(9:06:55) CarlosRuiz ha cambiado el tema a: 1 - Centralized Application Dictionary Management / 2 - Commit committee can't cope with the review of all commits
(9:07:45) red1 [email@example.com] ha entrado en la sala.
(9:12:39) Michael_Judd: yes - ok - so I can build that in then .... or you can as long as we co-ordinate with the depracted methods ..
(9:13:10) Michael_Judd: hi carlos..... thanks for your support this week ....
(9:14:05) Michael_Judd: I could use some help with ..... http://www.adempiere.com/wiki/index.php/Commit_Best_Practice
(9:14:15) Michael_Judd: if any one is willing ...
(9:15:30) red1: its ok.. and if we add Trifon's point of selecting a group of good committers only to have direct trunk commit
(9:15:45) red1: i think what Carlos meant by committer layer is that
(9:16:05) richard-4layers: and how do the others work?
(9:16:54) CarlosRuiz: Hi Michael, you're welcome
(9:16:54) CarlosRuiz: Hi everybody, we're going to start in 44 minutes, ok?
(9:17:04) teo_sarca: i think strict rules will be better then 5 commiters that should review the code
(9:17:15) Michael_Judd ha cambiado el tema a: 1 - Centralized Application Dictionary Management / 2 - Commit committee can't cope with the review of all commits 3 / Adempiere Foundation
(9:17:56) Michael_Judd: tend to agree teso - keep it open - then we just need to police the rules ......
(9:18:08) Michael_Judd: that's easier to do ....
(9:18:22) Michael_Judd: but depends on setting good rules ....
(9:18:49) Michael_Judd: Carlos - do I have the wrong time? Or do you need to defer the meeting ?
(9:19:48) CarlosRuiz: yes, you have the wrong time :-)
(9:19:56) mark_o_: nope. the meeting shall start at gmt 15:00
(9:20:02) teo_sarca: michael, are you using Windows VIsta ?
(9:20:02) CarlosRuiz: GMT doesn't have DST
(9:20:05) mark_o_: and now is gmt 14:20
(9:20:09) mark_o_: :-)
(9:20:19) Michael_Judd: teo - I haven't been on windows for at least 6 years .... ;)
(9:20:31) teo_sarca: oh oh :))
(9:20:37) mark_o_: :-)
(9:21:00) vpj-cd [n=Horus@22.214.171.124] ha entrado en la sala.
(9:21:05) karsten-thiemann: hi victor
(9:21:13) vpj-cd: Hi everybody
(9:21:15) red1: i go with Trifon's suggestion.. let ppl be committers until we know who is who
(9:21:17) teo_sarca: hi
(9:21:20) richard-4layers: sorry, but i don't understand how a lot of people can work rapidly enaugh if only fews can commit...
(9:21:24) vpj-cd: Hi karsten
(9:21:28) richard-4layers: Hi victor....
(9:22:15) vpj-cd: the first rules we agreement is no break the trunk
(9:22:18) CarlosRuiz: waiting for the start of the meeting in 38 minutes ... meanwhile does someone know how Linux work ? AFAIK Linux have very few committers
(9:22:22) vpj-cd: I think is good rule
(9:23:17) karsten-thiemann: I forgot our summer time :)
(9:23:54) red1: or maybe linux has more developers but they dont commit
(9:24:03) red1: committers can be finalists
(9:24:04) richard-4layers: Wouldn't it be better to have different level of commitment... one "dev" for everybody, and one "checked" for administrators...
(9:24:09) karsten-thiemann: I thought we would be still gmt+1 but in fact we are +2..
(9:24:09) CarlosRuiz: that's the point red1
(9:24:38) red1: perhaps we can arrange the trunk to be (1) development (2) Production
(9:24:50) red1: so there is 2 step before finalist
(9:26:24) Michael_Judd: isnt this issue here that there needs to be some QA / testing ..... ?? Perhaps that is a service - how we can differentiate our version ... ?
(9:26:37) vpj-cd: red1 I think the trunk should a rock
(9:26:41) Michael_Judd: QA costs time and money - it's a natural service ...
(9:27:04) karsten-thiemann: so it would be more like having a general branch with lots of committers (like the trunk is now) and then few committers who merge this branch with the trunk
(9:27:17) vpj-cd: to somebody want commit in trunk need your development are tested
(9:27:19) dladwig_Idalica [firstname.lastname@example.org] ha entrado en la sala.
(9:28:02) CarlosRuiz: still 23 minutes for the meeting ...
(9:28:02) red1: i agree with karsten
(9:28:15) red1: i agree in 23 mins
(9:28:27) richard-4layers: that's excactly what I'm saying....
(9:29:25) red1: vpj-cd, what do u mean as a rock? u mean rock solid? unbreakable?
(9:29:32) richard-4layers: a general level where the most are enabled for branches and contributions and an administration level for trunk and tags....
(9:29:48) vpj-cd: yes Red1 rock solid
(9:30:04) red1: will there be an issue of too many branches?
(9:30:13) vpj-cd: I think the development can have a branch to development improves or modules
(9:30:33) red1: then it back to the finalists overworking to verify
(9:30:43) vpj-cd: also we agreement that all new development should be integrate via 2pack
(9:30:53) red1: we shuld look out and invite more good committers
(9:30:53) Michael_Judd: who decides what gets promoted from testing to final ?
(9:31:00) CarlosRuiz: 9:30 here - 30 minutes to go (excuse me) ... it's better to wait if others join
(9:31:09) red1: i think there are more from time to time
(9:32:14) vpj-cd: CarlosRuiz no problem is is only a introduction :-)
(9:32:26) karsten-thiemann: we are just thinking loud
(9:32:28) CarlosRuiz: it's better to wait for others
(9:32:33) CarlosRuiz: the discussion will be richer
(9:32:40) red1: Michael_Judd, i think whats more important is WHO not what... we need more hi quality committers
(9:32:58) red1: this is water cooler talk :P
(9:33:49) vpj-cd: red1 the main issue is adempiere do not convert in frankenstein
(9:34:05) red1: one thing for sure Mike, u and me still dont make it to the top cut
(9:34:18) red1: the more we commit the more we break :(
(9:34:34) red1: this is the world league, amigo
(9:35:56) red1: vpj-cd, what u mean by frankenstein?
(9:37:10) Michael_Judd: never thought I was a developer .....
(9:38:17) vpj-cd: if many people add new functionality but this fuctionality exiset
(9:38:31) Michael_Judd: the trouble is that the people who commit dictate the product development - so there needs to be a feed back mechanism from customer to committer - or else we loose the open source advantge ...
(9:39:28) vpj-cd: then adempiere can increase with discontrol
(9:39:57) red1: we can try to review what worked and what didnt work too well in the past few mths
(9:40:59) vpj-cd: I think different in adempiere is a application critical mission
(9:41:01) CarlosRuiz: open source advantage is not that everybody commit
(9:41:14) Michael_Judd: just found http://www.adempiere.com/wiki/index.php/SVN_Commit_Policy - really useful ....
(9:41:44) red1: thats the example.. Mike 'committed' a new policy when there is already one before
(9:41:57) Michael_Judd: It feels like suddenly Adempiere is becomming Compiere ....
(9:42:07) red1: like Carlos said, in wiki, its easy to operate but in trunk its not.. things break
(9:42:31) red1: Compiere doesnt hold developers' meeting
(9:42:32) CarlosRuiz: I have heard that before Michael :-) exactly every time you revert a commit ;-)
(9:43:00) CarlosRuiz: revert a commit to protect product is not becoming Compiere
(9:43:18) vpj-cd: yes I remember when Carlos rever some of my changes
(9:43:48) Michael_Judd: Well what I really needed was an easy to follow guide of how to commit - that would have solved all problems as I don't mind following rules but when there are no rules to follow it makes the process of commit very expensive for everyone ...
(9:43:55) red1: we re trying to formulate an open policy to ensure commits are done according to good measures
(9:44:19) CarlosRuiz: it's not so easy
(9:44:27) red1: just like in Berlin, during security discussion, Teo mentioned that its about taking measures... not questioning the openness
(9:44:29) CarlosRuiz: you can follow all rules, and still commit low quality cod
(9:44:39) Michael_Judd: If I refresh the svn head in a fresh project, then overlay my changes, check that it compiles and works and then commits then thats fine....
(9:44:44) CarlosRuiz: you can follow all rules and broke adempiere
(9:44:48) Michael_Judd: Carlos - I agree .... just look at my code ...
(9:45:03) CarlosRuiz: you can broke business processes too, without broking code
(9:45:11) red1: mike, i hear the code u did last nite didnt work all the way.. was that so?
(9:45:26) Michael_Judd: I know too well that I need to do some more work on the code
(9:45:40) vpj-cd: ut I am totaly agree in trunk is unbreak
(9:45:40) Michael_Judd: Actually red1 - the code you gave me was broken - so I had to make a fix
(9:45:47) red1: so perhaps that can be done with one further step before
(9:45:59) Michael_Judd: that fix was partial - but it was better than the broken code in the svn ...
(9:46:14) Michael_Judd: then teo gave me the solution I was looking for and I committed that over the top ...
(9:46:28) red1: well thats the evolution of code fixing
(9:46:39) ***CarlosRuiz waiting another 14 minutes to start ...
(9:46:47) Michael_Judd: yes - the point that I hear everyone saying is don't brak the trunk
(9:47:13) Michael_Judd: and that's a great rule - if we can tell people how to check that properly then that solves problem #1
(9:47:29) red1: in project implementation we can have a development sever, separate from the live server
(9:47:35) Michael_Judd: I'm really looking forwardto the satrt of THIS meeting ;)
(9:47:53) red1: but the point is that mike, 'our' code went in without any QA
(9:48:09) Michael_Judd: well - perhaps the trunk shouldn't be the QA ....
(9:48:10) red1: u trusted me, thank you for that
(9:48:17) Michael_Judd: perhaps QA should be a separate process ....
(9:48:31) Michael_Judd: perhaps releases should be QA'd ...
(9:48:36) Michael_Judd: and then we should have testing ....
(9:48:38) red1: precisely
(9:48:58) red1: trunk = nitely build
(9:48:58) Michael_Judd: so trunk, testing, release .... I'm not the developer but I understand that this is common in the industry ;)
(9:49:23) red1: to stablise, extract from trunk, stabilise, release
(9:49:57) red1: but who to do that?
(9:50:04) richard-4layers: tags must be QA'd... trunk shoul be.
(9:50:14) Michael_Judd: that is the question ;)
(9:50:17) vpj-cd: in adempiere we need have the vision that function consultant say that is necessary development
(9:50:19) CarlosRuiz: industry = Linux? Linux have very few committer that receive contributions, not commits
(9:50:57) karsten-thiemann: before you commit you should do qa by yourself
(9:50:59) Michael_Judd: Linux - you mean the kernel - and the kernel is very stable ....
(9:51:02) vpj-cd: fist we have the problem and try solve from focus functional , if this no is solve then we can think in solve via development
(9:51:02) red1: i think Eclipse is a better reference, cos it has multiple projects
(9:51:34) richard-4layers: How do commiter in Linux receive contributions?
(9:51:37) red1: QA rule says that someone else must do the QA, not yourself :D
(9:51:59) karsten-thiemann: before you commit you should be quite sure that your code doesn't break the rest - so it should compile and you should have tested it's functionallity..
(9:52:10) mark_o_: @ red1: thats what the lazy developers say!
(9:52:11) red1: JBoss is another example, where i read that the committer are few
(9:52:15) Michael_Judd: Richard - in Linx - a team of 600 programmers paid by IBM get the code via a couple of branch moderators ...
(9:52:20) mark_o_: because testing is not so funny!
(9:52:26) karsten-thiemann: red1 - yes but you need to do some basic tests
(9:52:31) mark_o_: to find your own failures
(9:52:37) hengsin [email@example.com] ha entrado en la sala.
(9:52:41) vpj-cd: I my expertise exist excellent development, but they do not always know business process
(9:52:56) red1: victor, how do u do QA for Libero?
(9:52:59) vpj-cd: then this peril want solve all with code
(9:53:01) vpj-cd: :-)
(9:53:02) Michael_Judd: I don't think people disagree with testing - but how do you do testing - that is what the CC should define ....
(9:53:21) hengsin: two type of testing
(9:53:26) vpj-cd: QA is when I go with a client in production Red1
(9:53:29) hengsin: 1. unit testing 2. functional test
(9:53:44) Michael_Judd: so these should be defined for every commit ?
(9:53:46) vpj-cd: in control environment
(9:53:53) hengsin: unit testing should properly be done using automated junit test, functional testing should be scenario driven
(9:54:04) hengsin: just my 2 cent
(9:54:04) richard-4layers: Michael_Judd: What I wonder is the way they exchange sources without committing.... maybe they have a secondary repository?
(9:54:09) vpj-cd: but the issue no is bad code
(9:54:32) fredtsang [n=chatzill@ADSL-TPLUS-6-42.intnet.mu] ha entrado en la sala.
(9:54:40) red1: junit to remove laziness
(9:54:41) vpj-cd: the issue is bad disaing
(9:55:28) vpj-cd: you can write a excellent code but this code is best practice
(9:55:44) red1: u mean bad design
(9:56:00) vpj-cd: in mexico exist many many development that try say are ERP
(9:56:27) vpj-cd: but they reality try reinvent the wheel :-)
(9:56:42) richard-4layers: Is the meeting officially begun ?
(9:56:51) usm88 [firstname.lastname@example.org] ha entrado en la sala.
(9:56:51) Michael_Judd: NO 4 mins !!!!!!!!!!!!!!
(9:56:54) Michael_Judd: ;)
(9:57:00) red1: i seen your demo and it shows elegant design, vpj-cd .. but how do u train others to look at it too? Now there is Tim following your codes and Rob Klein wrote to us saying he is concerned
(9:57:03) richard-4layers: Oh k'mon
(9:57:14) vpj-cd: yes design i sorry
(9:57:37) red1: haha.. been trying to figure out what code u saying :D
(9:57:37) richard-4layers: what demo?
(9:57:41) vpj-cd: yes Red1 but I do not reivent the wheel
(9:57:51) red1: in Berlin, victor demo the Libero
(9:57:53) vpj-cd: I use the standard of APICS
(9:58:01) Bahman [n=Bahman@vps89.zoner.fi] ha entrado en la sala.
(9:58:16) red1: so is 2pack, the maker Rob said about his design issue
(9:58:17) Bahman: Hi all!
(9:58:21) vpj-cd: is my based I do no reinvent the wheel Ñ-)'
(9:58:23) vpj-cd: Ñ-)
(9:58:28) mark_o_: hi bahman
(9:58:30) Bahman: Looks like I'm not late!
(9:58:35) vpj-cd: hi bahman
(9:58:39) mark_o_: nope. 2 mins
(9:58:39) Michael_Judd: Hi Bahman ...
(9:58:39) red1: thus we can agree design is important, but from coder to coder exist difference
(9:58:56) red1: and we need to communicate up and down
(9:59:10) red1: in WEB 2.0 we have to communicate!
(9:59:31) vpj-cd: yes this this reason that my idea is use some similar to sun using with java
(10:00:00) vpj-cd: many company can give you idea and create new technology
(10:00:04) mark_o_: it's now 15:00 GMT!
(10:00:15) CarlosRuiz: flat
(10:00:16) muthah_ [email@example.com] ha entrado en la sala.
(10:00:16) CarlosRuiz: :-)
(10:00:20) richard-4layers: what excactly about sun rules?
(10:00:26) vpj-cd: Carlos now we can start :-)
(10:00:58) CarlosRuiz: looks like there is more interest in 2 than in 1 first
(10:00:59) vpj-cd: richard they use
(10:01:06) CarlosRuiz: so, we could start discussing the second point
(10:01:12) vpj-cd: Java Community process
(10:01:15) vpj-cd: http://jcp.org/en/home/index
(10:01:17) CarlosRuiz: to not interrupt the previous loud thinking :-)
(10:01:32) richard-4layers: Ok, I know it...
(10:01:33) karsten-thiemann: :P
(10:01:44) CarlosRuiz: currently we have this schema
(10:01:45) vpj-cd: this maintenance the control of java
(10:01:54) CarlosRuiz: many people commit, CC review
(10:01:57) richard-4layers: But Sun is even too much organized
(10:02:14) CarlosRuiz: when I say CC I think just Hengsin, Teo and me are revieweing (normally I don't see Victor and Trifon revieweing)
(10:02:28) red1: but trifon did, toh seldom
(10:02:34) CarlosRuiz: so we have a lot of committers and 3 persons reviewing
(10:02:53) CarlosRuiz: this schema can work while you have few committers, with many people committing this schema will fail
(10:02:57) vpj-cd: yes I review the main commit Carlos
(10:03:00) CarlosRuiz: and I think is failing now
(10:03:11) karsten-thiemann: and it is not garanteed that a commit is reviewed
(10:03:13) CarlosRuiz: ok, I've never seen Victor or Trifon reverting ;-)
(10:03:13) vpj-cd: I help to Michel to fixed a patch that break the trunk
(10:03:15) vpj-cd: :-)
(10:03:21) Bahman: Gotta mention that there's not lots of committers...about 12~15 at most.
(10:03:29) CarlosRuiz: yes, but growing
(10:03:38) CarlosRuiz: and 12-15 are now beyond the capacity of 3 reviewers
(10:03:45) richard-4layers: Here is a frsh new one :-)
(10:03:47) CarlosRuiz: and as Karsten said, there is no guarantee
(10:04:08) vpj-cd: now I think is very simple
(10:04:36) karsten-thiemann: but there is a functional review - cause lots of persons are testing/using the actual trunk
(10:04:50) CarlosRuiz: not sure, but now that Hengsin is working on AJAX, he's not reviewing thoroughlly
(10:04:50) karsten-thiemann: so problems are very fast reported
(10:04:52) vpj-cd: when the commit is the Carlos , Teo , or Trifon I have many confidence, but I know you work
(10:04:56) mark_o_: that's why the trunk must be stable
(10:04:57) red1: and also report back in bug trackers.
(10:05:01) Bahman: What if we maintain trunk as the stable one, to which only CC can commit and have another branch, say devel, for other committer to commit into...CC later can review and merge the good ones into trunk.
(10:05:29) red1: then we must point ppl to that branch to test it for us
(10:05:30) karsten-thiemann: that is even more work for CC
(10:05:34) needle58 [firstname.lastname@example.org] ha entrado en la sala.
(10:05:38) Michael_Judd: kasten - its too late for user to test functional .... they need to be hadnled the same way as compile bugs ...
(10:05:54) red1: so its a matter of organising those, and let community know where is where
(10:06:06) vpj-cd: yes I think gold rule is trunk should be unbreak
(10:06:16) richard-4layers: trunk and tags
(10:06:22) teo_sarca: agree with victor
(10:06:28) Michael_Judd: agre also with victor ...
(10:06:32) mark_o_: me too
(10:06:38) richard-4layers: yep
(10:06:41) teo_sarca: richard, basically we don't use tags/branches :)
(10:06:42) vpj-cd: to integrate something into trunk should are test to 2 tester I think
(10:06:44) Bahman: Perhaps enforcing test units as a part of commits can help.
