Transcript of NetFIX defining chat

Topic: Definition of NetFIX process
Date: March 6th, 2009
Time: 4:00 PM Central European Time
Duration: 2 hours
Conversation: netbeans
Chat Client: NetBeans Developer Collaboration module
  • Tushar Joshi (India)
  • Melongo Annabel (USA)
  • Jiri Kovalsky (Czech Republic)
  • Michael Santos Nascimento (Brazil)
  • Jean Morissette (Canada)
  • Michel Graciano (Brazil)
  • Anuradha Gunasekara (Sri Lanka)
  • Saptarshi Purkayastha (India)
  • Rajiv Perera (Sri Lanka)

  • Agenda:
    1. CA (Annabel Melongo, Tushar Joshi, Rajiv Perera, Jean Morissette, Enrico Scantamburlo)
    2. Temporary initiative: April - June
    3. mailing list?
    4. Regular coordination meetings?
    5. How much time per week? Weekends or working days? Feature specialization or no focus?
    6. Volunteer for coordination?
    7. Organization: easy-to-fix issues in Wiki, pick one, attach patch, mark as fixed
    8. Open discussion


  • Signed Contributor Agreement is required in order to participate in the NetFIX program, Jiri to ask those who didn't send their CAs off-line.
  • Almost everyone wants to contribute on an ongoing basis i.e. overall disagreement about NetFIX being a time limited initiative.
  • Do we replace inactive NetFIX participants? What would be the criteria? We will watch progress first and decide later when the dust settles down.
  • Jiri to setup mailing list and subscribe everybody from the chat.
  • Michael Nascimento: "I prefer contributing small enhancements. I want to join NB team and follow the same rules." Michel also wants to implement small features.
  • There will be no regular meetings. We will meet on demand and everyone will be entitled to call for a meeting. We only find ideal time e.g. Friday, 3:00 p.m. UTC
  • Distribution of the team: 1/2 will contribute on weekends, 1/2 during working days, 2 contributors 16 hours/week, others ~8 hours/week.
  • Since nobody can be an expert in everything, NetFIX participants will specify their preferred functionality areas. TBD via e-mails.
  • Michel Graciano was elected as NetFIX team community representative expected to watch progress, help others and escalate important issues to Jiri.
  • Jiri to create a new netfix keyword and setup new Wiki page with "free-to-netfix" issues. Jiri to inform NetBeans Engineering about this.
  • Process:

  • NetFix developer wants to fix some issue
  • if the issue has netfix keyword, s/he is entitled to assign the issue to her/himself and start working on the fix
  • if the issue does not have netfix keyword, s/he must ask module owner for permission to take ownership of the issue via adding a comment
  • if module owner agrees, NetFIX developer must update Wiki pool with "free-to-netfix" issues and then s/he can start working on the fix, in other case an explanation of the denial must be provided
  • when the bug fix is ready a patch must be attached to the issue and the Wiki page must be updated
  • module owner is expected to review the patch at his/her earliest convenience and either integrate it or reject it with justification
  • 4:02:10tusharvjoshi joined the conversation
    4:03:33Jean Morissette joined the conversation
    4:04:01Jiri Kovalsky Hello guys, I am trying to get everyone here but so far it looks like only three of us are online.
    4:05:07Jean Morissette How many have confirmed their precense?
    4:05:12mister__m joined the conversation
    4:05:37Jiri Kovalsky 8 people wanted to come.
    4:05:53mister__m i'm here hi everyone
    4:05:59Jiri Kovalsky Saptarshi told me he was in the middle of forests with a weak Internet connection.
    4:06:05mister__m sorry for being late
    4:06:09Jean Morissette Hi =)
    4:06:18Jiri Kovalsky Hi Michael
    4:06:20tusharvjoshi Hi to everyone from Tushar
    4:06:53Jiri Kovalsky Pretty bad that Anuradha is not on-line. :-|
    4:07:27Jiri Kovalsky Let me try to see if he and Rajiv are online via Skype.
    4:07:59Jiri Kovalsky Maybe they are on another unexpected meeting.
    4:08:29Jiri Kovalsky I suggest we wait 2 more minutes and then start, ok?
