to consensus within a few days then we may be able to get it into 4.2 [23:18] time check: +75 min [23:19] i need to sign off at +120 min [23:19] THAT will be a killer feature itself. Both long and short term. [23:19] i would dig metasearch [23:20] i think a %SEARCH that groks a query language to search meta data is more valuable [23:20] i agree that having more than one search tag is suboptimal [23:20] right now it is very complex and implementation dependent to do the regex search on form fields [23:20] and always wanted to do away with the second [23:21] but its going to depend on bugbear number 2 [23:21] backwards compatibility [23:21] regex in meta is above 5 on the Nerdometer. The new search will bring it down to 3-4 where it belongs. [23:21] some of our twiki apps are really dependant on how SEARCH does its thing [23:21] lol [23:21] good point kenneth [23:21] regext search is nerdometer level 8 [23:22] i thought it's 10. [23:22] :P [23:22] 10 is doing the same in haskell [23:22] na, there are at least 3 or more developers that can write code in it [23:22] i'd like to have a few weeks for these new features before we feature freeze [23:22] see what cc has up [23:23] not enough time for template unification I am afraid [23:23] he and I are more of the release when ready part [23:24] lets do a quick poll: [23:24]