(10:07:12) richard-4layers: teo, what u mean "we don't use branch" ?
(10:07:20) CarlosRuiz: and there are types of contributions:
(10:07:20) CarlosRuiz: - bug fixing
(10:07:20) CarlosRuiz: - small new functionalities or enhancements
(10:07:20) CarlosRuiz: - medium/big new enhancements
(10:07:23) richard-4layers: what about libero?
(10:07:24) vpj-cd: the developer can create a branch to create new functionality
(10:07:33) CarlosRuiz: the current rules say: bug fixing doesn't need permission to commit
(10:07:40) karsten-thiemann: hmm - but until now the trunk _is_ very stable, isn't it?
(10:07:40) teo_sarca: richard, we tag the trunk, branch the trunk but never bug fix the tags/branches ....
(10:07:45) vpj-cd: libero is new module and extend adempiere
(10:07:48) CarlosRuiz: medium/big need forum post, votation, etc
(10:07:59) red1: libero is and addon
(10:08:06) CarlosRuiz: ok
(10:08:16) CarlosRuiz: my idea is to have a layer of committers
(10:08:18) richard-4layers: yes, but it's directly done in branch...
(10:08:21) Michael_Judd: I think that is everyones understanding Carlos
(10:08:27) CarlosRuiz: few committers in trunk
(10:08:35) CarlosRuiz: but when I say few, it's not CC
(10:08:39) richard-4layers: as everything should be done; before branch, then reviewd and goes to trunk....
(10:08:41) CarlosRuiz: we can have maybe 8-10 committers
(10:08:57) vpj-cd: I think in trunk we need only set fixed bugs, important enhancements but need will be test
(10:08:58) CarlosRuiz: there are many good developers currently not in CC but we can change this to become them committters
(10:09:03) red1: ok agree
(10:09:16) hengsin_ [email@example.com] ha entrado en la sala.
(10:09:28) Bahman: And patches and bug fixes are reviewd and committed by them CarlosRuiz?
(10:09:29) vpj-cd: and medium / big new enhancements or new module as libero we need use 2pack
(10:09:54) Michael_Judd: does 2pack work on postgresql yet ? Last time I tried it was broken ...
(10:09:57) CarlosRuiz: Bahman, you must be in :-) not by "them"
(10:10:01) karsten-thiemann: maybe it would be easier to have a pair-committer rule
(10:10:13) red1: bahman, carlos already mentioned the rule of bug fixes need not review
(10:10:28) vpj-cd: yes Michael Tim are work and we are get run libero with 2pack
(10:10:36) Bahman: But some of them will easily break the code.
(10:10:36) vpj-cd: in postgresql also
(10:10:38) karsten-thiemann: so that everybody who wants to commit something needs to get the ok from another commiter - but we still have lots of committers
(10:10:41) Bahman: Or functionality.
(10:11:03) red1: karsten has point but how about thinking in teams of groups?
(10:11:04) Michael_Judd: ok victor - will start working with 2pack again ... ;)
(10:11:17) karsten-thiemann: and if you break something your commit rights can be taken away
(10:11:23) red1: groupings by code areas they focus on
(10:11:40) richard-4layers: I agree with functional groups
(10:11:41) vpj-cd: yes michael Tim create a new importer class
(10:11:41) Bahman: That's better karsten! Executional guarantee :-)
(10:11:46) red1: like the lead maintainer idea that we have anyway
(10:11:54) vpj-cd: include component nested
(10:12:09) hengsin ha salido de la sala (quit: Connection timed out).
(10:12:16) hengsin_ ahora se llama hengsin
(10:12:17) vpj-cd: Libero is application complex because use all AD and resource of Adempiere
(10:12:22) richard-4layers: still in each group there must be more then one CC
(10:12:48) karsten-thiemann: and it will still be a bottleneck
(10:13:06) CarlosRuiz: but a good bottleneck
(10:13:15) karsten-thiemann: I liked the 3.2 process
(10:13:29) Bahman: Does it slow the growth down? Like what happened to Compiere?
(10:13:30) karsten-thiemann: to have a feature freeze and bug fixing period
(10:13:33) CarlosRuiz: the point is not to make the process too quick, but make it too good
(10:13:35) teo_sarca: karsten, any measure generates a bottleneck, including breaking the core....
(10:13:44) richard-4layers: how many CC are there now?
(10:14:00) teo_sarca: 5 ?
(10:14:04) CarlosRuiz: not count in current CC, possibly we need to change this
(10:14:10) richard-4layers: and how many functional areas ?
(10:14:20) CarlosRuiz: better count : how many good developers we have that know how to touch the core?
(10:14:32) vpj-cd: yes is the issue carlos
(10:14:41) red1: the committer layer will replace CC, CarlosRuiz ?
(10:14:47) AS [n=AS@126.96.36.199.klj04-home.tm.net.my] ha entrado en la sala.
(10:14:47) vpj-cd: to touch the core need know functionality
(10:14:56) karsten-thiemann: right
(10:15:01) CarlosRuiz: yes, it could be, we can call it whatever we want
(10:15:03) richard-4layers: and architecture a lot
(10:15:29) richard-4layers: it's very important not to break architecyure too, not only code
(10:15:30) red1: we cannot escape the cycle of communication between functional, tester and committer
(10:15:34) Bahman: The code is the architecture richard-4layers :-)
(10:15:46) richard-4layers: C'mon... it's not
(10:15:51) CarlosRuiz: richard have a point
(10:16:01) richard-4layers: I'd love if it was
(10:16:11) CarlosRuiz: I have seen many contributions fixing things that are not broken
(10:16:20) Michael_Judd: there is too much focus on the code ..... IMHO
(10:16:20) CarlosRuiz: or implementing functionality that already exists
(10:16:26) vpj-cd: yes richard we need evolve but maintenance the own past
(10:16:30) Michael_Judd: Carlos is right on ....
(10:16:40) Bahman: They just don't know the functionality well CarlosRuiz, as Victor said.
(10:16:55) red1: so must the committer layer has functional expert?
(10:16:55) CarlosRuiz: exactly, is not all about code
(10:17:26) Michael_Judd: Adempiere needs more Carlos' and more knowledgable functional experts ....
(10:17:30) CarlosRuiz: I think is not necessary
(10:17:32) vpj-cd: yes issue with this approach is a developer can try solve a issue that do not exist :-)
(10:17:33) Michael_Judd: that can talk to coders ....
(10:17:33) karsten-thiemann: so how many functional experts do we have?
(10:17:34) CarlosRuiz: we need more communication
(10:17:49) karsten-thiemann: and how many of them know/understand the code?
(10:17:49) Bahman: Agree with CarlosRuiz.
(10:17:50) hengsin: when we develop, we should spend more time thinking , less time code, so functional and architecture knowledge is important
(10:18:04) red1: we have a good bug tracking system... we can see good communication going on there (at least)
(10:18:18) vpj-cd: I think that the community need more education with information
(10:18:23) CarlosRuiz: and more time communicating with others, IRC - forums - trackers
(10:18:25) richard-4layers: I think taht everyone should make his job...
(10:18:32) Bahman: That's what I wanted to talk about Victor...
(10:18:34) red1: perhaps we instill more discipline.. like the Commit Process Mike wrote in the wiki also said so
(10:18:39) vpj-cd: we need create documentation about architecture of adempiere
(10:18:41) Bahman: We need to breed some new developers.
(10:18:56) CarlosRuiz: every time I have one idea - I think is good, and when put in forums, there are always points from others that make it better
(10:18:56) richard-4layers: A technicien is not a functional and if he tries to be... well, it doesn't always work
(10:19:13) richard-4layers: I can write some Victor...
(10:19:29) karsten-thiemann: I added this to commit best practice: Before committing code there has to be
(10:19:30) karsten-thiemann: * a (confirmed) bug report - for bug fix commit
(10:19:30) karsten-thiemann: * or a (discussed) feature request with at least 3 votes for the new feature
(10:19:33) vpj-cd: yes java have a increase form order
(10:19:42) karsten-thiemann: I think that is about communicating
(10:19:44) red1: also notice that our adempiere-team meeting today is abit different from what it used to be
(10:19:49) Michael_Judd: red1 - the contributor agreements from the Foundation require us to have some process .... ;)
(10:19:59) Bahman: Very good richard-4layers! That's what we need to have a new wave of know-how developers.
(10:20:09) vpj-cd: ie some want create a new module to cash flow
(10:20:34) vpj-cd: the the people that know this subject have a discussion about this
(10:20:36) red1: what about Trifon's idea? Allow more to commit, but allow them to accept that if they break, they step back
(10:20:50) CarlosRuiz: returning to the question: how many developers we have that know how to touch core?
(10:21:05) karsten-thiemann: 1-2 :)
(10:21:11) Michael_Judd: but it is increasing ;)
(10:21:12) vpj-cd: they set general lines to development have a base to begin you development
(10:21:13) teo_sarca: red1, Trifon's idea sounds good but i don't know how to apply it :)
(10:21:17) richard-4layers: pessimistic!
(10:21:18) CarlosRuiz: red1, I think opposite, give people commit permissions when they demonstrate that can do it well
(10:21:27) CarlosRuiz: Karsten, you're humble
(10:21:35) red1: 1) Carlos, (2) Hengsin, (3) Karsten (4) Johannes ...
(10:21:42) richard-4layers: We can use SourceForge reputation for it....
(10:21:50) CarlosRuiz: Bahman, Karsten, Victor, Trifon, Teo ...
(10:22:05) karsten-thiemann: I work with the code since two years - but I still don't know all areas
(10:22:16) CarlosRuiz: neither me
(10:22:20) Bahman: Same as Karsten here.
(10:22:23) karsten-thiemann: I guess thats the same too most of us
(10:22:31) red1: but u know that already
(10:22:35) CarlosRuiz: yes, we don't have here Jorg Janke :-)
(10:22:38) red1: of not touching what u do not know
(10:22:48) red1: or at least u know how to go about it asking from ppl
(10:22:51) CarlosRuiz: even better
(10:22:53) karsten-thiemann: I hope so red1
(10:22:55) moyses [firstname.lastname@example.org] ha entrado en la sala.
(10:23:00) CarlosRuiz: we have developers that know how to research what they don't know
(10:23:05) Michael_Judd: well after making a few mistake - I now know a lot more - it will be the same when anyone new comes in ....
(10:23:06) usm88: JJ always claim that he is the only one who know the mega design :)
(10:23:08) red1: ah! i forgot (6) JJ
(10:23:38) CarlosRuiz: yes, we have a distributed brain, even better than JJ ;-)
(10:23:42) karsten-thiemann: hmm - but I don't think that he really does..
(10:23:47) richard-4layers: JJ has a design??? :-)
(10:23:57) CarlosRuiz: hahahahahaha
(10:24:05) red1: we are standing on his design
(10:24:08) red1: or what used to be
(10:24:21) moyses: JJ is a new CC member?
(10:24:23) richard-4layers: JJ is Jorge Janke, right?
(10:24:23) moyses: ;)
(10:24:28) CarlosRuiz: yes, and we're trying to preserve some good things
(10:24:32) CarlosRuiz: and replace not so good
(10:24:53) red1: yes Richard, he is
(10:24:59) CarlosRuiz: or John Jairo
(10:25:00) red1: been inviting him to be in this
(10:25:12) Bahman: CarlosRuiz, what you wanted to conclude from that question?
(10:25:16) richard-4layers: did you?
(10:25:33) CarlosRuiz: I think currently we can have maybe 10 skilled developers
(10:25:35) red1: we did... right in our main .org page
(10:26:07) CarlosRuiz: with 10 developers you can have a committer layer that in principle is not a bottleneck
(10:26:07) red1: carlos, can u list those 9 others down (again)
(10:26:14) Bahman: And you suggesting they become the new CL?
(10:26:19) Bahman: Alright...got it.
(10:26:26) red1: it seems so
(10:26:33) vpj-cd: yes but if trunk is only to bugs
(10:26:50) red1: lets try that idea, and review again next time
(10:26:55) Michael_Judd: I think victor's got a point - only bug fixes should go to the trunk ...
(10:26:56) Bahman: And what will be the policy to bring in new developers?
(10:26:57) vpj-cd: this will are minus
(10:27:06) CarlosRuiz: and the idea is to grow the committers when they demonstrate that they know how to touch the core
(10:27:16) red1: agree Carlos
(10:27:19) Bahman: How they demonstrate Carlos?
(10:27:28) karsten-thiemann: with bugfixes
(10:27:33) red1: exactly how u demonstrated Bahman
(10:27:35) CarlosRuiz: as you did Bahman, contributing good code
(10:27:36) Michael_Judd: so how do new committers apply to become committers ? and to whom ?
(10:27:38) vpj-cd: if we need touch the core to some improve then use a branch and 2pack
(10:27:54) red1: yep those buigfixes says what is good at
(10:28:02) CarlosRuiz: I think the important is not how much you know, but how much you learn
(10:28:03) vpj-cd: the community begin to test
(10:28:07) teo_sarca: About "only bug fixes should go to the trunk" , where will be the FR ?
(10:28:10) CarlosRuiz: and HOW YOU LEARN
(10:28:12) Michael_Judd: not really red1 - jsut what problems they have come accross ...
(10:28:15) Bahman: Good point Carlos!
(10:28:17) vpj-cd: even to set in production
(10:28:29) karsten-thiemann: good question from teo_sarca
(10:28:29) Michael_Judd: sometimes - I need help solving a problem but no one is interested in helping - so I try to do it myself ....
(10:28:39) red1: back to the question, do we have functional experts involving in this CL?
(10:28:40) vpj-cd: then the new enhancement can set into trunk
(10:28:42) vpj-cd: o via 2pack
(10:28:48) vpj-cd: we can say that is stable
(10:28:50) richard-4layers: Let's talk in practice: how can I become a committer ? And what will I do right now?
(10:29:06) CarlosRuiz: I think trunk must evolve with enhancements and functionalities
(10:29:18) vpj-cd: but I think that each enhancement need get you maturation
(10:29:20) CarlosRuiz: branches doesn't have the visibility to be enoughly tested
(10:29:21) karsten-thiemann: I agree Carlos
(10:29:32) red1: richard-4layers, once u fix bugs, others study them, and from there much can be seen to make a fair judgement
(10:29:41) hengsin: we need to classify the enhancement and bug fixed
(10:29:43) red1: has been this way ...
(10:29:50) hengsin: as in what is the level of risk
(10:29:52) richard-4layers: I'm going to help Tim with 2Pack issue... how can I share source code?
(10:30:01) teo_sarca: and we need to classify FR too ?
(10:30:01) red1: originally CC was Victor, Carlos and Trifon...
(10:30:02) hengsin: if it is high risk then it should goes into branch first
(10:30:05) CarlosRuiz: good point Hengsin
(10:30:29) teo_sarca: and what means maturity ? a partial working FR (without breaking bugs) or a complete working FR ?
(10:30:31) hengsin: yes, everything, bug fix can be high risk as well, you don't want people to fix 1 bug but introduce 100 new one!
(10:30:34) jsSolutions [email@example.com] ha entrado en la sala.
(10:30:35) red1: 2pack is in trunk, and Tim first started with 2 fixes
(10:30:37) CarlosRuiz: high risk, or just experimental must be in branch
(10:30:38) richard-4layers: I agre with jengsin
(10:30:58) red1: then Carlos studied them and he knew that it was fine for 2pack
(10:31:02) Bahman: Right Carlos.
(10:31:10) red1: but now Rob Klein says not so sure
(10:31:37) red1: and he is studying those new codes, his issue is "referencing' in the codes
(10:31:51) CarlosRuiz: what is 'referencing' ?
(10:31:52) red1: but then that is the process'
(10:31:59) Bahman: So we're gonna have trunk for low-priority bugs and FR and devel branch for high-priority ones.
(10:32:06) CarlosRuiz: nope
(10:32:09) red1: i cc the email to team
(10:32:18) hengsin: Bahman, not priority but the level of risk involve
(10:32:21) CarlosRuiz: I think the trunk must evolve
(10:32:29) Bahman: Got it hengsin.
(10:32:45) red1: 2pack has an undo function.. guess that is a referencing issue if not done well
(10:32:54) CarlosRuiz: just we want to clarify how to evolve in an ordered way
(10:33:04) vpj-cd: yes Red1 but Tim go the hand
(10:33:09) hengsin: I guess referencing as in how to resolve dependency between elements
(10:33:25) red1: what i do is when i see someone else touching another person's code, i inform the original owner to respond
(10:33:26) vpj-cd: I have to invest many time , and I am happy because now he can contribute
(10:33:35) red1: so that he doesn't get broken off from what he did
(10:33:56) red1: so what Tim did was constant communicating right from the start
(10:34:23) red1: i cant say for a few others... some hardly communicate
(10:34:30) CarlosRuiz: I think undo have never worked, have design problems
(10:34:45) CarlosRuiz: but it deserves another talk, not here :-)
(10:34:46) karsten-thiemann: that might be true
(10:34:53) richard-4layers: I think it's the issue he asked me to watch at...
(10:35:21) CarlosRuiz: ok, are we ready to conclude something about 2?
(10:35:22) vpj-cd: I only explit the issue
(10:35:28) CarlosRuiz: can we propose who can be committers?
(10:35:32) vpj-cd: he undestand he try solve :-)
(10:35:39) red1: for example, this richard here, has been seen recently and actively so now... that is a right way to go... soon he gets the grasp of things.. and he shuld be in
(10:36:14) red1: and now even Marco Lombardo, he just wrote in wiki, but it shows that he been following our development right from the bottom
(10:36:22) red1: that was amazing
(10:36:36) CarlosRuiz: I think there are posterita guys very skilled
(10:36:37) richard-4layers: Marco is excellent!
(10:36:39) Bahman: New functionalities and enhancements like RE should be in branches?
(10:36:41) vpj-cd: I think is simple fist need definite basic rule
(10:36:42) CarlosRuiz: but not working on trunk
(10:36:54) karsten-thiemann: I still like the pair committer rule :)
(10:37:05) CarlosRuiz: how is that Karsten?
(10:37:09) red1: so a 5 X 2 pairs?
(10:37:11) richard-4layers: what about Piero Berritta?
(10:37:13) karsten-thiemann: not for bugfixes but for new features
(10:37:25) vpj-cd: It hink the new enhancement only go trunk when is stable
(10:37:30) karsten-thiemann: you need to get your code reviewed by another committer
(10:37:42) red1: well if u can get him richard... he doesnt show up much beyond just giving his works
(10:37:49) CarlosRuiz: yes, but who are committers?
(10:37:53) richard-4layers: I'll try...
(10:38:02) red1: those 10
(10:38:05) karsten-thiemann: and you have to mention him on the committ notice
(10:38:10) vpj-cd: remember we talk critical business application
(10:38:57) teo_sarca: Carlos, can you build a list ?
(10:39:05) red1: but again, comminication... changes need to be published.. and comments taken in
(10:39:31) CarlosRuiz: I think that's the most important issue, a previous (not post) communication
(10:39:32) karsten-thiemann ha salido de la sala (quit: Read error: 104 (Connection reset by peer)).