    4:08:35tusharvjoshi ok
    4:08:36Jean Morissette sure
    4:09:03mister__m agree
    4:10:15Jiri Kovalsky In the meantime I have shared the meeting agenda with you.
    4:10:24mister__m btw
    4:10:26mister__m jiri
    4:10:41mister__m michel is having trouble logging in
    4:11:07Jiri Kovalsky Hm, strange. He was doing fine last Friday.
    4:11:27mister__m yep
    4:11:35mister__m he's restarting nb right now
    4:11:38Jiri Kovalsky Great, at least Melongo will be joining us.
    4:11:44amelongo joined the conversation
    4:11:45Jiri Kovalsky OK.
    4:12:22amelongo hi everybody
    4:12:33Jiri Kovalsky Hello Melongo! Let's wait also for Michel.
    4:12:39hmichel joined the conversation
    4:12:39Jean Morissette hi, nice to meet you =)
    4:12:58hmichel Hi people
    4:13:02Jiri Kovalsky OK, I think we should start.
    4:13:16Jiri Kovalsky Hello everyone, let me welcome you in today's chat dedicated to NetFIX - a brand new initiative to help with bug fixing NetBeans IDE.
    4:13:28Jiri Kovalsky I would like to emphasize that the primary goal of this meeting is to brainstorm the NetFIX idea and try to define its rules.
    4:13:31tusharvjoshi Hello Everyone
    4:13:49Jean Morissette Hi everyone =)
    4:13:51Jiri Kovalsky The first item on the agenda are Contributor Agreements. In order to contribute patches and then obtain push access rights to our Mercurial repository you must sign the CA. According to my ivestigation, the following persons don't have it signed yet: Annabel Melongo, Tushar Joshi, Rajiv Perera, Jean Morissette, Enrico Scantamburlo.
    4:14:13Jiri Kovalsky Could you guys please sign the CA and deliver it to me as soon as possible? The instructions are here:
    4:14:25tusharvjoshi Sure will do that
    4:14:30amelongo sure
    4:14:31Jean Morissette I will fax it to you after our meeting
    4:14:37mister__m i can confirm i have signed it
    4:14:59Jiri Kovalsky Excellent.
    4:15:02hmichel I already signed too
    4:15:38Jiri Kovalsky OK, so I am gonna contact those whose CAs are missing off-line after the chat.
    4:15:54Jiri Kovalsky FYI, once your CA is checked and accepted, your entry will show up here:
    4:16:05Jiri Kovalsky Secondly, I would like to highlight that NetFIX program should be a time limited activity similarily to NetCAT. In particular from April to June.
    4:16:21Jiri Kovalsky Does anybody have a problem with this?
    4:16:35Jean Morissette Can I ask why?
    4:16:40amelongo I have no issue
    4:16:41mister__m depends on how we define NetFIX
    4:16:44tusharvjoshi I dont see a problem
    4:17:08Jiri Kovalsky That's right. It depends on how we define it but let me answer Jean's question.
    4:17:10mister__m i guess your question will be answered easily by the time we finish this brainstorm
    4:17:43sunbiz joined the conversation
    4:17:52Jiri Kovalsky My experience with NetCAT program shows that it's better to have them temporary because of two reasons.
    4:19:00sunbiz Hi guys... sorry got a bit late with my conn
    4:19:13Jiri Kovalsky Community usually contributes in cycles: people come, contribute and go. It's because of other duties. They feel they have some time for couple of weeks/months so that's the first reason.
    4:19:31sunbiz do we have a log of this chat somewhere so that I could get on the same page ???
    4:20:47Jiri Kovalsky Secondly, it was always good to refresh the NetCAT team with new members with every release because even the most useful participants were half efficient in the next rounds.
    4:21:12Jiri Kovalsky Do you think Jean you would like to support NetBeans on an ongoing basis?
    4:21:29Jean Morissette This was my intention
    4:21:34Jiri Kovalsky I will send it to you Saptarshi. Wait a second.
    4:21:41mister__m even though i am not jean, i surely intend to contribute on a ongoing basis
    4:21:49Jiri Kovalsky OK, what are other's intentions?