(10:39:34) Bahman: Yes red1, outta communication new developers are born!
(10:39:35) karsten_thiemann [firstname.lastname@example.org] ha entrado en la sala.
(10:39:43) red1: the release notes so far that Carlos did has made things visible
(10:39:51) vpj-cd: we need more development document
(10:40:00) red1: i suggested AS to assist in doing the secretarial work for Release notes
(10:40:37) CarlosRuiz: there can be posterita committers?
(10:40:39) red1: the CL is a tip of an iceberg working here
(10:40:43) CarlosRuiz: in trunk I mean?
(10:40:45) fredtsang: yeah sure carlos
(10:40:57) red1: posterita is now in branch
(10:40:57) fredtsang: I can definitely recommend someone
(10:41:20) red1: i m communicating with Shameem closely on pair testing
(10:41:36) red1: he has helped me to review the codes (just started)
(10:41:47) CarlosRuiz: who could be, Fred?
(10:41:56) fredtsang: ashley
(10:42:15) fredtsang: and vishee
(10:42:17) fredtsang: can be
(10:42:18) red1: there are 3 projects from posterita
(10:42:25) hengsin: how about sendy, fred ?
(10:42:31) red1: POS, Webstore, ZK
(10:42:42) fredtsang: yes sendy also
(10:42:45) fredtsang: on ZK
(10:42:54) fredtsang: ashley can touch everything
(10:43:01) fredtsang: and he has very good vision
(10:43:02) red1: Shameem is on POS rite?
(10:43:23) fredtsang: yes, he is responsible
(10:43:36) fredtsang: but he still has to check a lot of things with our seniors
(10:43:53) fredtsang: so I can make him responsible for this
(10:44:01) red1: i think for posterita... they considered the commiters
(10:44:13) red1: the app itself is highly matured
(10:44:43) fredtsang: we do have some other issues
(10:44:50) fredtsang: on adempiere itself
(10:45:22) CarlosRuiz: what do you think on this list:
(10:45:22) CarlosRuiz: Committers: Hengsin, Teo, Trifon, Bahman, Karsten, Johannes, Phib, Ashley, Colin, Fer_luck, CarlosRuiz
(10:45:22) CarlosRuiz: In the waiting list: Kontro, Tim, Vishee, Sendy, Victor
(10:45:48) CarlosRuiz: anyone else to be on the committer or waiting list?
(10:45:54) fredtsang: and shameem just for posterita
(10:45:58) fredtsang: carlos
(10:46:13) red1: POS side that is
(10:46:17) fredtsang: yes
(10:46:22) fredtsang: right :)
(10:46:22) Michael_Judd: I'd like to put myself forward ....
(10:46:33) fredtsang: yes i would welcome that
(10:46:40) red1: i think Mike is the SME for Accounting
(10:46:48) red1: very much needed
(10:46:54) karsten_thiemann: right
(10:47:03) Bahman: Yes.
(10:47:04) CarlosRuiz: is different to be a functional reviewer than a committer
(10:47:11) CarlosRuiz: committer must be developer
(10:47:14) red1: Mike for FR
(10:47:40) red1: but he loves coding.. and he is good only when he doesnt ask me for advice
(10:47:43) fredtsang: ok mike would be good to prevent some commits
(10:47:44) Michael_Judd: Carlos - are you a developer or a FR ?
(10:47:50) CarlosRuiz: anyway, I think Mike is not a developer (as he declared), but his code is enoughly good
(10:47:52) CarlosRuiz: what is a FR?
(10:47:58) CarlosRuiz: ah ! Functional reviewer
(10:47:59) karsten_thiemann: yes we need a functional reviewer layer for feature requests
(10:48:00) red1: Carlos is both
(10:48:11) red1: Colin too
(10:48:18) fredtsang: does FR look at the code after or before?
(10:48:21) Michael_Judd: red1 - my point ...
(10:48:22) fredtsang: commit
(10:48:29) muthah_ ha salido de la sala (quit: No route to host).
(10:48:31) karsten_thiemann: before developing
(10:48:37) red1: i think coders refer to FR first
(10:48:38) fredtsang: ok good
(10:48:39) karsten_thiemann: they look at the idea
(10:48:49) red1: whats the point of coding things the other way?
(10:49:06) CarlosRuiz: ok, what do you think on this list?
(10:49:06) CarlosRuiz: Committers: Hengsin, Teo, Trifon, Bahman, Karsten, Johannes, Phib, Ashley, Fer_luck, CarlosRuiz
(10:49:06) CarlosRuiz: In the waiting list: Kontro, Tim, Vishee, Sendy, Shameem
(10:49:06) CarlosRuiz: Functional Reviewer: Colin, Michael_Judd, Victor
(10:49:07) red1: one thing i read from JJ's notes...
(10:49:25) karsten_thiemann: I think we should always have a Feature Request for new functionality - followed by a discussion
(10:49:43) karsten_thiemann: and the FR give their comments here
(10:50:00) karsten_thiemann: after voting for new functionality the coding starts
(10:50:10) Bahman: The discussion maybe in forums also...so a link would suffice.
(10:50:16) Bahman: may be
(10:50:16) karsten_thiemann: and committers review the code before it goes to trunk
(10:50:18) red1: i think posterita shuld be its own commiters - just like libero
(10:50:20) jsSolutions: yes, FRs should comment on the requirements and review the funtionality through the gui once it is available
(10:50:31) CarlosRuiz: red1, the list is just for trunk
(10:50:41) red1: is 2pack in trunk?
(10:50:41) CarlosRuiz: anyone can ask for a branch for experimental
(10:50:46) CarlosRuiz: yes, 2pack is in trunk
(10:50:58) red1: so Tim cannot commit to trunk just for 2pack?
(10:51:27) ***CarlosRuiz thinking ...
(10:51:40) red1: perhaps put some rules onto such apps
(10:51:50) teo_sarca: carlos, i think that there can be some commiters for some modules ....
(10:51:57) teo_sarca: like 2pack
(10:51:57) red1: posterita i m pretty confident their whole gang know what they doing under Ashley
(10:52:03) vpj-cd: Ye Carlos I accept to functional review
(10:52:04) richard-4layers: Who are the Comitters and CC for libero?
(10:52:15) CarlosRuiz: so, Tim's case, we can have a specific module committer, or make Tim pass through committer filter while he becomes a full committer
(10:52:15) red1: libero is under Victor
(10:52:34) jsSolutions: my concern is if module touches other pieces
(10:52:45) jsSolutions: module maintainer can have limited focus
(10:52:46) red1: precisely.. thats the rule
(10:52:57) fredtsang: yeah i agree
(10:53:07) CarlosRuiz: yes, initially I would prefer to start without specific module developers
(10:53:21) CarlosRuiz: but Tim can be promoted later
(10:53:41) teo_sarca: carlos, but sooner or later this will be a must
(10:53:51) CarlosRuiz: ¿?
(10:53:56) teo_sarca: to have modules commiters
(10:53:57) Bahman: A module head and module developers....module developers commit the code which change other parts of ADempiere through module head.
(10:54:00) red1: i tot u made Tim a maintainer for 2pack? wonder if maintainers still apply
(10:54:21) CarlosRuiz: yes, and Alejandro/Johannes jasper committers
(10:54:22) Bahman: I mean head should confirm them and head is a member of CL.
(10:54:27) jsSolutions: bahnamn- good
(10:54:27) red1: thus they can have responsiblities and there is delegation
(10:54:33) teo_sarca: with the mention that module commiters DON'T change the core !
(10:54:43) red1: thats rite teo
(10:54:46) hengsin: yes, we need module lead, the rest is just communication and design issue
(10:55:01) red1: and in the FR of accounts for example.. we need Mike
(10:55:15) teo_sarca: if a module commiter need changes to core, he makes a request
(10:55:23) Bahman: To the head.
(10:55:27) red1: yep
(10:55:28) CarlosRuiz: what are adempiere modules?
(10:55:38) karsten_thiemann: :) good question
(10:55:42) Bahman: If Head confirms he commits it on his own :-D
(10:55:43) teo_sarca: :)))
(10:55:44) teo_sarca: true
(10:55:52) richard-4layers: I think for each FR module there should be a functional expert...
(10:55:58) Bahman: Currently Libero!
(10:56:09) red1: posteritas
(10:56:18) CarlosRuiz: what are adempiere modules in trunk?
(10:56:18) teo_sarca: hey, basically adempiere is not modular
(10:56:23) karsten_thiemann: the rest
(10:56:29) teo_sarca: but we can drag some lines
(10:56:30) teo_sarca: for the start
(10:56:32) teo_sarca: so we have 2pack
(10:56:34) richard-4layers: looks is a module
(10:56:36) teo_sarca: application dictionary
(10:56:44) teo_sarca: postgresql port
(10:56:45) richard-4layers: webStore is a module
(10:56:50) Bahman: Agree with Teo.
(10:56:55) CarlosRuiz: AD is not a module
(10:57:02) teo_sarca: is the core
(10:57:04) richard-4layers: webUI (serverApps) too
(10:57:05) teo_sarca: AD is the core
(10:57:07) karsten_thiemann: agree CarlosRuiz
(10:57:12) teo_sarca: but you can see it as a module
(10:57:13) red1: shuld nominate here also an architect team
(10:57:19) Bahman: But those modules are so bound to core that maintaining them as module makes the situation more complex.
(10:57:21) Bahman: IMHO
(10:57:31) karsten_thiemann: accounting
(10:57:36) karsten_thiemann: maybe
(10:57:54) hengsin: workflow
(10:57:55) CarlosRuiz: ok, maybe we can define what core is?
(10:58:08) richard-4layers: Here comes another question: dbPort to me is the persistence layer shared in the whole app... So why moving sources to base?
(10:58:08) teo_sarca: you don't need functional knowledge to maintain the AD
(10:58:10) karsten_thiemann: core is what not is a module :)
(10:58:14) Bahman: Yes, this way's better Carlos.
(10:58:29) CarlosRuiz: hehehe
(10:58:45) teo_sarca: i mean business process knowledge
(10:58:47) CarlosRuiz: can we delimit core?
(10:58:48) vpj-cd: I think 2 part inportant AD and java code core
(10:58:55) karsten_thiemann: how about the printing engine
(10:58:59) moyses: Indeed
(10:59:01) moyses: are we guided just by developers?
(10:59:10) red1: kernel contributor shuld be here :>
(10:59:13) moyses: or by business process knowledge? accounting knowledge? et all
(10:59:17) Bahman: Now we have 100 modules and 10 developers!
(10:59:29) karsten_thiemann: and 2 FR
(10:59:29) vpj-cd: yes
(10:59:33) teo_sarca: 10 modules / developer :)
(10:59:55) red1: well no wonder OB has 70
(10:59:58) hengsin: Bahman, how you get the 100 ?
(11:00:04) teo_sarca: but there are not 100 modules...
(11:00:08) CarlosRuiz: do we have modules really?
(11:00:13) vpj-cd: but exist functional enhanced and architectural enhanced
(11:00:26) red1: M_classes mostly
(11:00:37) Bahman: Just exaggerating hengsin!
(11:00:56) teo_sarca: Carlos, imho there are no concrete modules, but we can look at them as modules (as much as posible)
(11:01:08) Bahman: I'd like the word functionality more than module, MHO.
(11:01:10) karsten_thiemann: we have some modules and the rest is more or less monolithc or has at least lots of connections
(11:01:20) CarlosRuiz: ok, we don't have sources divided by functional modules
(11:01:24) teo_sarca: we need to start somehow to decentralize this big maze
(11:01:25) Bahman: Right Karsten.
(11:01:43) vpj-cd: ok we need summary and continue advance , but need set checklist
(11:02:26) CarlosRuiz: ok, supposing that we have 10 committers, 5 in waiting list, 3 FR
(11:02:30) richard-4layers: Don't you think we can replace swing UI with a SWT one just changing looks module? So it is a module...
(11:02:33) CarlosRuiz: what must be the procedure to commit?
(11:02:53) red1: what mike pulled out from the wiki?
(11:03:02) CarlosRuiz: 1 -> just committers can touch the trunk
(11:03:10) CarlosRuiz: nobody else allowed
(11:03:11) moyses: what is need to become a FR?
(11:03:12) karsten_thiemann: 1. start with confirmed bug report or with a discussed feature request
(11:03:15) Michael_Judd: http://www.adempiere.com/wiki/index.php/SVN_Commit_Policy
(11:03:45) hengsin: SWT port is not tough, just the matter of there are people committed to it :)
(11:03:57) Michael_Judd: So what happens when I find a critical bug that effects a customer?
(11:04:01) CarlosRuiz: what is the role of a committer?
(11:04:20) CarlosRuiz: how a committer is going to receive a contrib?
(11:04:21) Michael_Judd: I will fix it locally - and because it is even more difficult than it is now - I won't bother putting it back in the trunk....
(11:04:30) Michael_Judd: Why should i ?
(11:04:36) teo_sarca: hehe
(11:04:36) CarlosRuiz: to not fix it next version
(11:04:42) karsten_thiemann: exactly
(11:04:49) Bahman: Specially the migration scripts!
(11:04:55) Bahman: Nightmares cloaked in SQL!
(11:04:58) karsten_thiemann: thats one reason why we are here
(11:05:09) teo_sarca: because you are selfish, don't want to maintain it :)
(11:05:18) Michael_Judd: well - I only need to be taught how to fish once - that will be the same for any new person ...
(11:05:30) Michael_Judd: yeah - I don't want to maintain ...
(11:05:34) red1: i think we shuld let the CL has final say and revert rights
(11:05:39) red1: more so than commit rights
(11:05:48) CarlosRuiz: nope
(11:06:02) CarlosRuiz: revert has popularity problems, hahahaha
(11:06:11) fredtsang: Is someone going to summarize the new procedures and write it in the wiki?
(11:06:11) richard-4layers: waht Judd means is maybe that making committing policy too complicate can make people lazy
(11:06:12) karsten_thiemann: there is no real difference between revert and commit
(11:06:16) Michael_Judd: It seems to me that with the new regime - I may as well just mirror everyone elses work locally and sell it under my own brand ....
(11:06:16) CarlosRuiz: honestly, every time I revert changes from someone is a problem for me
(11:06:24) Michael_Judd: Something like how adempiere started ?? ;)
(11:06:24) red1: but CarlosRuiz u rtold us about reverting as important measure
(11:06:43) CarlosRuiz: yes, but is a problem
(11:06:53) Bahman: Hmmm...Mike, it's that complex?
(11:06:59) red1: i think commit be as open as in wikipedia.. but the seniors (CL) have final say
(11:07:01) mark_o_: but it's the only way to learn
(11:07:03) moyses: the problem
(11:07:05) moyses: is
(11:07:07) red1: bad example
(11:07:07) CarlosRuiz: nope
(11:07:13) Michael_Judd: I've been watching the conversation and thinking - it is already costly enough to commit - if it is more costly - why do it - that is what some other people do
(11:07:15) red1: just testing...
(11:07:30) moyses: that where the FR enter in the whole picture? who will check for example if Carlos is reverting correctly or not?
(11:07:34) CarlosRuiz: Michael, we're talking about code, quality, security for your own customers
(11:07:38) red1: i got reverted before...
(11:07:50) hengsin: yes, that was me :)
(11:07:51) red1: felt hurt... but for a day
(11:08:10) Michael_Judd: what you are talking about it limiting my business and creating a monopoly over the code if I was being suspicious ...
(11:08:15) CarlosRuiz: yes, revert has alwyas a personal feeling
(11:08:22) red1: but we have to live by it
(11:08:27) CarlosRuiz: monopoly?
(11:08:28) Michael_Judd: and the next move I would make if I were you would be to set up a development team that had access to committ ;)
(11:08:43) red1: so far i see our developers here are quite mature to it
(11:08:43) vpj-cd: I think new functionality come via sponsored
(11:08:45) teo_sarca: personally i write mails before reverting :)
(11:08:52) vpj-cd: as now have the list in wiki
(11:08:56) red1: Victor got reverted the very first days
(11:08:58) CarlosRuiz: no red1
(11:09:09) Michael_Judd: like Kontro said in berlin - he is worried about exlcusvity ...
(11:09:10) vpj-cd: but born when existe the need
(11:09:11) Michael_Judd: me too ...
(11:09:29) CarlosRuiz: I'm worried about quality
(11:09:36) Michael_Judd: me too .....
(11:09:42) vpj-cd: yes I am agree with Carlos
(11:09:46) fredtsang: if we follow the example of wikipedia
(11:09:55) Michael_Judd: I'm also worried about the way we are trying to achieve it - and whether that will stiffle the project ...
(11:09:58) moyses: this will become somewhat bureaucratic in a way
(11:10:02) richard-4layers: It's better to lose some committers then stability...
(11:10:02) fredtsang: it's extremely open
(11:10:05) red1: quality can come from reverts
(11:10:05) vpj-cd: this is a critical business application
(11:10:13) CarlosRuiz: no red1
(11:10:17) Michael_Judd: One of the issues from Berlin is there is not enough developer - now we are making another bottleneck ...
(11:10:17) CarlosRuiz: reverts are not working
(11:10:19) red1: fredtsang, ever tot why?
(11:10:35) vpj-cd: no is a stuff to experimental
(11:10:38) Bahman: Even Athene (the pure democracy) was lead by a senate...but if talking to senate and impressing them becomes too complex and slow, then ancient Athene becomes current Iran :-)
(11:10:43) fredtsang: I can sense michael point
(11:10:48) fredtsang: and his concern
(11:11:00) CarlosRuiz: yes, me too, but I have problems every time I revert someone's code
(11:11:01) vpj-cd: Michil and adempiere no is exclusive
(11:11:08) red1: ok let us ask how many is many?
(11:11:17) CarlosRuiz: and there is another point
(11:11:17) vpj-cd: the development need give work
(11:11:24) fredtsang: we could do it the other way round
(11:11:26) red1: many can pose new problems
(11:11:34) fredtsang: not who we elect
(11:11:38) fredtsang: but who we kick out
(11:11:41) CarlosRuiz: reviewing commits is a hard task
(11:11:43) vpj-cd: Adempeire now accept contribution posteriata, libero , etc
(11:11:43) Michael_Judd: yes fredtsang - my point - QA in a branch ...
(11:11:56) red1: i think Carlos is sensitive person, he doesnt like to kick out
(11:11:58) Bahman: good point fredtsang.
(11:12:00) moyses: I agree with Michael
(11:12:08) CarlosRuiz: not because I'm sensitive
(11:12:11) vpj-cd: do not is esclusive
(11:12:15) CarlosRuiz: it's because people feel it personal
(11:12:31) CarlosRuiz: if a simple revert make people feel this way, imagine a kick out
(11:12:31) red1: sorry!
(11:12:37) Michael_Judd: ;) was that aimed at me Carlos ???
(11:12:39) Bahman: What if we put a disclaimer in our SVN commit policy page Carlos?
(11:12:48) fredtsang: well
(11:12:51) CarlosRuiz: You and every people I have reverted before
(11:12:52) fredtsang: if we have rules
(11:13:06) fredtsang: about 'kicking out'
(11:13:11) fredtsang: it can be temporary
(11:13:12) red1: i think those who felt hurt for more than a day wont last in a bazaar
(11:13:14) fredtsang: after violation
(11:13:20) CarlosRuiz: Michael, it was the first time for you, but maybe the fifth for me :-)
(11:13:20) Michael_Judd: Do you ever make mistakes Carlos ?
(11:13:21) fredtsang: of the commit rules
(11:13:26) Bahman: If the rules are defined and written and published then there will be no personal matter.
(11:13:27) CarlosRuiz: all the time Mike
(11:13:34) fredtsang: yes bahman
(11:13:42) fredtsang: i agree
(11:13:43) teo_sarca: agree Bahman !!!
(11:13:49) Michael_Judd: The trouble I had was I couldn't find the rule - they were creaed after I loked ...
(11:13:53) CarlosRuiz: not agree
(11:13:55) moyses: I second that Bahman
(11:13:56) red1: why dont we make a rule that reverts can happen by the CL?
(11:13:57) fredtsang: and if the 'punishment' is clear. then you can take your own responsibility
(11:14:01) Michael_Judd: If I could find the rules and follow a process there would not be a problem.
(11:14:05) fredtsang: instead of delegating that responsiblity elsewhere
(11:14:18) Michael_Judd: I was let down when I couldn't find this information and couldn't get any response
(11:14:19) CarlosRuiz: let me finish another point
(11:14:19) Bahman: Right fredtsang!
(11:14:34) CarlosRuiz: revieweing commits is a really hard task
(11:14:49) CarlosRuiz: I need to download / run migration scripts / imagine what people tried to fix
(11:14:52) CarlosRuiz: then make a test case
(11:14:57) CarlosRuiz: if something fails
(11:15:04) CarlosRuiz: I need to think if is easier to fix or to revert
(11:15:11) CarlosRuiz: and if I revert I need to cope with all the personal feelings
(11:15:14) CarlosRuiz: and explain a lot
(11:15:16) CarlosRuiz: and .....
(11:15:27) CarlosRuiz: that's just reviewing one commit
(11:15:38) CarlosRuiz: suppose now that 5 developers changed the same class
(11:15:39) Bahman: That's why I suggested test units and test procedures as mandatory part of commit or pathces.
(11:15:40) Michael_Judd: What would people say if we made it someones job (paid) to review commits ?
(11:15:40) red1: u need ppl to delegate to to do that revert then
(11:15:50) CarlosRuiz: the point is
(11:15:55) red1: u review, found something wrong post it out, someone else second it
(11:15:56) Michael_Judd: Carlos - this is a bazaar .....
(11:15:59) CarlosRuiz: when we grow in committers
(11:16:00) red1: just like in wikipedia
(11:16:07) CarlosRuiz: commits are going to be very correlated
(11:16:17) CarlosRuiz: and it becomes a Frankenstein
(11:16:19) CarlosRuiz: as Victor said
(11:16:22) fredtsang: agree michael
(11:16:23) CarlosRuiz: is not like wikipedia
(11:16:23) moyses: we must have rules to follow, more than one person should agree, I believe it can't be just one person decision to revert back something
(11:16:31) CarlosRuiz: code is harder to fix
(11:16:38) red1: i still havent absorb the frankenstein point fully but never mind
(11:16:55) red1: thus we need more eyeballs
(11:17:05) moyses: we need more eyeballs and rules
(11:17:08) CarlosRuiz: I think it can't be just one person decision to commit something
(11:17:10) vpj-cd: I remember my fist patch , and when Carlos revert
(11:17:11) red1: and they come when we somehow influence them via this bazaar
(11:17:13) moyses: parameters to measure something
(11:17:13) fredtsang: right moyses
(11:17:21) Michael_Judd: it is easy (now I know) to make sure code commits don't break the trunk - but who teaches you / shows you this ?
(11:17:35) vpj-cd: I fist are mad, but I undestand that is necesary
(11:17:40) richard-4layers: Ehi people I have to go soon, but I wanted to hear about point 2.... Where are we?
(11:17:46) fredtsang: things get extremely personal in wikipedia, if you follow the discussion, but the articles are good
(11:17:57) CarlosRuiz: wiki can't be compared with code
(11:17:57) vpj-cd: I am happy that Carlos was make this
(11:18:01) Michael_Judd: I agree victor - and you are here now because you commited got reverted and then commited again correctly ...
(11:18:03) Bahman: Yes..the outcome matter fredtsang!
(11:18:16) Michael_Judd: and that's why all developers are here - but what about in the future if the process changes ?
(11:18:29) vpj-cd: because he takecare the trunk
(11:18:30) vpj-cd: :-)
(11:18:45) moyses: I am not a developer
(11:18:52) moyses: I am just a FR
(11:18:56) CarlosRuiz: do you want quality?
(11:19:02) hengsin: we should compare with other opensource development like linux, eclipse, etc. they do have quite some rules ...
(11:19:10) moyses: but my main complain with Victor was that they don't test the changes to the code
(11:19:21) CarlosRuiz: do you want the security that someone skilled reviewed your code?
(11:19:33) red1: lets try accepting what Carlos propose for a time period and then agree to review after a time
(11:19:40) Bahman: What if Carlos goes to a journey for 3 months (just for example)? You can review the ADempiere commit status while Carlos was not at home some days ago...
(11:19:57) Bahman: This makes me think "We cannot and shouldnot depend on a few."
(11:20:05) red1: we had fun when Carlos was not around :D
(11:20:05) Michael_Judd: I feel we are approaching this from the wrong direction. The point is that QA is costly and no-one really wants to do it as it isn't very fun. Why not allow the commercial comapnies bear the cost .... ?
(11:20:12) Michael_Judd: and QA a branch ....
(11:20:17) moyses: I believe that behind a commit there must some samples showing how was corrected the bug or added the new functionality
(11:20:21) Michael_Judd: and stop stiffling new ideas ...
(11:20:24) red1: hmm can u do that mike?
(11:20:47) CarlosRuiz: I don't think so
(11:20:52) CarlosRuiz: imagine a branch with 200 commits
(11:21:01) CarlosRuiz: not reviewed before
(11:21:16) red1: any objections to the 10 + 5 + 2 in pairs framework?
(11:21:18) CarlosRuiz: some commits will broke the purpose of other commits
(11:21:33) vpj-cd: ok who make this in other project
(11:21:38) hengsin: michalel, trunk must of reasonable quality not to scare off all the potential commercial implementors for adempiere :)
(11:21:41) vpj-cd: we can adopted some model ?
(11:21:49) Bahman: Well, that's a sign that we gave the commit rights to the wrong person!
(11:21:58) Bahman: That's our mistake not his! MHO.
(11:22:00) red1: take for instance sponsored development.. it went thru some wiki process.. then its left to the CL who can refer to that wiki
(11:22:05) vpj-cd: how are to do TinyERP or SugarCRM ?
(11:22:53) CarlosRuiz: I repeat the question: do you want quality? do you want to have the security that someone reviewed code in Adempiere?
(11:23:02) CarlosRuiz: or do you prefer to allow all code to come in
(11:23:06) CarlosRuiz: and review it in a later stage
(11:23:11) red1: one golden rule i read in PM book, "Be bit better is good enoiugh for now"
(11:23:20) CarlosRuiz: with the possiblity that it can't be fixed because it becomes a Frankenstein
(11:23:25) moyses: before talking about wuality you must set the parameters against what quality is
(11:23:28) Michael_Judd: hengsn - but how do I get changes made without wasting time? In the future - I won't do the extra work to get the code back i nto the trunk as it is too costly. I will put it in to our repository and leave it there. This is what is already happening. If you are a development house - it is to your advantage because you can charge each implemntor time and time again for the same fix
(11:23:38) Michael_Judd: It is a good way to make money from Adempiere ;)
(11:23:40) Bahman: Later does not necessarily mean a week or month later Carlos...it can be a day later.
(11:23:40) vpj-cd: yes Carlos I want Quality is business aplication
(11:24:07) moyses: If you want to commit your changes you must document that your code changes are at least tested in a way
(11:24:12) vpj-cd: if we want experimental or play we can create a branch :-)
(11:24:15) CarlosRuiz: who is going to review a day later, Bahman?
(11:24:24) Michael_Judd: Yes carlos - I want quality too ..... but where should we do quality ?
(11:24:27) Bahman: The CL and FR.
(11:24:28) hengsin: Michael, that won't work for long term, the market doesn't have only one solution :)
(11:24:43) fredtsang: what about that initial idea of having 2 trunks?
(11:24:45) CarlosRuiz: can the CL and FR review 50 commits in one day?
(11:24:48) fredtsang: would it solve some issues?
(11:24:56) fredtsang: about open/quality
(11:24:59) Bahman: Guess not Carlos.
(11:25:01) Michael_Judd: At the expense of bug fixes fixing broken funtionality, making new functionality - if you raise the bar too high - people will keep their work private ...
(11:25:12) CarlosRuiz: that's the point Bahman, we're growing fast
(11:25:28) red1: some here thinks that no bar is the way to go (me)
(11:25:35) Bahman: With current situation they can't review...I again insist upon test functions and scenarios to be mandatory parts of commit patches.
(11:25:39) CarlosRuiz: I don't think so Michael, we're completely open for 2pack modules
(11:25:49) CarlosRuiz: 2pack modules don't pass through CL or FR
(11:26:03) red1: but it got a big boost from head of CC
(11:26:21) red1: i remember there was a small debate between 2pack and ADCK
(11:26:30) red1: and then Victor weighs in finally to use it
(11:26:30) Michael_Judd: I probably find 10-15 bugs a day - I don't have tme to log them all. I fix maybe 2-3 a day ... I don't have time to commit them.
(11:26:42) red1: pass it along..
(11:26:59) Michael_Judd: I have thousands of lines of code that I need to fix up for the project - but no time to do this ...
(11:27:15) moyses: I support the QA branch
(11:27:18) Bahman: That's the least to do when talking about bugs Mike IMHO.
(11:27:23) Michael_Judd: these bugs are the same bugs everyone else is fixing - but becausethe bar is too high - we do the same work many times
(11:27:29) Michael_Judd: isn't the answer to lower the bar
(11:27:34) teo_sarca: as fred said, maybe is time to consider (discuss) about a stable tag, what you think ? (yes, no, maybe)
(11:27:36) Michael_Judd: to get more input and then to control that ?
(11:27:38) red1: or train more?
(11:27:43) CarlosRuiz: low the quality?
(11:27:51) Bahman: Train more red1!
(11:28:01) red1: somehow we got good trainees !
(11:28:08) fredtsang: I am not sure if we lower bar, we lower quality
(11:28:14) fredtsang: we certainly may create more work
(11:28:17) Michael_Judd: raise the bar <> quality ...
(11:28:18) fredtsang: as there are more newbies
(11:28:28) CarlosRuiz: raise
(11:28:31) Michael_Judd: microsfot have very few committers - that <> quality
(11:28:32) CarlosRuiz: is for quality
(11:28:45) red1: precisely fredtsang .. think European Union post Poland
(11:29:03) red1: we may have to compromise.. but then thats me
(11:29:06) CarlosRuiz: Microsoft problem is not about commits, is about marketing, we don't have such problem
(11:29:19) Michael_Judd: I think you make rules for Poland and then police them hard in the ist few years - then there is no longer a need to police them hard...
(11:29:50) CarlosRuiz: Fred, who is going to review the more work?
(11:29:52) red1: and the british pound remains..
(11:29:52) fredtsang: A personal experience we had recently
(11:29:53) CarlosRuiz: if we lower the bar?
(11:29:58) moyses: the process that Carlos explained about reviewing every commit must be documented
(11:30:17) CarlosRuiz: and newbies are the more likely to commit bad code or duplicate
(11:30:19) moyses: I believe that every developer may apply that or something similar to that before commiting
(11:30:41) CarlosRuiz: it's dangerous to have 500 newbies!!!
(11:30:47) CarlosRuiz: with a low bar
(11:30:47) fredtsang: yes
(11:30:48) Michael_Judd: CarlosRuiz - I am suggesting making some rules about committing and then as long as people meet the rules allow the commit.... lower the bar - and then QA in a branch ...
(11:30:51) fredtsang: that's extreme
(11:30:56) moyses: it could be wonderful to have 500 newbies!
(11:31:02) moyses: with a good couch!
(11:31:05) red1: this Open Source idea is viral
(11:31:10) Bahman: But that's the price we should pay to have 10 skilled developers Carlos.
(11:31:28) CarlosRuiz: yes Bahman
(11:31:36) CarlosRuiz: but you don't need to low the bar to have them
(11:31:45) CarlosRuiz: you can coach them without allowing commits
(11:31:48) CarlosRuiz: this is another point
(11:31:51) red1: ok lets define the bar
(11:31:53) Bahman: Once there are another bunch of 10 skilled developers there's gonna be less pressure on you and active CC members.
(11:32:01) moyses: how can you coach them Carlos?
(11:32:08) red1: i think the way Carlos define it is quite low
(11:32:18) Michael_Judd: Carlos has no time !!!! He spends it all reverting code ;)
(11:32:19) CarlosRuiz: as Victor coached Tim?
(11:32:24) red1: he said, if he is like Tim, then he can be in
(11:32:35) mark_o_ ha salido de la sala (quit: Read error: 104 (Connection reset by peer)).
(11:32:40) red1: and now Tim coaching AS
(11:32:45) hengsin: isn't it newbies should start with submitting patches first ? that's how I get started
(11:32:51) red1: or rather. everyone else coaching AS
(11:32:53) Michael_Judd: But really - Carlos doesn't have enugh time - nor do I! This is a problem because we are open source - otherwise our license would cover the cost
(11:32:59) teo_sarca: me too
(11:33:09) red1: haha
(11:33:18) moyses: we are running circles here
(11:33:27) fredtsang: moyses right
(11:33:28) CarlosRuiz: that can be solved having paid committers ;-)
(11:33:28) hengsin: and from time to time, we can give commit right to people with good patches
(11:33:31) teo_sarca: yes
(11:33:37) Michael_Judd: What about the pairs idea? What about having someone that can review and coach ..
(11:33:42) red1: i m telling u ... i translated the spanish words... Carlos said the bar is really low
(11:34:04) CarlosRuiz: red1, is a demand/offer problem
(11:34:11) Michael_Judd: CarlosRuiz Thats what I worry about - unless it is centralised ...
(11:34:40) red1: Michael_Judd, proposed a good point on formulating the coaching process along with it
(11:34:52) red1: that can be part of the bar
(11:34:53) fredtsang: What about an initiation program for the new committers
(11:34:55) fredtsang: ?
(11:34:56) CarlosRuiz: do we need to open commit to coach?
(11:35:06) CarlosRuiz: what's the point?
(11:35:11) red1: if u sin, u get to pray in the temple
(11:35:13) vpj-cd: yes Carlos I coach Tim with agreement the give your time to Adempiere :-)
(11:35:13) Michael_Judd: If I could ask someone about some code before I commit I would. I have tried and no-one helps ... so I commit and then I get eyeballs ;)
(11:35:13) Bahman: Good fredtsang! We need to breed new developers!
(11:35:45) CarlosRuiz: or you could ask in Forums Michael
(11:35:47) red1: have to agree with Mike, sometimes bad codes draws ppl
(11:35:49) moyses: Agree fredtsang!
(11:35:59) red1: just like a bad compiere
(11:35:59) vpj-cd: I am try help to AS but I ask he that make a tutotial
(11:36:17) CarlosRuiz: we get out of the point
(11:36:20) richard-4layers: I am doing it too
(11:36:21) vpj-cd: with you expetice then AS create a new page in wiki
(11:36:26) CarlosRuiz: having new committers <> allowing commit for everyone
(11:36:54) richard-4layers: tutoring AS...
(11:37:04) red1: i m seeing CarlosRuiz the CC or CL more as police cum coaches which is what we need as rules enforcers
(11:37:10) CarlosRuiz: what about -> new committers can be proposed by everyone and we vote for them to be included in CL ?
(11:37:29) richard-4layers: that's good
(11:37:33) red1: ok
(11:37:33) hengsin: new comer should start with submitting patch to the project.
(11:37:38) Bahman: Let's look at this subject from another view (previous one): How new committers will be bred?
(11:37:40) CarlosRuiz: -> to become a committer you must show at least n contributions ?
(11:37:40) red1: thats right
(11:37:53) moyses: I believe in the coaching idea
(11:37:56) fredtsang: bahman nice perspective
(11:38:05) red1: coaching has been going on fine...
(11:38:15) moyses: CC may take a newbie developer and check their patchs
(11:38:17) red1: hardly see anyone shouting here
(11:38:28) hengsin: we don't need a hard rule for that, just in general, if your patch are good, we will give you commit right.
(11:38:31) CarlosRuiz: CC?
(11:38:34) Bahman: How many coaching processes do we have red1?
(11:38:43) red1: one
(11:38:49) Michael_Judd: I think we are solving the wrong problem the wrong way and think we should tkae a step back andlook at the big problem
(11:38:57) Michael_Judd: if not - then there's no point really ...
(11:39:07) Bahman: So, how it's fine?! I don't see developer training in ADempiere.
(11:39:07) CarlosRuiz: agree Michael
(11:39:13) CarlosRuiz: I repeat the question: do you want quality?
(11:39:17) moyses: agree bahman!
(11:39:20) Bahman: You must live on your own to become a developer.
(11:39:22) Michael_Judd: I repeat YES!!!!
(11:39:23) CarlosRuiz: do you want review on code entering Adempiere?
(11:39:32) Michael_Judd: Is this the way to get qulity - I don't think so !
(11:39:37) CarlosRuiz: do you want the security of review?
(11:39:39) moyses: Bahman touched a very importan point!
(11:39:43) vpj-cd: If you support to Adempiere the community support to you :-)
(11:39:50) red1: ok let me try... can i propose Mike Jude to be in the CL?
(11:40:01) Michael_Judd: Its not the point red1
(11:40:17) Michael_Judd: the point is what we should do to make the project really good....
(11:40:21) red1: i tot we just passed that rule to propose (testing..)
(11:40:26) Michael_Judd: and we are trying to solve the wrong problem here today
(11:40:27) moyses: we are not solving the problem from its roots
(11:40:38) CarlosRuiz: ok
(11:40:46) CarlosRuiz: 2 - Commit committee can't cope with the review of all commits
(11:40:48) Michael_Judd: exactly - we are trying to come up with a quick fix and not looking at the bigger problem
(11:40:54) CarlosRuiz: this is a potential quality problem
(11:41:00) CarlosRuiz: that's the reason of my questions
(11:41:04) red1: well.. mike.. i tot u want to be in the CL.. so what is the real point?
(11:41:10) Michael_Judd: yes - so how many people are committing daily .... ?