    4:22:08amelongo I  can contribute on an ongoing basis too
    4:22:21tusharvjoshi I see myself contributing on a ongoing basis also
    4:22:45sunbiz I would like to fix my fav IDE, which I've been getting a lot from...making me more productive... I intend to try to give back something its given me!!
    4:23:13Jean Morissette What I was hoping is to help fix bugs and maybe in the future get a voice for more high level feature, collaborating with NetBeans developers ultimatly.
    4:23:44Jiri Kovalsky I see. So do we want to define some rules when somebody wants to resign?
    4:23:55Jean Morissette Like Apache open source projects by example
    4:24:09Jiri Kovalsky Or his/her participation in NetFIX program will get to zero?
    4:24:43Jiri Kovalsky What aspects do you like most Jean?
    4:24:47mister__m i guess thee is no need to resign actually
    4:24:56sunbiz yes.... I agree we need to define some rules if this is going to be an ongoing contribution thing
    4:25:23mister__m there are valid reasons to remove someone from the program
    4:25:25Jiri Kovalsky You know there are situations when people feel they can't contribute any longer.
    4:25:28sunbiz because then we need to make room for others when some people don't participate as much...
    4:25:28mister__m for misbehaviour
    4:25:30amelongo Maybe we can set a standard for a number of bugs fix on a monthly basis and if that standard isn't met that we ask the developer why? if the team isn't happy with the answer then the developer is replaced
    4:25:32mister__m no problem
    4:25:43tusharvjoshi If one see that he/she will not be available for development for certain period, we can provide a off period for that person who can join again later when available
    4:25:46mister__m if they can't contribute at the moment, they just don't get assigned any tasks
    4:26:13sunbiz I like amelongo's suggestion!
    4:26:27mister__m except...
    4:26:34mister__m some tasks are larger than others
    4:26:40Jiri Kovalsky OK, I would like to hear Jean's take on this. Jean?
    4:27:03Jean Morissette I also think tht we neeed some kind of mesurement
    4:27:22Jiri Kovalsky Of course, large assignments are time consuming so we should take this into account.
    4:27:28mister__m well
    4:27:33mister__m i don't like this idea
    4:27:36mister__m :-P
    4:27:37Jiri Kovalsky Great, so what will that criterion be?
    4:27:48mister__m someone might get temporarily busy
    4:28:04mister__m they shouldn't be excluded just for that
    4:28:08Jiri Kovalsky How do you define that "temporarily" Michael?
    4:28:18mister__m couple of weeks maybe
    4:28:34sunbiz I think 1 P1 || 2 P2 || 3 P3 || 4 P4 bugs a month should b a good start
    4:28:38Jean Morissette Maybe we could inspire ourself from Apache rules which are based on meritocracy.
    4:28:44mister__m yep
    4:28:51mister__m meritocracy is much better
    4:29:05mister__m and it is not related to how often u can contribute
    4:29:19tusharvjoshi I agree with Saptarshi for measurement
    4:29:41mister__m measurement is tricky
    4:29:47mister__m because
    4:29:48sunbiz and then as we all start contributing more... we increase the criteria
    4:29:56mister__m priority != difficulty
    4:30:03mister__m patches are atomic
    4:30:12mister__m so one issue might require a month of work
    4:30:29mister__m how do you track if a person is working or not?
    4:30:33sunbiz yes... and we will consider those as a community I think
    4:30:50amelongo I think we're talking about the same thing. A P1 priority of course is more demanding that a P3; so if a developer can only fix a P1 on a monthly basis, I don't see why he/she would be asked to leave.
    4:30:57Jiri Kovalsky Well, I would like to say though that we should start with Easy-to-fix bugs first which should not require months of work.
    4:31:00mister__m no
    4:31:10mister__m P1 is not more demanding than P3
    4:31:14mister__m depends
    4:31:18mister__m P1 is a blocker
    4:31:20mister__m basically
    4:31:23Jean Morissette I agree with mister on that
    4:31:25mister__m P3 is a nice-to-have
    4:31:34mister__m blocker might require 3 lines of code to get fixed
    4:31:38Jiri Kovalsky Such big integrations would also require significant time on side of NetBeans engineers which is what I would rather like to avoid in the beginning.