(11:41:32) hengsin: zero
(11:41:38) moyses: the problem is that the commits are not passing the QA
(11:41:39) Michael_Judd: red - I don't want to feel excluded and I don't want to increase the 'cost' to me of contributing to the proejct
(11:41:42) vpj-cd: I think the we can commit if we have 2 people with +1 vote
(11:41:46) moyses: but who defined that QA?
(11:41:48) CarlosRuiz: nobody from CC is paid to do that, nobody has the security that a commit is reviewed
(11:41:56) vpj-cd: and a vote include the commintment they test :-)
(11:41:57) Michael_Judd: Exactly!
(11:42:02) moyses: who trained those guys about good adempiere code development?
(11:42:28) vpj-cd: we need also more info to developer
(11:42:37) CarlosRuiz: again
(11:42:39) moyses: we are not looking at the whole picture here
(11:42:40) CarlosRuiz: we go out of the point
(11:42:50) Bahman: We are short of skilled developers:
(11:42:50) Bahman: 1. How does this effects the quality of ADempiere?
(11:42:50) Bahman: 2. Can it be compensated by a developer training programme?
(11:42:50) Bahman: 3. How we can expand the skilled developers set?
(11:43:05) moyses: excellent points Bahman!
(11:43:07) red1: bahman, moyses, coaching is a context here...
(11:43:10) vpj-cd: they need know how do not break the trunk
(11:43:16) Michael_Judd: we also need a easy way for a FR to work with a developer ....
(11:43:18) moyses: so much better question that just Do you want quality?
(11:43:23) Michael_Judd: like a left and right hand ...
(11:43:26) Michael_Judd: working together
(11:43:33) red1: its not formal.. but coaching went on in a highly charged environment.. and that is better than any army i seen
(11:43:34) CarlosRuiz: hey, you must be joking
(11:43:57) CarlosRuiz: if you think a developer trainee program can solve the quality problem, you don't know developers ;-)
(11:44:00) moyses: totally agree Michael
(11:44:06) Bahman: Absolutely right Mike!
(11:44:26) Bahman: It does solve Carlos!
(11:44:29) Michael_Judd: developers are nothing without the fr and fr is useless without the developer
(11:44:37) Bahman: How the people in list became developers?
(11:44:42) moyses: Carlos, you must give them more power thorugh more knowledge
(11:44:43) CarlosRuiz: some developers are FR
(11:44:48) red1: they try, and try..
(11:44:51) CarlosRuiz: some FR can develop
(11:44:53) moyses: you must be kidding with your comment Carlos
(11:44:56) Michael_Judd: some special people can do both - but if we want more people to contribute we need to make it easier for them to do it - and that means more collaboration
(11:44:59) Bahman: it costed me more than 500 hours...
(11:45:05) CarlosRuiz: power = open commit?
(11:45:10) red1: and bahman really tried
(11:45:19) CarlosRuiz: we're again out of focus
(11:45:22) Bahman: that's why we can't expect every one to become a developer on his own.
(11:45:30) Bahman: We must push them...help them climb.
(11:45:35) moyses: your focus is too narrow Carlos, pardon me
(11:45:41) Michael_Judd: Bahman - look at my last 6 years - look at the guys I sent to Compiere taining porgrammes (what a laugh)
(11:45:43) CarlosRuiz: my focus is on quality
(11:45:46) moyses: we are pointing out something really important
(11:45:55) fredtsang_ [n=chatzill@ADSL-TPLUS-6-42.intnet.mu] ha entrado en la sala.
(11:45:56) CarlosRuiz: yes, but out of the focus of this meeting
(11:45:57) red1: i missed that moyses
(11:46:06) AS6: Carlos's point on quality make sense. And certainly Michael_Judd's view are very vital to OSS community. Both make sense. A rigid environment may deter contributors from nearing Adempiere , who may have a high future value towards Adempiere. The question to ask is, how can we have a fair balance between quality and flexibility to submit? There must be way to this, if we could accept the fact that both flexibility and security is importa
(11:46:12) CarlosRuiz: if you want we can discuss about the adempiere university
(11:46:15) richard-4layers: I'm sorry but i have to quit.
(11:46:16) vpj-cd: yes Carlos I think is quality is fist
(11:46:17) CarlosRuiz: this won't solve the quality problem
(11:46:30) red1: thanks richard-4layers ...
(11:46:32) richard-4layers: Let me know final decisions... See you soon
(11:46:40) Bahman: Bye richard-4layers!
(11:46:45) teo_sarca: bye bye
(11:46:51) red1: we will publish a long wiki
(11:46:53) moyses: you are looking at short carlos
(11:46:53) vpj-cd: this this reason good rule the do not break the trunk
(11:47:00) moyses: you are not looking at the long term
(11:47:02) dladwig_Idalica: I reccomend a tirage process--- we take all code but depending on your proven level or quality you may have to wait and go through a triage
(11:47:03) CarlosRuiz: Moyses, I'm trying to solve a problem
(11:47:06) Michael_Judd: it might solve the quality question actually ...
(11:47:08) moyses: you want to solve the problem now
(11:47:09) richard-4layers ha salido de la sala (quit: "See ya").
(11:47:11) CarlosRuiz: not reconstructing the world
(11:47:13) red1: i still think that what carlos propose is not a high bar.. so let us try it out
(11:47:20) Bahman: Who is gonna publish that red1? We are running short of personnel here.
(11:47:24) moyses: i dont agree red1
(11:47:26) CarlosRuiz: when solved the problem I'll try to reconstruct the world :-)
(11:47:27) red1: carlos
(11:47:31) vpj-cd: then we have that to go evolve but without the quality
(11:47:43) CarlosRuiz: ok, let me show in other way
(11:47:47) Michael_Judd: I don't think this is a quick fix ....
(11:47:48) CarlosRuiz: can we just drop CC
(11:47:49) moyses: then problems arises just like some bad decisions from JJ
(11:47:57) red1: now there are 10 plus committers... an open voting system..
(11:48:00) CarlosRuiz: it's a lower bar
(11:48:03) CarlosRuiz: we can drop the CC
(11:48:04) red1: a process of sanity
(11:48:08) CarlosRuiz: and let this the bazaar way!!!!!
(11:48:12) red1: ppl like Tim getting in
(11:48:12) CarlosRuiz: like a wiki!!!!
(11:48:14) Michael_Judd: What would be great for quality - wouldbe to get a FR Expert and a developer to work together (or in groups)
(11:48:19) red1: and soon AS
(11:48:25) red1: Bahman is already in
(11:48:28) Michael_Judd: Then the collaborative effort would be committed as a workign group ....
(11:48:32) red1: i would nominate Moyses for FR
(11:48:36) CarlosRuiz: agree?? we can start today without CC?
(11:48:41) red1: i prefer Mike to be the Chancellor
(11:48:48) moyses: I like Michael Judd approach
(11:48:59) AS6: Bahman: Dont worry about the limited personnel. Adempiere is open to the 6.6 Billion people. More will come. Its just that we need conducive environment.
(11:49:00) red1: wait till i land in London
(11:49:02) moyses: If I am accepted as a FR who would be the right hand?
(11:49:04) jsSolutions: I missed a lot of this, but we cannot agree her to drop CC
(11:49:08) moyses: because I will be the left
(11:49:23) red1: some had too many beers jsSolutions
(11:49:25) Bahman: They should learn AS...we must transfer what we know to the next wave....
(11:49:29) vpj-cd: what is the issue with current approach
(11:49:31) jsSolutions: outcome of this discussion must be posted for consideration and vote
(11:49:33) red1: Mike started it :>
(11:49:34) jsSolutions: in forums
(11:49:37) vpj-cd: somebosy set a bug
(11:49:40) vpj-cd: some fixed
(11:49:45) Michael_Judd: We need to organise ourselves a little to make it easier to get more people involved - that is lowering the bar to get involved will increase the quality ...
(11:49:46) vpj-cd: and set patch
(11:49:46) moyses: whoo will be my partner in this modell, who will be de developer that I will work closely?
(11:49:50) CarlosRuiz: I'm exaggerating Joel
(11:50:00) CarlosRuiz: trying to get people again in the point
(11:50:03) vpj-cd: 2 people test and give +1 vote then this go trunk
(11:50:07) jsSolutions: i know u CR, I'm backing u
(11:50:08) red1: too much European tour
(11:50:09) CarlosRuiz: 2 - Commit committee can't cope with the review of all commits
(11:50:13) vpj-cd: what is the issue withis aproach?
(11:50:35) CarlosRuiz: thanks Joel :-)
(11:50:41) moyses: the problem is that you dont test enough your code victor
(11:50:46) vpj-cd: Carlos give the review the community
(11:50:51) moyses: and you commit it!
(11:50:56) CarlosRuiz: 2 - Commit committee can't cope with the review of all commits
(11:50:56) CarlosRuiz: Solution: low the bar -> allow everyone to commit !!!
(11:50:57) vpj-cd: if I vote the rule is I vealidate the patch
(11:51:05) red1: we must help victor test
(11:51:22) red1: the contributor can never be guilty
(11:51:23) CarlosRuiz: 2 - Commit committee can't cope with the review of all commits
(11:51:23) CarlosRuiz: > Solution -> Drop CC
(11:51:27) Bahman: Carlos, that's not the ultimate solution...we're just discussin :-)
(11:51:30) vpj-cd: if this is teste to more the a people is simple only apply the patch file
(11:51:38) moyses: who will be the guilty red1?
(11:51:41) red1: yes for the last 2 hrs
(11:51:42) Michael_Judd: because victors code isn;t i nthe trunk - I dont test it. I have looked at it - and there are many accounting issues in there ...
(11:51:44) CarlosRuiz: yes, but we're really out of focus
(11:51:49) red1: the others moyses
(11:51:50) vpj-cd: then the contribute have find the support the other memeber to tester you code
(11:51:52) Michael_Judd: but I don't look at it if it is not in the trunk ....
(11:51:54) red1: those who stood by
(11:52:03) red1: those outside this room perhaps
(11:52:03) CarlosRuiz: motion to focus on the problem -> 2 - Commit committee can't cope with the review of all commits
(11:52:03) CarlosRuiz: can we?
(11:52:03) vpj-cd: this way all work I think
(11:52:08) Michael_Judd: what about another solution carlos ?
(11:52:26) CarlosRuiz: I'm open to hear, that's why I'm here
(11:52:32) moyses: what do you want then carlos?
(11:52:34) Michael_Judd: What about having some commit groups rather than people ....
(11:52:42) Michael_Judd: the groups can organise themselves
(11:52:47) vpj-cd: if somebody do not tenst then he can not vote
(11:52:49) Michael_Judd: and they must follow rules
(11:52:53) moyses: you rather tell us and we just will tell YES right away
(11:52:57) teo_sarca: should we work in branches, and maintain a stable branch ???
(11:52:58) moyses: is that fine with you?
(11:53:27) moyses: I like teo_sarca idea
(11:53:36) Bahman: Hmm...
(11:53:38) vpj-cd: ok this my summary idea
(11:53:43) CarlosRuiz: I have a question : Do you think I'm being too strict with trunk?
(11:53:47) Michael_Judd: who will merge the branches to the trunk ? it is even more work ....
(11:53:54) vpj-cd: 1.- all bug should are report via sf
(11:53:58) CarlosRuiz: is the person the problem?
(11:54:13) Bahman: Mike's idea looks interesing but how groups are organised?
(11:54:17) Bahman: You have an idea?
(11:54:17) Michael_Judd: the problem is getting the right skills to review contributions ....
(11:54:26) teo_sarca: come on. i want that everybody to commit and also to have a stable trunk. it's not posible
(11:54:34) vpj-cd: 2.- to propose a fixed bug is necessary create a patch file
(11:54:37) red1: mike, for example on Accounting rules.. can u organise a team?
(11:54:45) red1: gather more accountants
(11:54:51) CarlosRuiz: remember Adempiere is not modular
(11:54:58) Michael_Judd: I have proposed a team - but we still need a developer to do stuff ...
(11:55:04) CarlosRuiz: we inherited that :-(
(11:55:06) vpj-cd: 3.- To apply this fix bug should have +2 votes
(11:55:15) moyses: I can be an accountant in that team
(11:55:17) Bahman: Sure we don't want everybody comitting Teo...we just want committing be simpler so that we can have new developers.
(11:55:26) CarlosRuiz: hey
(11:55:28) vpj-cd: 4.- the user that give you vote need test to you vote is valid
(11:55:33) CarlosRuiz: new developers <> easy of committing
(11:55:44) Bahman: I know Carlos :-)
(11:55:50) vpj-cd: 5.- this is is valid via test then is simple we apply the patch in trunk
(11:55:52) CarlosRuiz: if a developer can't fill a bug tracker and submit a patch, he's too lazy
(11:55:55) Bahman: I mean easing the experience :-)
(11:55:57) Michael_Judd: I had an issue that I asked everyone about - and no-body else had that problem. No one was selling items ove 999.99 through credit card payment....
(11:55:57) vpj-cd: and fish :-)
(11:56:00) Michael_Judd: so who will review ?
(11:56:25) moyses: I agree with that Michael
(11:56:26) red1: but ppl did
(11:56:28) CarlosRuiz: where you asked?
(11:56:32) red1: ppl tested your code
(11:56:35) red1: and u made it better
(11:56:40) red1: i see that as the process
(11:56:53) red1: he asked me privately after no one answered in the IRC
(11:57:00) CarlosRuiz: IRC is not the site to discuss, is a starting point, but forums is the site
(11:57:03) red1: but Mike did go around and got around
(11:57:15) Michael_Judd: I asked in skype - I asked Joel, I asking in irc twice, I asked the tsang brother when they came to my office in London
(11:57:23) CarlosRuiz: did you ask in forums?
(11:57:24) Michael_Judd: so I feel like i asked everyone .....
(11:57:45) moyses: as long as you don't face the problem you don't show the same concern carlos
(11:57:48) moyses: trust me!
(11:57:53) vpj-cd: if you want you patch go trunk you need find people that test you code
(11:57:59) vpj-cd: the this generate Coesion in comunnity
(11:58:01) Michael_Judd: Don't think I asked in forums because I had asked all of the people actively workign with adempiere
(11:58:10) CarlosRuiz: nope
(11:58:20) CarlosRuiz: all = not me
(11:58:23) moyses: what happened to the problem with the cash journal problem? It is still around
(11:58:25) red1: yes but a post to the forum must be done .. this is WEB2.0 remember?
(11:58:27) Bahman: Disagree Victor!
(11:58:31) Bahman: Look at this: https://sourceforge.net/forum/message.php?msg_id=4375694
(11:58:46) Michael_Judd: the trouble is that Carlos can do both things - and he doesn't understand the frustration of those other poeple who are experts in just one area
(11:58:46) Bahman: The author has a good post and a previous one...
(11:58:47) karsten_thiemann: sorry - got to go - guess I have to read thousands of lines in my log tomorrow :)
(11:58:52) red1: hmm.. i think its better to have code review sessions also
(11:58:57) Bahman: and it's been there in SF for more than 2 weeks...
(11:59:00) red1: why dont we have boot camps
(11:59:04) Bahman: but no one reviewed.
(11:59:13) vpj-cd: yes but then he no make you work
(11:59:31) red1: i tried a swing problem with Bahman... and he managed... got the kick out of it
(11:59:40) vpj-cd: he need explain more people in #adempire
(11:59:46) Michael_Judd: exactly bhman - it happens all the time
(11:59:52) vpj-cd: and try sele you patch fix
(11:59:53) CarlosRuiz: Solution: low the bar
(11:59:53) red1: even Tim got to posting in forums
(12:00:01) CarlosRuiz: push people to review broking Adempiere
(12:00:12) red1: yep
(12:00:12) vpj-cd: if he get the support then get Coesion :-)
(12:00:13) Michael_Judd: make it easier for people to contribute
(12:00:17) Michael_Judd: let them work with groups
(12:00:23) Bahman: As Carlos said: IRC is the starting point...you _have_ to ask in forums Victor....that's what the author did two times.
(12:00:28) red1: been happening
(12:00:29) Michael_Judd: set up a process that supports the group model....
(12:00:32) moyses: Solution: lower down the curve for developing and creating better code for adempiere
(12:00:59) red1: let it be a long curve.. ppl will climb in due time
(12:01:00) vpj-cd: because Tim start with participate now are working in 2pack
(12:01:00) CarlosRuiz: Bahman -> it can mean he didn't know how to ask, or is not from interest of anybody out there
(12:01:04) Michael_Judd: then this reduces the work Carlos needs to do reviewing
(12:01:05) fredtsang_: Have rules!!!
(12:01:13) vpj-cd: because Adempiere and libero give some to he interest
(12:01:18) moyses: yes red1 maybe 2 or 3 years from now
(12:01:33) red1: moyses, we are 3 or 4 years from then!
(12:01:42) vpj-cd: now What is my interest in Tim ok Thim you need work to libero is ready in 2pack
(12:01:48) Bahman: He does know how to ask (in MHO) ... it's about WF...should be of interest of someone in CC Carlos...that's the backbone of ADempiere.
(12:01:52) Michael_Judd: for instance - I have tried to discuss the price lists inc vat (ex vat sales) issue - it effects everyone. i have a solution for UK cutomers ...
(12:02:02) vpj-cd: the I get Coesion
(12:02:05) moyses: but we need more people, more eyeballs, more contributors, we can not just depend on red1, victor, carlos, hensing, et all
(12:02:06) Bahman: If we can't review patches about WF...then why do we have commit rights.
(12:02:19) Michael_Judd: If I ould discuss with a programmer I could put it back in to trunk - but I have to pay to get a programmer becuase noone in the community wants to help
(12:02:19) vpj-cd: to me another gold rule if you give I give :-)
(12:02:32) red1: but we getting more ppl
(12:02:36) vpj-cd: this each to Adempiere comunnity
(12:02:44) Bahman: Are they developers red1?
(12:02:53) red1: u are
(12:02:56) Bahman: Are they walking in teh path to become one?
(12:03:09) Bahman: (I may be slefish) but how many like me?
(12:03:15) CarlosRuiz: -----> AS6: Carlos's point on quality make sense. And certainly Michael_Judd's view are very vital to OSS community. Both make sense. A rigid environment may deter contributors from nearing Adempiere , who may have a high future value towards Adempiere. The question to ask is, how can we have a fair balance between quality and flexibility to submit? There must be way to this, if we could accept the fact that both flexibility and security is importa
(12:03:21) red1: Tim is, AS is, Richard, Marco, ... bad memory
(12:03:28) Michael_Judd: And as Carlos said - not every solution is to write more code..... groups helps to solve this ...
(12:03:35) moyses: red1 if we follow this path then in 9 monts we will have a new bahman!
(12:03:41) fredtsang ha salido de la sala (quit: Connection timed out).
(12:03:54) red1: with a bad back i m afraid
(12:03:57) CarlosRuiz: "can we have a fair balance between quality and flexibility to submit?"
(12:03:59) moyses: thats great!
(12:03:59) Bahman: :-)
(12:04:03) fredtsang_: carlosRuis AS6 point is good
(12:04:17) Bahman: Now coming to point Carlos :-)
(12:04:19) Michael_Judd: I thnk we can get a balance Carlos - but we may need to work at it over time ....