    4:31:41mister__m P3 might require months of work
    4:32:13sunbiz yes... I think Jiri makes a valid point
    4:32:23tusharvjoshi I can see the point mentioned by mister__m
    4:32:38sunbiz may b the wiki can give good hints on what NetFIX participants can target
    4:33:06Jean Morissette Yes, I think taht we should take that in phase at first and adjust after a certain period of time. We could start with mesurement and evolve towaard meritocracy after a certain period. This would help get some momemtum at the start.
    4:33:20Jiri Kovalsky I think nobody questions that priority does not determine complexity of a fix.
    4:33:28mister__m ok, jean, but how do we measure?
    4:33:56mister__m and what is the point of measuring it?
    4:33:59Jean Morissette How do you mesure in Sun internal?
    4:34:40Jiri Kovalsky And what if we only watch the progress for first two months and decide the criterion later?
    4:34:50Jean Morissette agree
    4:34:57Jiri Kovalsky Measure what? How much efficient engineers are?
    4:35:10tusharvjoshi I like the idea of watching first and then again come back to decide the criteria
    4:35:12mister__m i really think this measurement discussion would be clearer later on
    4:35:35Jiri Kovalsky Yes, our discussion is very vague at this point.
    4:35:41sunbiz   Does Sun measure efficiency ???
    4:35:48mister__m once we define how we will organize issues, contributions and eventual Mercurial access
    4:36:08Jean Morissette I just want that we have the possibility to keep teh project alive and have hope to get integration with Sun developer at some point. This is my personall motivation. Otherwise, I could just sumbit patchs.
    4:36:48Jean Morissette I would like to learn from you
    4:36:52Jiri Kovalsky Of course, our managers want to measure progress but it's hard to define some concrete formula for that.
    4:36:56mister__m looking at the agenda,jiri, i have the impression your concept of netfix is different from what i had in mind
    4:37:00mister__m so
    4:37:01Jiri Kovalsky It's per manager, per project etc.
    4:37:26Jiri Kovalsky OK, so let's go to the next item.
    4:37:32mister__m i thought netfix was about getting good developers to contribute to netbeans and eventually become part of the team, i.e., follow the same rules
    4:37:35Jiri Kovalsky Now, I have one question for you. Do you think it would be beneficial if I setup a mailing list for communication of NetFIX program members?
    4:37:45mister__m yes (for mailing list)
    4:37:52Jean Morissette agree with mister
    4:37:53tusharvjoshi yes @ mailing list
    4:38:13sunbiz yes... agreed
    4:38:16Jean Morissette +1 mailing list
    4:38:22amelongo Jiri, why don't we start with a "at least one bug a month" strategy and after that, when we get acquainted with the code, we can move on to other measurements.
    4:38:24Jiri Kovalsky Michael, hold your horses. NetFIX is a brand new project that must first show some results.
    4:38:33hmichel I think so, mainling list is important
    4:39:11amelongo I agree with mailing list too
    4:39:20Jiri Kovalsky I really don't know how this can evolve but of course if some people qualify to be great contributors they could take over some functionality if we don't have our own resources for that.
    4:39:38Jiri Kovalsky So, is good for everybody?
    4:39:46sunbiz yes
    4:39:48Jean Morissette yes
    4:39:57tusharvjoshi sounds perfect
    4:39:58amelongo sure. I like that
    4:40:03rajivderas joined the conversation
    4:40:16mister__m i understand jiri... so what i had in mind was creating a netfix user in issuezilla which you would control and use it to determine which issues are on our hands
    4:40:28sunbiz Hi Rajiv
    4:40:51mister__m the wiki page would tell which issues are assigned to each contributor
    4:41:01Jiri Kovalsky Melongo, let's see how much we can get out of the NetFIX first and then setup the right expectations in terms of efficiency.
    4:41:13rajivderas hi sunbiz
    4:41:15Jiri Kovalsky Hello Rajiv! Do you know anything about Anuradha?
    4:41:33rajivderas he is on the way
    4:41:47anuradha joined the conversation
    4:42:01Jiri Kovalsky Hm, netfix user. I will have to think about it.
    4:42:22Jiri Kovalsky Because that way we will not see who individual did what ... :-\
    4:42:29Jiri Kovalsky Welcome Anuradha!
    4:42:42sunbiz guys... my cell phone battery is dying out.... so I'll b gone any moment... I hope I can get the transcript of the chat
    4:42:54anuradha hey sorry for being late
    4:42:57mister__m ok, so we might use our iz users
    4:43:00sunbiz Hi Anuradha
    4:43:03mister__m and a netfix keyword
    4:43:13rajivderas sorry for the late entrence
    4:43:14mister__m or status whiteboard
    4:43:29anuradha so where are we ?
    4:43:34Jiri Kovalsky Sure, I think creating new netfix keyword will not be a problem.
    4:43:51Jiri Kovalsky I will send you the transcript Anuradha!
    4:44:04Jiri Kovalsky Next item on the agenda are regular meetings. Do you see a need for them? We meet bi-weekly with VOC Captains to discuss the latest issues.
    4:44:04anuradha sure thanks
    4:44:07hmichel I like that idea, assign issues for our users and mark issue with an keyword is a good way to track things
    4:44:29rajivderas could you send the  transcript also
    4:44:41sunbiz yes... I think regular meetings are a must... bi-weekly sounds good enough
    4:45:00anuradha +1 for bi-weekly
    4:45:08mister__m is there a need for a meeting if we have a mailing list?
    4:45:21amelongo Since we already meet bi-weekly at VOC, I would suggest we meet once a month at Netfix.
    4:45:22mister__m why not just having a status report as jiri does for netcat?
    4:45:23sunbiz and we need to flag ppl missing them... and then may b some limits to how many we can miss
    4:45:38tusharvjoshi I suggest a meeting per month, and mailing list to handle most of the discussions
    4:45:57mister__m all this kind of controls seem unnecessary to me, really
    4:46:00mister__m this is open-source
    4:46:12rajivderas weekly might be goog since then we have some time period
    4:46:19Jiri Kovalsky Meetings are faster than mailing list communication.
    4:46:31rajivderas thats true jiri
    4:46:44mister__m i agree jiri, but what is the point of the meeting if we will be talking in the ml for two weeks?
    4:46:46sunbiz hmm... then it already is open-source and you can today contribute patches with the SCA-signed
    4:46:59rajivderas but having them regulerly is eating time
    4:47:05anuradha both mailing list and meetings is I prefered
    4:47:17hmichel I think meetings are necessary when we had a agenda, or a problem to define where the mainling list is not enough
    4:47:21mister__m @sunbiz: yes/no, since there is no promise it will get accepted
    4:47:27mister__m i agree with hmichel
    4:47:40rajivderas  i also agree with hmichel
    4:47:46Jiri Kovalsky What if we don't set any period and we would only agree on some time good for everyone and meetings would be organized only on-demand basis. Everyone would have a right to call for a meeting if s/he wants to discuss something.
    4:47:46hmichel the another problem, is the timezones
    4:48:00hmichel really hard to meet all people in the same time
    4:48:05mister__m @Jiri: agreed
    4:48:15anuradha @Jiri +1
    4:48:18tusharvjoshi agree @ Jiri
    4:48:34hmichel @Jiri +1
    4:48:48rajivderas but the demand should be of valid state
    4:48:51Jiri Kovalsky Great, so we have a consensus on that.
    4:49:05Jean Morissette I feel that it's a good idea to make the point after some time internal. let's start with demand-based meeting.
    4:49:14Jean Morissette interval*
    4:49:19Jiri Kovalsky Now, something more difficult. How much time can you afford to spend on bug fixing?
    4:49:43Jiri Kovalsky And when do you usually expect to work on bug fixing? Weekends? Working days?
    4:50:23Jiri Kovalsky Do you prefer being focused on just couple of modules or do you think you could cover whole codebase?
    4:50:35rajivderas Weekends
    4:50:45amelongo for me it would be mostly on week-ends; and maybe late at night when I can't sleep
    4:50:58mister__m i would do my contributions mostly on working days
    4:51:00Jean Morissette Each developer should be assined one or some module only imo
    4:51:21mister__m @jean: it's hard to divide the team in modules
    4:51:21tusharvjoshi I see this contribution as an extension to my working days, so working days for me
    4:51:28amelongo Why don't we start with modules first, and then progress to the codebase...