(12:04:29) red1: AS points point to the fact that we have both schools here
(12:04:30) fredtsang_: yes
(12:04:43) moyses: indeed
(12:04:48) fredtsang_: and both solid points
(12:04:52) red1: we have both quality and opennes concerns
(12:04:54) fredtsang_: need to find a way to reconcile
(12:04:57) moyses: we will reach nowhere land :)
(12:05:04) Michael_Judd: So can we organise in some sort of groups and then self regulate the commits ?
(12:05:05) red1: any grey ppl around?
(12:05:05) vpj-cd: I think the community is build in base to network
(12:05:28) Bahman: *Land sighted moyses!*
(12:05:38) fredtsang_: can you explain more your point michael
(12:05:53) fredtsang_: elaborate
(12:05:58) vpj-cd: in my example Victor support to Tim , Tim support to AS6 , AS6 now can support another user
(12:05:58) red1: open source army with no leader .. thats what groups mean!
(12:06:05) Michael_Judd: OK - just one sec - end of the day and my 1800 meeting arrived .... be back in a minute
(12:06:10) moyses: I support a) a QA branch, b) groups between developers and FR, c) make an easy path to become a developer with better rules and documentation
(12:06:11) Bahman: Agree with fredtsang_, I don't get how groups are organised and what's their role.
(12:06:56) fredtsang_: yes what are the steps
(12:06:59) moyses: I guess he refers to FR and developers woking together
(12:07:01) fredtsang_: to become a committer
(12:07:10) fredtsang_: and also how to lose the commit rights
(12:07:24) vpj-cd: the network you know the people and take confidence
(12:07:33) red1: like victor trying to point out, i think its about informal groupings
(12:07:34) vpj-cd: they give more the comunnity
(12:07:53) vpj-cd: the join is with force
(12:07:56) Michael_Judd: do we need to take away commit rights if we get a group workign together - then we only need to take away rights if individuals break the rues
(12:07:59) Michael_Judd: rules
(12:08:09) vpj-cd: and the comunnity is know and solve you need
(12:08:13) Michael_Judd: if the group agrees - then one person can do the commit on behalf of the group
(12:08:26) red1: wiki grew so strong due to its opennes plus open rules
(12:08:31) moyses: not everything can be informal red1
(12:08:33) red1: thats what i read
(12:08:48) red1: well this debate is
(12:08:59) red1: no one stood guard at the door
(12:09:07) Bahman: Mike, all the people in a group have commit rights?
(12:09:07) vpj-cd: I remember you comment Carlos about the how the cell is autofixed
(12:09:14) moyses: the CC is an example that you have been supporting!
(12:09:15) vpj-cd: I think the comunnity is =
(12:09:34) Michael_Judd: it doesn't really matter who has the rights as long as they work with the group
(12:09:35) fredtsang_: the problem with the one person committing for the group
(12:09:41) fredtsang_: is the responsibility he must take
(12:09:43) red1: yes... reverse entrophy
(12:09:45) needle58 ha salido de la sala (quit: "Leaving.").
(12:10:00) Bahman: So it's sort of pair committing Mike? Am I right?
(12:10:05) vpj-cd: entrophy rigth Red1
(12:10:06) red1: yesw the grop is the bigger cell
(12:10:13) moyses: I like Mike idea
(12:10:13) red1: they correct the other cells
(12:10:17) vpj-cd: the I think we only need basic rule
(12:10:28) red1: so lets repeat the basic rules
(12:10:29) moyses: in that way the group must test before commiting
(12:10:31) fredtsang_: yes simple rules
(12:10:38) vpj-cd: and a philosophy
(12:10:39) red1: show up or else
(12:10:55) vpj-cd: ie if you give to Adempiere the Community give you
(12:11:21) vpj-cd: we need build a network with more developers
(12:11:38) red1: we are victor.. just hang on to these cells
(12:11:44) vpj-cd: but this is a process as happened with Tim
(12:11:47) Bahman: Right Victor. That's the question: How new developers are selected.
(12:11:57) moyses: thats important victor
(12:12:02) vpj-cd: is easy to you good work
(12:12:04) moyses: we need to document that process victor!
(12:12:24) teo_sarca: and you will support them
(12:12:29) Bahman: Right moyses! Documentation will make ADempiere continue its life or not! MHO
(12:12:30) fredtsang_: we could have something like that: Any current committer
(12:12:33) vpj-cd: yes Moy the document is part the rule basic :-)
(12:12:33) fredtsang_: can invite anybody
(12:12:38) moyses: since the very begining! how do I set up eclipse, and all those questions that frequently arises on the forums
(12:12:39) fredtsang_: but then he is responsible
(12:12:43) fredtsang_: for whatever work the newbie
(12:12:46) fredtsang_: puts in
(12:12:55) CarlosRuiz: like a mentor?
(12:13:03) fredtsang_: don't know
(12:13:04) vpj-cd: yes a mentor Carlos
(12:13:05) Bahman: Good fredtsang_!
(12:13:06) moyses: exaclty like a mentor!
(12:13:09) Bahman: Good idea.
(12:13:19) vpj-cd: if we have a newbie the will be a metor in future
(12:13:35) fredtsang_: you are responsible for those you invite
(12:13:42) Michael_Judd: but its not a junior and a senior - it is a group of experts
(12:13:48) vpj-cd: then when a newbie have 3 newbie the is convert to metor
(12:13:55) fredtsang_: yes
(12:13:56) vpj-cd: to merit own :-)
(12:14:02) Michael_Judd: some are developers and others are fr - and then there are junios of each type
(12:14:06) usm88 ha salido de la sala (quit: "Leaving").
(12:14:08) red1: the groups also need time to get into being
(12:14:11) Michael_Judd: that should integrate and learn from the groups
(12:14:17) Michael_Judd: like LUG's perhaps ???
(12:14:25) Bahman: subtle point Mike!
(12:14:29) red1: Jug u mean?
(12:14:31) vpj-cd: yes whe you integrate to group
(12:14:38) vpj-cd: the understand the context
(12:14:38) moyses: Linux User Group
(12:14:56) vpj-cd: then you can begin with contribution
(12:15:10) fredtsang_: and check with your mentor
(12:15:14) fredtsang_: before committing
(12:15:23) Bahman: Contributing docs at the first step I suggest.
(12:15:45) red1: so u think this groups can answer the questions of quality assurance?
(12:15:46) moyses: yes documentation is a very important part of this process
(12:15:56) CarlosRuiz: I don't think so
(12:16:08) red1: that means we can pose to them those responsibilty
(12:16:12) moyses: in the long term red it will bring more capables developers who will be able to check the code
(12:16:15) red1: let them be delegated such
(12:16:23) vpj-cd: this newbie have the commit moral to bring another newbie
(12:16:28) red1: i tink if u give a boy a rifle, he might just shoot u
(12:16:30) Bahman: Agree with moyses...can't rely on a few.
(12:16:34) moyses: that means that as the times pass they will be able to make better commitments!
(12:16:41) red1: unless we teach him a point or two
(12:16:42) vpj-cd: the the network the community increase :-)
(12:16:47) CarlosRuiz: ok
(12:16:56) Bahman: Come on red1! We're not talking about rifles :-)
(12:17:00) moyses: so the need to check all those commits will be less
(12:17:00) CarlosRuiz: it sounds to me like layers of committers
(12:17:07) red1: sorry bahman.. getting tired here
(12:17:09) moyses: red1 you are totally lost today!
(12:17:12) Bahman: The shoot can be simply healed (reverted) :-)
(12:17:13) CarlosRuiz: first layer take care of second layer ...
(12:17:13) moyses: :)
(12:17:33) CarlosRuiz: second layer of third layer ... and so on
(12:17:42) red1: my conclusion is more of u must come forth strongly in support of those in front doing all the hard work
(12:17:44) fredtsang_: like the army
(12:17:46) fredtsang_: general
(12:17:48) moyses: indeed the network will grow up as victor pointed out
(12:17:48) fredtsang_: caporal
(12:17:50) fredtsang_: soldier
(12:18:03) fredtsang_: u only have to reports within the same rank
(12:18:03) CarlosRuiz: it sounds better for me
(12:18:07) Bahman: This looks more fit to a pirate ship ;-)
(12:18:21) red1: thats the spirit lads
(12:18:23) CarlosRuiz: the current list of 10 committers can be organized in two levels
(12:18:23) fredtsang_: :)
(12:18:36) vpj-cd: yes
(12:18:41) red1: yes but we expect that 10 to double soon
(12:18:46) red1: since we re not having rifles
(12:18:47) CarlosRuiz: yes but in layers
(12:18:59) fredtsang_: and whoever the invite
(12:19:00) red1: aye aye sir
(12:19:01) Michael_Judd: I think we aregetting very narrow again - why should people be restricted if they are just administering on behalf of a group and following the rules ?
(12:19:08) fredtsang_: they must be responsible for their code
(12:19:14) vpj-cd: CorlosRuiz the levet can be depend the if touch core
(12:19:17) Michael_Judd: I agree fred ...
(12:19:19) CarlosRuiz: it sounds good, thanks Fred!!!!
(12:19:26) red1: i think many wouldnt even care about others codes
(12:19:35) red1: look at Trifon.. he does his ADCK mostly
(12:19:53) red1: and the translators .. they arent affected i think
(12:19:57) moyses: Agree fredtsang_
(12:20:20) red1: but there are those who does care about the rest
(12:20:25) red1: Carlos and Hengsin.. etc...
(12:20:32) red1: and they need help
(12:20:43) CarlosRuiz: what I see is this
(12:20:50) red1: is that what u mean Carlos?
(12:20:59) CarlosRuiz: a first level of committers, i.e. : Teo, Hengsin, me
(12:21:05) CarlosRuiz: a second level of committers
(12:21:26) red1: there must be central level overall app wide caretakers... thats what i understand from the 1st level
(12:21:26) CarlosRuiz: Teo take the responsibility of review the commits from 3 people
(12:21:30) CarlosRuiz: Hengsin other 3
(12:21:32) moyses: I like the layers idea...
(12:21:34) CarlosRuiz: me other 3
(12:21:48) fredtsang_: for it to work we must have access to the one above and they must make time
(12:21:49) CarlosRuiz: so we have a second layer of 9 people
(12:21:55) red1: it has to come together somewhere
(12:22:00) Michael_Judd: still not sure that will work - what happened to the groups idea? o
(12:22:10) red1: groups below them i guess.. ok?
(12:22:15) fredtsang_: yes
(12:22:18) red1: lets try out
(12:22:18) fredtsang_: michael
(12:22:19) CarlosRuiz: third layer can be 27 people
(12:22:21) moyses: we may need to define how can you become a first or second layer commiter :)
(12:22:30) red1: now thats viral
(12:22:35) moyses: or even a third layer
(12:22:38) fredtsang_: it's like you have a leader in that team
(12:22:40) vpj-cd: yes a new member is add to invite
(12:22:48) vpj-cd: as google mail :-)
(12:22:59) red1: higher layer means u can afford to take care of more
(12:22:59) Bahman: Each member can be promoted to a higher level through a positive vote from all the immediate higher layer members.
(12:23:09) vpj-cd: yes red1 one is viral a network
(12:23:10) Michael_Judd: so who do I talk to when I need something done from a developer perspective ? How do these propesed layers work?
(12:23:13) CarlosRuiz: it sounds good for me, what do you think Hengsin, Teo ?
(12:23:27) CarlosRuiz: Michael -> forums
(12:23:31) jsSolutions ha salido de la sala (quit: "hasta la vista").
(12:23:35) vpj-cd: this is force some similar that happened with baazar
(12:23:36) Michael_Judd: Carlos - bit laugh in our office at your last comment !
(12:23:36) fredtsang_: in michael case
(12:23:37) fredtsang_: i would say
(12:23:51) fredtsang_: they need one on their own
(12:23:54) fredtsang_: to be independent
(12:24:00) red1: mike.. u mean i dont get the laughs nowadays?
(12:24:04) moyses: yes fred is right!
(12:24:05) Michael_Judd: And what happens when there are not takers in forums - or newbies offer inappropraite comments and Carlos is too busy ?
(12:24:26) CarlosRuiz: forums are not newbies
(12:24:28) CarlosRuiz: forums is everyone
(12:24:29) Bahman: What's your suggestio Mike?
(12:24:32) fredtsang_: they must make time for you
(12:24:35) fredtsang_: or promote you
(12:24:46) red1: JJ knew all too well.. he gave up
(12:24:53) CarlosRuiz: Michael, I think you can't have a committed developer as you want
(12:24:56) CarlosRuiz: not without payment
(12:24:56) moyses: good question Michael what can we do about it?
(12:25:05) Michael_Judd: My suggestion is to have a group where we people can organise themselves by skills and commit when they are ready ....
(12:25:14) Michael_Judd: rather than having a cathedral ....
(12:25:28) Bahman: Hmm...
(12:25:29) CarlosRuiz: does it look like a cathedral?
(12:25:33) vpj-cd: yes but a newbies have to give some the community
(12:25:36) CarlosRuiz: it looks like a viral pyramid
(12:25:45) moyses: I believe that both approaches can coexist!
(12:25:49) CarlosRuiz: I like the idea from Fred, it can become viral
(12:25:50) red1: so CarlosRuiz u agree with Mike's idea of groups somewhere along the layers?
(12:25:53) CarlosRuiz: and we can have developers mentored
(12:25:54) vpj-cd: yes Carlos is a network
(12:25:59) CarlosRuiz: no
(12:26:05) CarlosRuiz: red1 -> no
(12:26:09) moyses: we may need the expert group about Accounting, Sales, etc
(12:26:10) CarlosRuiz: I think groups don't work
(12:26:14) CarlosRuiz: layers can work
(12:26:17) Michael_Judd: it looks just like my old university department
(12:26:23) red1: lol
(12:26:39) ***red1 laughs alone cos everyone is sleeping here
(12:26:44) Michael_Judd: there are the really old timers who control everything - and then young lecturers who have influence and then students who just pay money ...
(12:26:58) red1: no wonder i didnt go to the university
(12:27:01) CarlosRuiz: it can sound ugly but you can't have my commitment without payment on specific issue
(12:27:02) vpj-cd: or logia mason :-)
(12:27:04) vpj-cd: jejejej
(12:27:09) Bahman: Mike's idea can be used this way: No more layers after layer 2...groups are organised by members of layer 2.
(12:27:25) red1: so u agree with Bahman?
(12:27:26) moyses: Agree with Bahman
(12:27:35) CarlosRuiz: it's like the post Bahman pointed, nobody has the obligation to answer
(12:27:48) teo_sarca: true
(12:27:49) dladwig_Idalica ha salido de la sala (quit: "Trillian (http://www.ceruleanstudios.com").
(12:27:49) CarlosRuiz: I don't agree
(12:27:51) Michael_Judd: hence nothing happens and the problem persists....
(12:28:03) CarlosRuiz: yes, Michael, unless you solve it
(12:28:07) CarlosRuiz: or pay someone to
(12:28:12) red1: perhaps Carlos started coaching more layers in the form of groups
(12:28:17) red1: and delegate to them
(12:28:19) CarlosRuiz: looke
(12:28:22) red1: we will support
(12:28:25) CarlosRuiz: I don't agree with second layer too big
(12:28:26) Michael_Judd: If I posted some expert nd detailed advice about how something fundamental doesnt work in adempiere - what would be the public response (with end users)?
(12:28:32) CarlosRuiz: because what I like from this approach
(12:28:36) CarlosRuiz: is that I can cope with 3 person
(12:28:40) CarlosRuiz: but I can't cope with 20 person
(12:28:53) hengsin: there is another issue here that we miss
(12:28:53) Michael_Judd: you are suggesting though that if I solve something myself - that you won't allow it in to the trunk ...
(12:29:03) Bahman: 2nd layer is composed of 9 members.
(12:29:05) Michael_Judd: I thought pyramid schemes ended i nthe '80's
(12:29:10) CarlosRuiz: Am I saying that we don't allow into the trunk
(12:29:19) CarlosRuiz: I'm saying that you can put into the trunk following a process
(12:29:26) Michael_Judd: We calledthem bottom of the harbour schemes in Australia (that's Sydney harbour)
(12:29:29) Bahman: _for example_ 9.
(12:29:47) hengsin: the commitment from people making contribution, enhancement, etc
(12:29:57) moyses: we will be like the girls selling avon products :)
(12:29:58) red1: i think its a small pyramid scheme
(12:30:07) CarlosRuiz: hahahaha, but Avon pyramid works
(12:30:12) red1: well we have to support whatever scheme
(12:30:14) CarlosRuiz: Amway
(12:30:16) teo_sarca: good one moyses :)))
(12:30:22) vpj-cd: yes Mike the network is preset toady
(12:30:25) Bahman: Yes...we are not totally in general-colonel-major-.... form...
(12:30:30) vpj-cd: en every place
(12:30:34) red1: i tink its a matter of perception
(12:30:35) Bahman: we just proposing some king of QA..
(12:30:41) CarlosRuiz: look
(12:30:43) hengsin: I'm ok with the group concept, if the group took full responsibility on their contribution
(12:30:44) red1: carlos is seeing his issues from up there
(12:30:47) Bahman: which may simply be removed in future (if required)...
(12:30:50) CarlosRuiz: initially the first and second layer can have commit permissions
(12:31:00) CarlosRuiz: then when third layer becomes strong
(12:31:01) red1: we re seeing from somewhere not so up
(12:31:07) CarlosRuiz: we give third layer commit permissions ...
(12:31:10) CarlosRuiz: it could be?
(12:31:49) red1: let me ask one stupid question, who can make our version releases, if Carlos or Hengsin is off?
(12:31:52) teo_sarca: what will be the scope of commiting ?
(12:31:52) fredtsang_: i think all the pyramid is saying is that
(12:31:56) fredtsang_: a trusted b
(12:31:58) hengsin: as for individual, without any recommendation from other committer, I still think the individual must start with submitting patches
(12:32:01) fredtsang_: so if b trust c now
(12:32:05) fredtsang_: then a is fine with c
(12:32:08) red1: perhaps we just need a hi level release comittee
(12:32:09) fredtsang_: and so on
(12:32:15) fredtsang_: simpler to manage
(12:32:24) red1: the rest been groups
(12:32:34) Bahman: Good point Teo...if the top layer should review all the commits again then this military hierarchy is non-sense, IMHO.
(12:32:44) fredtsang_: yeah
(12:32:47) CarlosRuiz: yes Hengsin, it must always be the start
(12:32:47) Michael_Judd: ah ha - the cathedral emerges ....
(12:32:49) fredtsang_: bahman
(12:32:59) red1: there must be someone hi up to commit finally
(12:33:06) red1: thus some priests
(12:33:11) teo_sarca: and one more point, the high layer won't do just review stuff, will commit also (BF, FR etc)
(12:33:14) Michael_Judd: and to think 2 years ago you all disapproved of doing this ?
(12:33:26) teo_sarca: so practically, the thinks are mixed
(12:33:36) red1: but there is a low level groupings like mike said that will assist in SME areas
(12:33:40) CarlosRuiz: I said, initially first two layers have commit permission, and in time third layer will have
(12:33:57) Bahman: I think layer 2 should be federal...