    4:51:38mister__m u should be able to contribute to whatever u feel comfortable
    4:51:43hmichel In my case, the P1 for me is issues that interfir in my day job, so, I will try to work on it at working days if possible, with no meansuring of time.
    4:51:55rajivderas @mister__m +1
    4:52:46Jiri Kovalsky Yes, but frankly and I don't want to offend anybody here it is very hard to be expert in everything. I hope you get my point.
    4:53:02rajivderas yup
    4:53:09hmichel @Jiri +1
    4:53:13tusharvjoshi I agree with Jiri
    4:53:13mister__m i agree with u jiri, but some fixes require just trivial Java knowledge
    4:53:19anuradha I prefer core,maven ( ) ,java
    4:53:23mister__m others require module knowledge
    4:53:40mister__m and i guess some people here might be interested in roughly the same modules
    4:53:51rajivderas @ mister__m good point
    4:53:57Jiri Kovalsky Looks like the team is devided into 1 half for working days and the other half for weekends. Great ...
    4:54:04tusharvjoshi Yes I also prefer core, platform, java and editor parts
    4:54:12amelongo but even though the bug might be simple, it really take time to know the code. That's why we should progress slowly
    4:54:22Jiri Kovalsky Why maven Anuradha? ;-)
    4:54:36rajivderas i agree with  amelongo
    4:55:20Jean Morissette Maybe we should let you know what modules we are interested and you would submit us bugs taht respect those preferences?
    4:55:21Jiri Kovalsky OK, so let's discuss the preferred functionality areas over the e-mail. OK?
    4:55:42mister__m i wonder if we really need to define this
    4:55:50sunbiz left the conversation
    4:56:11Jiri Kovalsky And one overlooked question: how much time per week can you on average dedicate to NetFIX? Please try to be conservative.
    4:56:14anuradha @Jiri I fixing my own bugs as I have some experience with bugfixing on netbeans code base will take P1 - P3 and 16 hours week if possible
    4:56:55amelongo 16 hours/ weekly for me is fine
    4:57:11tusharvjoshi I see contributing 8 hrs per week and more some weeks according to the schedules
    4:57:25Jean Morissette I would say 8h/week
    4:57:28Jiri Kovalsky So you basically don't care what you get Michael? And are you sure you will be able to fix anything?
    4:58:06Jiri Kovalsky Michel? Rajiv?
    4:58:10anuradha I say 5 bug's per week
    4:58:39hmichel I think the right point here, is that each one can choose the issues that are relevant for each one
    4:58:47mister__m no, rather the opposite
    4:58:55rajivderas have start from simple fixes
    4:59:01Jiri Kovalsky Yes, that was my idea @ michel.
    4:59:01mister__m i don't understand how choosing modules will help
    4:59:08hmichel not necessarily in the same module
    4:59:30hmichel for example, I had filed some issues that are typo, it is really easy to solve
    4:59:50Jiri Kovalsky You don't have to study different code. You get familiar with one area and fixing in that area will take you less time then @Mister_m
    4:59:54rajivderas the hours of contribution by my self may vary
    4:59:59amelongo @mister_m: modules choosing help in that, you familiarize yourselft with that module and the better you know it, the easier will be to fix bugs
    5:00:34mister__m it should only matter for very specific bugs
    5:00:55Jiri Kovalsky Michael, Rajiv - how much time per week can you afford?
    5:00:56mister__m the easy-to-fix will be, in general, not related to module knowledge, for instance
    5:01:11mister__m and should also give u very little module knowledge
    5:01:13rajivderas 8 or 16
    5:01:17mister__m 8h/week
    5:01:53hmichel I understand rhe amelongo point, but, there are any problem if I found a issue at maven module and post a patch, even another one are the module owner? I don't think so. I think if someone want to focus on a module, no problems, and if someone will just choose issues ramdonly, by importance or whatever reason, I can't see problem too
    5:02:29hmichel I think I can work between 4h/8h per week
    5:02:34Jean Morissette If we can choose our bugs, we are not forced to have preassigned modules; this will happen naturally.
    5:02:59amelongo @hmichel: I don't have any issue with that. But an expert in an module will do the work quickly though and more efficiently
    5:03:03rajivderas @jmori good point
    5:03:14anuradha any case if someone need help on issue lets help other buy giving hints and pointers
    5:03:26mister__m agree with jean
    5:03:34rajivderas @ anuradha +1
    5:03:50tusharvjoshi I agree with Anuradha, and jmori
    5:03:56Jiri Kovalsky OK, thanks for your honest answers guys. Now would somebody be interested in coordinating this effort?
    5:03:58mister__m the mailing list will also be a source of advice @anuradha
    5:04:09anuradha typo buy - > by
    5:04:38Jiri Kovalsky I don't insist on that of course ...
    5:05:01Jiri Kovalsky I only wanted to know if there is somebody really interested in this.
    5:05:07tusharvjoshi Jiri, what actions are expected in coordinating the effort?
    5:05:16mister__m what do u envision as coordination activies?
    5:05:18rajivderas i think  anuradha is a good candidate
    5:05:35mister__m i vote for hmichel
    5:05:39mister__m :-P
    5:05:55mister__m but would like anuradha as well, for his commit experience
    5:06:45tusharvjoshi Jiri, are you asking for a self interest or nomination?
    5:06:49Jiri Kovalsky Well, I would naturally serve as liaison on NetBeans engineering side but that person would be responsible for monitoring the progress, helping others with less experience etc.
    5:07:32Jiri Kovalsky Anuradha, would you be interested in this? Or do you prefer only fixing?
    5:07:48Jiri Kovalsky I asked about self interest @tushar.
    5:08:05hmichel I can help about it too, sonce it is a VOC captain interesting too
    5:08:10anuradha I prefer only fixing
    5:08:34Jiri Kovalsky NetFIX coordinator would be required to escalate to me everything what the team feels is urgent etc.
    5:08:37mister__m +1 for hmichel :-P
    5:08:41rajivderas @ anuradha
    5:09:15hmichel I think it is similar responsability, and I can ask for help when necessary of course
    5:09:23tusharvjoshi +1 for hmichel
    5:09:31amelongo @rajivderas: really funny... don't force people to do what they don't want...
    5:10:28rajivderas @amelongo to do things right you should have the right person
    5:10:31Jiri Kovalsky So, what about you Michel? Do you think this would be satisfying for you? Or do you also prefer fixing?
    5:10:42hmichel no problems to me
    5:10:43amelongo if michel, if he wants it, why not?
    5:10:49anuradha anyway will help other with bug hunting but on coordinating job's NO!
    5:11:11anuradha + 1 hmichel
    5:11:33Jean Morissette hmichel +1
    5:11:43amelongo hmichel +1
    5:12:03hmichel :-)
    5:12:48Jiri Kovalsky So, I take it as you accept this nomination Michel... Speak up please unless it's yours! :-)
    5:13:16hmichel I do
    5:13:27Jiri Kovalsky OK, thanks!
    5:13:34Jean Morissette Congratulation hmichel =)
    5:13:51Jiri Kovalsky Now the most interesting part. Definition of the process itself.
    5:14:04amelongo We have a manager....
    5:14:33anuradha @Jiri definitely
    5:14:52mister__m i suggest we assign issues to ourselves, add a netfix keyword and provide a patch
    5:15:07Jiri Kovalsky So, my idea was that there would be a Wiki page with easy-to-fix bugs, you would put your name into "Owner" column, and once you provide a patch you would mark "Status" of the bug as "Patched".
    5:15:12mister__m once we coordinate with module owner
    5:15:14anuradha let's manage wiki page too
    5:15:21mister__m easier to track
    5:15:35mister__m so i guess we ask the module owner for permission to take over
    5:15:36tusharvjoshi wiki seems easy and I agree
    5:15:40rajivderas wiki would be helpful
    5:16:03hmichel first of all, we need to have a place (wiki is a good place) where we can enumarate issues 'good to have' fixed by community
    5:16:22Jean Morissette Jiir, you would feel the bugs on the wiki page I presume?
    5:16:22Jiri Kovalsky I understand that Wiki page mainly as a source of bugs and also as a one page proving we are progressing.