(12:34:05) red1: i read Animal Farm, mike
(12:34:13) red1: more equal than others
(12:34:17) hengsin: 2 year ago, I'm not involve with compiere :)
(12:34:20) Bahman: what good is breaking the community into layers while the top layers have to review all commits again!
(12:34:22) teo_sarca: orwell :)
(12:34:30) CarlosRuiz: no
(12:34:46) red1: 3 years ago there also only Victor, Mike and red1 :>
(12:34:52) CarlosRuiz: I'll review commits just from my three mentored
(12:34:55) CarlosRuiz: that's the difference
(12:35:03) CarlosRuiz: I can't guarantee you that I review all commits
(12:35:07) red1: (joking)
(12:35:15) CarlosRuiz: but I can guarantee you that I review commits from my 3 mentored 2nd layer
(12:35:26) red1: bahman.. its only reviewing.. not shooting!
(12:35:31) Bahman: So you're facing 20/3 commits per day, right?
(12:35:44) Bahman: I know red1! I mean it doesn't help the top layer...
(12:35:47) red1: the top has to delegate somehow
(12:35:51) CarlosRuiz: yes, but I'll have strong communication with my 3 mentored group
(12:35:56) red1: and we have to help them
(12:35:57) Bahman: the pressure will be on again,
(12:36:04) red1: we are all in the same boat
(12:36:11) moyses: agree with bahman
(12:36:12) Michael_Judd: so what's wrong with as Bahman says - policing the commit rules and facilitating the development groups .... helpng to get more people in to the project
(12:36:26) Bahman: That's what I'm talking about red1!
(12:36:27) red1: thats why i m learning up posterita...
(12:36:28) Michael_Judd: make it easer to geet new skills rather than trying to make it harder to get now skills ?
(12:36:39) CarlosRuiz: again the same wrong point -> having more people doesn't mean allowing everyone to commit
(12:36:40) vpj-cd: I am in Compiere from 2001 - 2002 :-)
(12:36:50) red1: we have to accomodate both left and right here
(12:36:53) vpj-cd: when JJ have compiere no exist in sf :-)
(12:37:06) CarlosRuiz: in fact the layer can guarantee a mentored training for newbies
(12:37:20) Bahman: Yes Carlos...
(12:37:29) vpj-cd: yes Carlos
(12:37:31) Bahman: But groups should act federally.
(12:37:32) red1: i think almost everyone here are in the layers.. rite CarlosRuiz ?
(12:37:37) vpj-cd: and quality
(12:37:46) CarlosRuiz: I don't understand "federally"
(12:37:59) Michael_Judd: so who is going to be at the top of the pyramid and egulate everyone else ?
(12:38:00) red1: a central command structure?
(12:38:11) Michael_Judd: regulate ...
(12:38:13) red1: i think regulate is too strong a word
(12:38:20) red1: coaching sounds more fairer
(12:38:20) Bahman: I mean there should not be a top layer slowing down the growth process.
(12:38:27) red1: precisely
(12:38:31) CarlosRuiz: Bahman, second layer can commit
(12:38:35) Michael_Judd: I think it is exactly the right word from what I have been watching ....
(12:38:37) CarlosRuiz: in future third layer can commit
(12:38:41) Michael_Judd: i.e. who do i have to pay ?
(12:38:43) jsSolutions [email@example.com] ha entrado en la sala.
(12:38:49) Bahman: but top layer will review Carlos, am I right?
(12:38:53) AS6: Carlos's suggestion sounds practical and is a reasonable measure to control on quality issues. And the approach is also giving mechanism for contributors to contribute.
(12:38:56) Michael_Judd: Will it be one of those people I tried to pay but didn't have time ?
(12:39:17) moyses ha salido de la sala (quit: "ChatZilla 0.9.78.1 [Firefox 188.8.131.52/2007051502]").
(12:39:22) red1: i think if anyone new come along for that matter, he can always have a word and he is in..
(12:39:33) CarlosRuiz: top layer review? possibly, it depends on the trust that I have on second layer ;-)
(12:39:54) Bahman: So what good is this layer story to you, hengsin and Teo?
(12:39:56) Michael_Judd: What if we dont trust the top layer - if the root of trust is not trusted ?
(12:40:07) nwessel_2 [firstname.lastname@example.org] ha entrado en la sala.
(12:40:09) CarlosRuiz: is this the case?
(12:40:11) Bahman: And what difference does it make to current situation?
(12:40:13) red1: so far no such case
(12:40:33) red1: no one got kicked out
(12:40:37) Michael_Judd: Personally - no - I'm not being personal. But there may be conflicts of interest ....
(12:40:49) hengsin: sounds a bit too complicated and demanding for me :)
(12:40:58) teo_sarca: for me too
(12:41:01) red1: u have to give a case
(12:41:12) Bahman: Carlos, all I'm trying to say is that, how this hirerarchical structure help ADempiere growth?
(12:41:13) CarlosRuiz: What I would try to get is a commitment of review/mentor people from few committers
(12:41:15) AS6: Carlos being at the top is simply business. Its nothing personal. However, if he conflicts, then the community will be "reprimanding" him.
(12:41:18) Michael_Judd: Whos work will layer 1 dal with first?
(12:41:24) Bahman: Everyone, I don't get the point.
(12:41:34) CarlosRuiz: I don't understand the point
(12:41:49) CarlosRuiz: what can be the "conflict" ?
(12:42:01) hengsin: I think we must not rush to put thing into the core/trunk, that wouldn't produce quality product.
(12:42:14) teo_sarca: agree
(12:42:20) Michael_Judd: Well who does layer 1 deal with first ? They have two things to commit - both very large - who do they deal with first?
(12:42:20) AS6: Agree with that.
(12:42:29) Michael_Judd: ANSWER: who every pays ....
(12:42:49) red1: can the small goups manage themselves on that?
(12:42:54) CarlosRuiz: hey, rule of voting is not dropped
(12:42:59) Michael_Judd: We are making things more difficult and not solving the real problem
(12:43:05) CarlosRuiz: we still haven't dropped any rule to commit
(12:43:12) red1: i think money has little effect here
(12:43:13) Michael_Judd: How do we get more people involved in the project and how do we help them to be effective ...
(12:43:17) CarlosRuiz: to allow functionality you must ask openly in forums and vote
(12:43:26) hengsin: for e.g, the workflow example Bahman mention earlier, I does take a look at it, it touches many place of the core thus need a carefull and detail review. so it tooks longer to goes into trunk ...
(12:43:28) Michael_Judd: If people can be more effective - carlos has to do less review as he can tust the groups
(12:43:51) Michael_Judd: The groups work is dispalyed by individuals commiting
(12:43:55) CarlosRuiz: that's what I'm trying to say Michael
(12:44:02) Bahman: That's what I was trying to say Mike!
(12:44:05) Bahman: Ah! :-)
(12:44:13) Michael_Judd: If an indivual breaks the trust of rht group - or the commit rules they can be disciplined ...
(12:44:21) CarlosRuiz: back to the roots:
(12:44:21) CarlosRuiz: 2 - Commit committee can't cope with the review of all commits
(12:44:29) CarlosRuiz: we're here trying to solve this
(12:44:30) Michael_Judd: Why do we have to go around making a cathedral and resticting thing ???
(12:44:37) CarlosRuiz: what's the cathedral point?
(12:44:41) CarlosRuiz: I don't see a cathedral here
(12:44:41) teo_sarca: Michael: have you faced with the situation when the core is broken, but also there are good bugfixes too, you want to deploy the trunk but you have to elimintate the issues. So basically if everybody can commit your work will be twice also.
(12:44:46) CarlosRuiz: I don't see restricting here?
(12:44:54) red1: a cathedral is over there > compiere
(12:45:00) Michael_Judd: Blocking commit rights is not the way to encourage this - making people responsible for their actions is ....
(12:45:02) hengsin: for group work, I would think that we need to do a high level review to ascertain the risk of their work
(12:45:03) CarlosRuiz: I see a process
(12:45:05) red1: not here mike :>
(12:45:19) red1: maybe some stonehenge
(12:45:23) CarlosRuiz: Michael, again, nobody is paid here, you can't make people responsible
(12:45:32) Michael_Judd: well looks like we have some ruble and the catherdral is build build - I'm trying to keep the bazzar ..
(12:45:35) moyses [email@example.com] ha entrado en la sala.
(12:45:39) CarlosRuiz: I have seen people broking trunk and forgetting about that
(12:45:43) hengsin: if it is of high risk, it should be treated just like individual high risk work, start in branch first.
(12:46:01) Michael_Judd: well if you break and forget that is one thing - that is not on ....
(12:46:05) CarlosRuiz: hengsin, teo_sarca, what's the issue you said
(12:46:06) Bahman: Hmm...A good point from Carlos: The amount of reviews is related to trust between a top layer member and his mentor.
(12:46:14) red1: perhaps the unpleansant work of exposing the breaks has to be done also somehow
(12:46:19) Michael_Judd: and why does that happen - is there a clear testing process for people to undertake before committing ?
(12:46:21) red1: to educate
(12:46:22) hengsin: also, the group must give their commitment to maintain their contribution.
(12:46:23) Bahman: So, it's their job to review...
(12:46:28) Michael_Judd: have we done our jobs to make that process clear and visible ?
(12:46:33) Bahman: And groups exist at layer 2.
(12:46:37) red1: ask ourselves
(12:46:38) CarlosRuiz: is that an excuse Michael?
(12:46:45) vpj-cd: yes example I ask to Teo help me to
(12:46:51) Michael_Judd: If a user rings for support too much - then I spend some time training and they stop ringing ....
(12:47:01) vpj-cd: improve collapse tab :-)
(12:47:33) Michael_Judd: Im not making excuses - Im trying to make it easier for people to be involved and to improve quality
(12:47:51) Michael_Judd: the way to improve waulity is to get more people contibuting - not less ...
(12:48:12) CarlosRuiz: nope
(12:48:15) red1: we have to look outward... getting more ppl
(12:48:15) Michael_Judd: you need to make a way where people can contribute and they can also get help
(12:48:18) AS6: Michael: Carlos's latest suggestion DOESNT prevent people from contributing.
(12:48:25) red1: within we can have small rules
(12:48:29) red1: small govt big bazaar
(12:48:29) CarlosRuiz: finally!!!
(12:48:42) Michael_Judd: Sorry - I thought one of the actions was to drop some poeple commit rights ....
(12:48:47) CarlosRuiz: it's tricky if you think that preventing bad commits prevent people from contribute
(12:48:56) CarlosRuiz: just prevent bad commits to enter in trunk!
(12:48:56) Michael_Judd: and that increases the 'cost' of commit ....
(12:49:12) Bahman: I'd suggest the 2 layers idea but no more layers...everything is organised inside the 2nd layer and top layer acts mostly as the consultant to layer 2...not the governer to layer 2.
(12:49:20) CarlosRuiz: what is the 'cost' of commit vs the 'cost' of fix a bad commit ??
(12:49:33) Michael_Judd: what is the cost of NOT getting the input ....
(12:49:39) teo_sarca: for me bad commits increase the costs too
(12:49:57) CarlosRuiz: is better not to have a functionality that to have a bad functionality
(12:50:00) Bahman: Not getting inputs is a tragedy.
(12:50:08) AS6: We will get inputs as long as we have mechanisms for receiving inputs.
(12:50:09) Michael_Judd: I agree - but is there anyone who consistently badly commits - who doesn't fix their problems if asked ?
(12:50:27) AS6: And carlos's suggestion is allowing for inputs.
(12:50:31) red1: its clear that there is no evil hand in the committing
(12:50:40) Michael_Judd: What makes one persons contributions less valid just because they can't do the last 10% of code ......
(12:50:41) hengsin: Michael, we don't want to get personal here but there are.
(12:50:49) moyses: no Carlos none is better
(12:51:08) moyses: you have been talking about quality none is better
(12:51:20) red1: middle road?
(12:51:28) teo_sarca: i think we need branches !!!
(12:51:32) CarlosRuiz: ok
(12:51:38) AS6: agree to sarca
(12:51:40) CarlosRuiz: can we rethink the layer proposal?
(12:51:42) Michael_Judd: philisophical difference ........ embrace to community to restrict access ....
(12:51:43) moyses: we need branches teo I second you!
(12:51:51) Bahman: Teo...how are the branches are to be merged?
(12:51:52) CarlosRuiz: what's the issue hengsin and teo have about?
(12:51:59) teo_sarca: but this will slow the dev. too :)
(12:52:11) hengsin: Michael, it is not about restricting access, it is about earning the trust of the community.
(12:52:31) teo_sarca: "what's the issue hengsin and teo have about?"
(12:52:31) teo_sarca: what you mean ?
(12:52:49) Michael_Judd: for me it is another problem all together - developing code is only a small part of this project
(12:52:51) Bahman: The community will die without fresh forces hengsin.
(12:53:04) vpj-cd: yes the branch is to enhanced medium and big
(12:53:07) Bahman: Whatever force. business expert or developer.
(12:53:09) CarlosRuiz: teo, you and hengsin said that layer approach is too demanding
(12:53:56) CarlosRuiz: IMHO, layer approach can bring more contributors, and mentored newbies will learn faster
(12:54:25) Michael_Judd: why dont you lay out a proposal in the wiki and the community can discuss ?
(12:54:26) teo_sarca: and everybody will commit anywhere in the trunk?
(12:54:39) Bahman: Agree with the mentor idea...as it trains fresh soldiers for the ship but can't agree with the top layer governing layer 2.
(12:54:58) red1: why dont we try the prototpe approach? Lets think about all these ideas, do our effort in making more things work and prove the concept...and get back here in few days to share our findings?
(12:55:03) teo_sarca: basically a proposal will be great because we are running in circles here :)
(12:55:09) Bahman: Teo, you said we need branches. How the branches are to be merged?
(12:55:10) hengsin: I don't think we can commit to always review everything, I prefer a model where people earn the commit right through patches or recommendation.
(12:55:34) teo_sarca: Bahman, we need to define by who, when and how
(12:55:51) CarlosRuiz: another 2 months to solve this? like Adempiere foundation?
(12:55:56) teo_sarca: but as i see there is a high requirement for a svn playground :)
(12:56:14) hengsin: teo: branch is for that
(12:56:22) CarlosRuiz: hengsin, the idea is if you can commit to always review commits from just 3 persons
(12:56:24) Bahman: I guess the "who" = "top layer"...sorry if I play with "top layer" too much,,,
(12:56:27) CarlosRuiz: not all commits
(12:56:39) Bahman: but don't want ADempiere tied up to ideas of a few...
(12:56:50) Bahman: no matter how that few help ADempiere grow...
(12:56:51) hengsin: sorry, really can't make that sort of commitment.
(12:56:55) red1: for example Victor gave a tab prototype idea, then Hengsin did... then its a matter of proving that it works further and further... any better idea?
(12:56:58) Bahman: Thans to those few :-)
(12:57:08) red1: bahman, u are in the top layer
(12:57:14) CarlosRuiz: ok, hengsin, what about not reviewing, but mentoring?
(12:57:26) Bahman: Aha! Good point Carlos!
(12:57:29) red1: yes mentoring is a better word
(12:57:38) red1: mentor can slap students.,.. not shoot them
(12:57:39) Bahman: Being a consultant to layer 2.
(12:57:47) fredtsang_: mb the word 'layers' is confusing
(12:57:50) fredtsang_: what about friends
(12:57:51) hengsin: mentoring is ok, but no hard number please, it have to depends on my own work schedule.
(12:58:09) CarlosRuiz: ok, it's getting better
(12:58:15) fredtsang_: so you invite friends
(12:58:16) red1: certainly not BPartners
(12:58:17) fredtsang_: to a house
(12:58:24) fredtsang_: of a friend
(12:58:30) red1: and Mike buys the beers and coke
(12:58:31) fredtsang_: so you make sure your friends behave
(12:58:31) Bahman: Camp 1 and Camp 2?
(12:58:37) fredtsang_: and respect the rules of the house
(12:58:42) fredtsang_: once you happy they know the rules
(12:58:46) fredtsang_: and behave you let them be
(12:58:48) CarlosRuiz: layer 1 can grow with people from layer 2 becoming experts
(12:58:54) fredtsang_: but everybody knows you invited them
(12:58:59) red1: and friends trust each other
(12:59:02) fredtsang_: so you will be embarassed
(12:59:08) fredtsang_: if they start to misbehave
(12:59:13) Bahman: Very important point Carlos!
(12:59:31) CarlosRuiz: again, just layer 1 and 2 have commit rights
(12:59:39) fredtsang_: I don't think it's a cathedral
(12:59:45) fredtsang_: but more of an initiation
(12:59:47) red1: its a lodge
(12:59:49) CarlosRuiz: but layer 1 can grow, so layer 2 can grow with layer 1
(12:59:51) fredtsang_: and watching over
(12:59:54) red1: (joking)
(12:59:54) hengsin: off topic here but I just find it funny people say submitting patch is difficult.
(12:59:55) Bahman: How you see layer/camp/house 2 Carlos?
(12:59:57) fredtsang_: your small group
(13:00:12) fredtsang_: and no one can impose friends on you
(13:00:14) fredtsang_: you choose them
(13:00:17) CarlosRuiz: don't get the point Bahman
(13:00:21) red1: my first work to JJ was a patch.. it was hard not hearing from him for 6 mths
(13:00:22) fredtsang_: specially if they are helping you out
(13:00:24) fredtsang_: on some stuff
(13:00:31) Bahman: I mean who are them?
(13:00:46) CarlosRuiz: ah, ok, we can vote on layer 1
(13:00:49) CarlosRuiz: to make it demochratic
(13:00:55) CarlosRuiz: then we can vote on layer 2
(13:00:57) red1: fredtsang_, is pointing to the infromal bazaar way... which is good
(13:00:58) Bahman: Good fredtsang_...it's getting clearer.
(13:01:44) CarlosRuiz: hengsin, teo, can you commit to mentor 3 persons each one?
(13:01:55) fredtsang_: hengsin
(13:02:04) fredtsang_: we can propose you sendy and ashley
(13:02:06) fredtsang_: ?
(13:02:13) teo_sarca: yes, i can mentor 3 persons
(13:02:18) fredtsang_: since you working with sendy already
(13:02:22) fredtsang_: make sense right?
(13:02:28) Bahman: Looks like the result of the elections is pre-assumed ;-)
(13:02:28) red1: sounds good
(13:02:30) hengsin: ok.
(13:02:48) CarlosRuiz: we can ask later Trifon if he can be in layer 1 too mentoring other 3 persons
(13:02:49) red1: its still the cave age for us Bahman.. pretty advanced
(13:02:53) hengsin: I doubt I need to mentor ashley thought :)
(13:03:03) red1: ashley shuld be a mentor
(13:03:03) hengsin: maybe the other way round.
(13:03:13) CarlosRuiz: Layer 1 -> Hengsin, Teo
(13:03:21) CarlosRuiz: any problem if I propose myself in layer 1 ;-)
(13:03:28) Bahman: Hmm..
(13:03:34) fredtsang_: ok he would be happy to hear that
(13:03:35) red1: can bahman be my source mentor?
(13:03:36) Bahman: I see no problem :-)
(13:03:56) Bahman: Let me check...
(13:03:57) CarlosRuiz: Layer 1 -> Hengsin, Teo, CarlosRuiz (we must ask later to Trifon)
(13:04:08) Bahman: How many participants present we have here?
(13:04:17) Michael_Judd: me still - but relucantly ...
(13:04:20) Bahman: Who aren't sleeping now?
(13:04:20) moyses: ummm Carlos Ruiz in layer 1?? not a good idea :)
(13:04:44) red1: anyway i got my little path.. posterita under shameem... guess ashley is behind that.. and hengsin here for advice..
(13:04:45) teo_sarca: not sleepy, just hungry :)
(13:04:49) Bahman: We are very few I guess for an election...
(13:04:58) Bahman: better arrange it by email.
(13:05:08) red1: it has to be a mutual and evolving thing
(13:05:23) CarlosRuiz: any proposal for layer 2?
(13:05:45) CarlosRuiz: Bahman, Karsten, Johannes, Phib, Ashley, Fer_luck, Kontro, Tim, Vishee, Sendy, Shameem ??
(13:05:55) Bahman: Many of us are absent here....I'd strongly suggest postposning election to a later time :-)
(13:06:06) red1: i think its just suggestion bahman
(13:06:09) CarlosRuiz: it's not an election
(13:06:13) moyses: we must use the forums :)
(13:06:14) red1: we have to put it down somehow
(13:06:24) Bahman: Alright then.
(13:06:28) fredtsang_: agree red1
(13:06:34) CarlosRuiz: maybe I would propose that layer 1 select the layer 2
(13:06:38) CarlosRuiz: between friends
(13:06:41) red1: which layer am i now?
(13:06:42) moyses: haha
(13:06:48) red1: or am i the little group?
(13:06:53) moyses: Carlos Ruiz and his friends in both layers
(13:06:53) red1: the cosy lodge?
(13:06:56) moyses: ha ha
(13:07:01) Bahman: moyses LOL
(13:07:03) red1: back to the ship?
(13:07:21) CarlosRuiz: what do you think?
(13:07:26) Bahman: So what do you think>
(13:07:27) Michael_Judd: mre lauchs from over here moyses
(13:07:28) moyses: we must put this in the forums
(13:07:32) Bahman: You'r faster Carlos :-)
(13:07:35) CarlosRuiz: layer 1 propose the layer 2 ??
(13:07:38) Michael_Judd: more laughs ;)
(13:07:49) moyses: it reminds me something like Fidel Castro in Cuba
(13:07:51) moyses: hehe!
(13:08:01) red1: i like Fidel.. before he went up..
(13:08:02) teo_sarca: where are you from moyses ?
(13:08:14) red1: not Cuba.. nearby there
(13:08:15) moyses: they have something called CDR
(13:08:22) Bahman: I'd suggest publishing the proposal and then voting for the way to elect layer 2.
(13:08:31) moyses: and in the end they just elect once again Casto
(13:08:37) moyses: I am from Mexico Teo
(13:08:38) Bahman: It should be first in Wiki I guess.
(13:08:50) red1: i think forum is more suited
(13:08:55) fredtsang_: who is volunteering to write it up?
(13:09:00) red1: cos many can easily track via emails
(13:09:02) CarlosRuiz: anybody want to propose someone else for layer 1 ??
(13:09:15) Bahman: I'll write a draft...
(13:09:16) Michael_Judd: no I dont agree so I'm not nominating anyone
(13:09:20) red1: 2am i cant say i want to write more
(13:09:28) moyses: I don-t agree either
(13:09:29) fredtsang_: cool bahman
(13:09:30) Bahman: but that needs editing.
(13:09:42) CarlosRuiz: Michael, can I ask you what you don't agree? the approach, or the people?
(13:09:42) moyses: Bahman will write a draft
(13:09:45) moyses: I agree with that
(13:09:45) Michael_Judd: so anyone else not agree ?
(13:09:54) red1: what do we do with the descending votes? :>
(13:10:00) red1: discenting
(13:10:20) Bahman: Mike, I think we should do the correction in an iterative way...
(13:10:25) Michael_Judd: this will stop people contributing - so I am against it ...
(13:10:36) Bahman: I don't think we can solve all out problems tonitgh...
(13:10:37) red1: i like the group idea for functionals and the top layer for mentoring and final say on commit and releases
(13:10:38) moyses: yes I agree with Bahman suggestion
(13:10:55) moyses: nothing should be decided now!
(13:11:00) Bahman: sure it has limitations but let another meeting 3 moths later solve some of its limitations.
(13:11:07) moyses: because there is a lot of people missing this discussion
(13:11:14) CarlosRuiz: hehehe, it doesn't need to be said
(13:11:17) fredtsang_: it might not be best solution, but is it an improvement over the current structure?
(13:11:22) red1: rather extreme view from England here, but can we allow some time to try?
(13:11:24) CarlosRuiz: we're always open to change things that are not working
(13:11:24) Bahman: Yes fredtsang_.
(13:11:27) Michael_Judd: we are walking down the wrong path - why walk further and then have to walk back ?
(13:11:29) Bahman: I hope so :-)
(13:11:39) red1: we have to learn from mistakes i guess
(13:11:54) red1: any latin version of that bahman?
(13:11:58) nwessel_2: walking slowly is ok as long as we walk at all
(13:12:03) Bahman: Of what?
(13:12:21) nwessel_2: I do not mind walking the wrong direction to find out that t was the wrong one
(13:12:29) red1: no love lost
(13:12:43) red1: its only a bazaar.. but a serious one so far
(13:12:52) red1: and an open one
(13:12:58) red1: we respect all views
(13:13:05) red1: but decisions has to be made
(13:13:11) nwessel_2: michael, are you afraid that we can not roll it back if it does not work?
(13:13:17) Bahman: "Cul peccare licet peccat minus" - One who is allowed to sin, sins less.
(13:13:17) Michael_Judd: YES
(13:13:37) red1: i m sure we call roll back
(13:13:43) red1: its called the Revert button
(13:13:44) nwessel_2: what do you think would prevent a roll back?
(13:13:49) moyses: we must bring this to the forums red1
(13:13:55) CarlosRuiz: What I'm sure is that current approach is not working, so I don't think we are going to roll back
(13:13:58) CarlosRuiz: we're going to evolve
(13:13:59) moyses: as you said all the views have to be heard!
(13:14:00) red1: who is stopping you?
(13:14:15) Michael_Judd: once something is put in place - it is too dififuclt to change. We have taken a long time on the foundation but found a very good solution ...
(13:14:19) moyses: you want to take decisions now now now!
(13:14:28) moyses: so hold on!
(13:14:28) CarlosRuiz: we found a very good solution?
(13:14:29) red1: carlos called for this meeting to evolve things
(13:14:37) moyses: dont take decisions!
(13:14:46) red1: then when?
(13:14:48) moyses: red1 repeat your own words, now carlos words!
(13:15:06) moyses: I have the sense that you repeat just what carlos said :)
(13:15:07) red1: what shuld we do then?
(13:15:15) nwessel_2: good point, red
(13:15:16) CarlosRuiz: I think we NEED to take decisions
(13:15:19) AS6: Moyses: Did you read the above log? No decision are to be made today. Only formulation of proposal which will be voted in forums.
(13:15:23) CarlosRuiz: just that decisions are never definite
(13:15:28) Michael_Judd: why do we need to take a decision today ?
(13:15:32) red1: i try defending everyone here
(13:15:32) nwessel_2: I agree, Carlos
(13:15:38) moyses: AS6 red1 indicated that decisions should be taked
(13:15:40) nwessel_2: at least, prepare decisions
(13:15:42) AS6: No carlos, the quorom is not enough
(13:15:54) vpj-cd: It hink we can follow our politic
(13:15:57) AS6: We should be getting inputs and voting from all to have a decision
(13:15:59) red1: to me its suggestion
(13:16:03) red1: subject to review
(13:16:06) vpj-cd: plublish adn voting :-)
(13:16:18) vpj-cd: this way include the comunnity
(13:16:18) CarlosRuiz: voting what?
(13:16:22) moyses: Agree with you AS6! there is not enough Quorom!
(13:16:22) red1: carlos must be out of his mind to force a hard coded decision :>
(13:16:26) CarlosRuiz: what I say is that current approach is not working
(13:16:33) CarlosRuiz: so a vote to maintain the current approach is not valid
(13:16:38) red1: quorum doesnt really arise
(13:16:43) red1: a meeting was called
(13:16:43) Michael_Judd: why not ?
(13:16:52) vpj-cd: no CarlosRuiz to new propouse
(13:16:56) red1: now we are publishing this idea in the forums
(13:17:05) red1: u have your say
(13:17:09) moyses: the voting is not about keeping the current state
(13:17:12) red1: many agreed including me
(13:17:15) Bahman: Mike, you mean another meeting, a thread in forums?
(13:17:17) moyses: it is about the new one proposed...
(13:17:20) AS6: It looks like we have come with a solution, let it be published and be decided in the forum, that we take into consideration the other members' view.
(13:17:30) CarlosRuiz: ok, let's do it, as red1 said, I would like to have problems solved, one thing I worry about this community is that never take decisions
(13:17:32) AS6: and get the votings
(13:17:34) moyses: agree with AS6
(13:17:35) red1: but meanwhile i like to see more 'protoype of ideas
(13:17:36) vpj-cd: yes agree with AS6
(13:17:49) Michael_Judd: who will answer my questions on foundation ?
(13:17:51) moyses: red1 you have lost your charm about politics!
(13:18:00) moyses: you have been charged to one side :)
(13:18:06) vpj-cd: is spirit of baazar
(13:18:10) Michael_Judd: contributor agreements ?
(13:18:24) red1: what carlos said here is just about his 2 issues
(13:18:29) red1: we are losing focus
(13:18:38) CarlosRuiz: we still need a second meeting for point 1 - Centralized Application Dictionary Management
(13:18:39) moyses: haha red1 repeat once again carlos!
(13:18:52) red1: yes i am
(13:19:00) nwessel_2: so what, moyses?
(13:19:02) moyses: we are lossing focus, so he will said next Who wants quality?
(13:19:14) red1: cos of his role and responsibility
(13:19:21) moyses: we need to put this in the forums as AS6 pinted out
(13:19:26) red1: this is meritocracy
(13:19:33) moyses: whats your point nwessel?
(13:19:34) AS6: Moyses: I can assure that we are not being personal here, its just business. The meeting was called today to address specific issue. And that is what we are trying to do. And please be assured, no one here is one-sided. Its just in the best interest of the community.
(13:19:35) red1: we shall
(13:19:53) red1: my aplogies for siding carlos
(13:19:54) Bahman: Alright...let's summarise the discussion....we're gonna put it in SF so that every one can criticise (in a reasonable duration).
(13:19:56) Michael_Judd: This is bad for business IMHO
(13:19:59) CarlosRuiz: I'm one sided
(13:20:00) red1: but i m no coward to admit so
(13:20:00) Bahman: Aren't we?
(13:20:06) CarlosRuiz: I'm one sided with good proposals
(13:20:08) red1: i can always take the popular stand
(13:20:10) red1: but i cannot
(13:20:16) red1: i must steer something
(13:20:20) moyses: yes cArlos you just do good -proposals!
(13:20:32) red1: to say nothing is worse
(13:20:39) moyses: I know you just sound as CarlosL
(13:20:42) red1: i still think the group idea by mike is good
(13:20:45) Bahman: I think as somebody else mentioned before we all had our words here....
(13:20:55) red1: it has something along islamic ways i must say
(13:20:57) Bahman: That's the spirit of community to win the voting or not...
(13:21:00) red1: but that is my personal opinion
(13:21:02) moyses: lets bring it to the forums
(13:21:14) moyses: no decisions here!
(13:21:14) Bahman: just like the security issues discussion in SF....
(13:21:16) red1: moyses can u do me a favour?
(13:21:32) red1: can u help to publish this in the forum.. its late here in Malaysia
(13:21:34) Bahman: So I'd suggest Mike putting down his mind in SF.
(13:21:45) CarlosRuiz: yes, I already said, let's do it in forums, just that I think WE MUST TAKE DECISIONS, and we're failing on this
(13:21:52) CarlosRuiz: but anyways, I can't change this easily
(13:22:11) AS6: What we can do is to establish a deadline to reach the decision.
(13:22:27) moyses: Bahman will do it red1
(13:22:31) CarlosRuiz: and I stick thinking that we're becoming Debian -> not taking any decision at all
(13:22:36) red1: i suggest let us walk this road, then review
(13:22:36) moyses: he just volunteer for it red1
(13:22:43) AS6: Bahman, i`m around if u need me to assist u.
(13:22:44) red1: syukran bahman
(13:23:01) Bahman: No problem for me if anyone else wants to take the job ;-)
(13:23:03) nwessel_2: I do agree Carlos
(13:23:09) red1: remember moyses u have nothing to fear from me :D
(13:23:12) Bahman: Thanks AS6! And welcome red1!
(13:23:21) moyses: neither you red1 from me :)
(13:23:32) red1: i m your friend
(13:23:47) red1: your only friend at times like this *chuckles*
(13:23:51) moyses: eres amigo!
(13:24:14) Michael_Judd: let try to make GOOD decisions ...
(13:24:23) CarlosRuiz: ok, I'll make a call for another meeting next tuesday -> 1 - Centralized Application Dictionary Management
(13:24:35) red1: and that can only be done with good open healthy brave objections !
(13:24:38) Bahman: 2. CC can't cope ... ;-)
(13:25:04) moyses: red1 you are always welcome at my table :)
(13:25:21) red1: byw CarlosRuiz when can we release the next 3.3?
(13:25:41) Bahman: Oh! Not now! Please!
(13:25:42) Bahman: :-P
(13:25:47) red1: hehe moyses.. we came a long way together..
(13:25:51) red1: not asking u bahman...
(13:25:55) AS6: OK, guys we are all brothers here from one family. Lets not take anything personal. There is no right and no wrongs. Only differences of views and opinions. Lets bring our conclusion to the SF and get the community voting. I believe all of us here will be very objective in addressing the issue that is to be addressed.
(13:26:06) teo_sarca: bahman is commiting a suicide :)
(13:26:13) CarlosRuiz: not sure, maybe we can ask in forums, I'm being accused of dictator
(13:26:15) fredtsang_: LOL
(13:26:15) Bahman: :-)
(13:26:23) red1: i better have a good sleep and tomorrow put on my rifle and sword
(13:26:24) Michael_Judd: Not accused !
(13:26:28) moyses: please hold on bahman!
(13:26:37) red1: the seas are rough this time of the year
(13:26:41) fredtsang_: bahman thanks for being brave!
(13:26:46) Michael_Judd: just tendencies ;) and I understand your problem - I jsut thing perhaps there is a better solution ...
(13:26:58) Bahman: Just think of my back and the length of the log!
(13:27:12) CarlosRuiz: we can make votations on when to release 3.3
(13:27:12) AS6: Red1: I had good rounds of laughs reading your amusing jokes, all the way. :D
(13:27:16) vpj-cd: what is summary :-0
(13:27:19) vpj-cd: :=)
(13:27:31) red1: its past 2 am.. i can only do that AS6
(13:27:35) Bahman: Find it in SF Victor ;-)
(13:27:38) red1: otherwise i will fall on this table
(13:27:47) vpj-cd: I like release 3.3 when libero is ready via 2pack
(13:28:00) fredtsang_: 3 hr + meeting
(13:28:01) AS6: red1: Yours is a glass table, u better not fall, otherwise...hehehhe :D
(13:28:01) red1: when is that victor?
(13:28:18) vpj-cd: I home 2 week
(13:28:20) moyses: we must test it first I believe victor!
(13:28:27) vpj-cd: we test and only are solve some issue
(13:28:47) Bahman: And prepare a good, all in one patch for testers Victor.
(13:28:48) moyses: how can we help to test it out?
(13:29:19) red1: hope Tim wont ask "Have u seen Victor?" again after this
(13:29:25) vpj-cd: you can use the tutorial of tim
(13:29:30) moyses: red1 LOL!!
(13:29:34) CarlosRuiz: maybe we need to vote if libero is included (I'm joking Victor)
(13:29:38) Bahman: *Signs of constructive discussion!*
(13:29:56) moyses: My vote to not included Libero :)
(13:29:57) red1: Goodnite my new dictator!
(13:29:59) Bahman: Should I put that in SF too Carlos?
(13:30:03) Bahman: ;-)
(13:30:09) CarlosRuiz: hahahaha
(13:30:15) moyses: I second Carlos here!
(13:30:18) red1: now i cannot stranggle u
(13:30:20) CarlosRuiz: I'm opening a votation to lunch
(13:30:31) Michael_Judd: good evening all ;) thanks for listenign to my comunisitic ideas ....
(13:30:33) Bahman: Votation to dinner and sleep :-)
(13:30:36) hengsin: Good morning: red
(13:30:41) fredtsang_: good evening all
(13:30:46) vpj-cd: jejejej
(13:30:46) Bahman: Bye all!
(13:30:47) CarlosRuiz: c u later, thank you very much to everybody
(13:30:50) teo_sarca: vote ! i am starving
(13:30:51) red1: a tout al heure
(13:30:56) Bahman: Very good and constructive discussion...
(13:30:57) red1: au voir!
(13:31:00) Bahman: wait...
(13:31:10) red1: no more latin please
(13:31:10) Bahman: What should I do to edit the draft?
(13:31:13) vpj-cd: no is necessary Carlos you only can instal or uninstall
(13:31:18) Bahman: Anyone wants to edit it?
(13:31:20) Bahman: red1?
(13:31:22) vpj-cd: this is aventage the use 2pack :-)
(13:31:22) red1: no need edit
(13:31:25) fredtsang_ ha salido de la sala (quit: "ChatZilla 0.9.78.1 [Firefox 184.108.40.206/2007051502]").
(13:31:28) moyses: bon appetite to all!
(13:31:29) red1: as it is.. warts and all
(13:31:36) Bahman: Then goodnitgh!
(13:31:39) teo_sarca: bye bye to all
(13:31:42) red1: let the dictator burn
(13:31:43) AS6 ha salido de la sala.
(13:31:49) ***Bahman is away: Dinner time!
(13:31:56) vpj-cd: ok
(13:32:04) red1 ha salido de la sala (quit: "zzz").
(13:33:26) moyses ha salido de la sala (quit: "Abandonando").
(13:33:30) hengsin_ [firstname.lastname@example.org] ha entrado en la sala.
(13:33:43) hengsin_ ha salido de la sala.
(13:48:05) karsten-thiemann [n=chatzill@p57A0B96D.dip0.t-ipconnect.de] ha entrado en la sala.
(13:49:11) karsten-thiemann ha salido de la sala (quit: Client Quit).
(13:50:07) hengsin ha salido de la sala (quit: Read error: 110 (Connection timed out)).