    5:16:36mister__m btw, not sure everyone knows, but michel is a voc (voice of community) representative
    5:16:40anuradha Jiri please make sure to coordinate module owners (maven owner is no problum I have commit rights too )
    5:16:45mister__m so he could enumerate things developers are concerned
    5:16:48hmichel maybe with status whiteboard in issuezila, running a query on issuezila and generating this page automatically
    5:16:51Jiri Kovalsky You know I will have to show something to our management... :-)
    5:16:54mister__m i mean, external developers, nb users
    5:17:40mister__m @jiri: an issuezilla query would be a good report
    5:17:44mister__m based on the keyword
    5:17:48mister__m no?
    5:17:58Jiri Kovalsky By "coordinate module owners" do you mean to let them know such thing is coming and give them heads up about upcoming pathes?
    5:18:07Jean Morissette I like the idea to have a wiki that reflect our progress
    5:18:32hmichel I think there are two points here
    5:18:35anuradha yep if not guys will lose the mortivation
    5:18:58amelongo @Jiri: I like the idea; but then we would need to define the module owners first.
    5:18:58hmichel when we talk about issue, it includes enhancements and bugs, just to make things clear to everybody
    5:19:04mister__m @jiri: we need to know if the team isn't working on something already
    5:19:06Jiri Kovalsky Well, it's not formatted properly. One would need to post process it. @Mister_m
    5:19:25hmichel or am I wrong?
    5:19:36mister__m i see jiri, but it would always be up-to-date
    5:19:50anuradha hey guys I have to go sorry
    5:20:03Jiri Kovalsky Bye Anuradha!
    5:20:07anuradha let me know what you guys agreed
    5:20:12anuradha bye
    5:20:18rajivderas sorry guys yup i also have to go
    5:20:18Jean Morissette bye =)
    5:20:32Jiri Kovalsky Bye Rajiv!
    5:20:41hmichel bye guys
    5:20:50amelongo @hmichel: if I understand, we will be concerned with bugs fixes ONLY; not enhancements
    5:21:13mister__m i want to work basically on enhancements :-P
    5:21:21mister__m small ones, but enhancements
    5:21:37mister__m most of them have been filed already (and not by me, btw :-P)
    5:21:52mister__m such as bean property refactoring
    5:21:54hmichel I don't think so, we have some issues at VOC that I really wish to see solved, as some refactorings not supported for example
    5:21:56Jiri Kovalsky I still think the Wiki page is useful because if you don't know what to do, you just go there and pick one. Besides it would also show that somebody works on something.
    5:22:07Jiri Kovalsky So we don't end up in a conflict.
    5:22:10mister__m then we create a netfix-candidate
    5:22:14mister__m keyword
    5:22:16mister__m no?
    5:22:38mister__m i guess even if we have the wiki page, issuezilla tracking would be desirable
    5:22:57hmichel @Jiri: I think wiki and issuezilla should work together
    5:23:03Jean Morissette agree
    5:23:09hmichel one workflow possible is something like that
    5:23:25Jiri Kovalsky @hmichel: Right. Of course, that's what we already agreed on: to have a netfix keyword.
    5:23:39hmichel people can mark a issues as candidate, and the module owner can say if it is possible or not
    5:23:55hmichel somethink like we have for patches
    5:24:06mister__m so
    5:24:10mister__m the process would be:
    5:24:14hmichel netfix-candidate, netfix-fixable, netfix-fixed, netfix-verified
    5:24:19mister__m we would create a netfix-candidate keyword
    5:24:30hmichel and we have wiki pages to track these
    5:24:33mister__m then assign it to any issues we want to work on
    5:24:36mister__m and update the wiki
    5:24:40Jiri Kovalsky The problem with "small enhancement" is that if somebody contributes such thing he MUST fix the feature at least in the current and next release!
    5:24:59hmichel and we, fixers, chosse anyone and start to work
    5:25:11Jiri Kovalsky I don't think that we need so many keywords.
    5:25:13mister__m i would mainly contribute editor hints and beans related functionality
    5:25:26hmichel @Jiri, it is just an idea
    5:25:32mister__m module owner would accept contribution
    5:25:38mister__m by moving to netfix
    5:25:51hmichel we need to work together with nb team, that is the idea
    5:25:55mister__m then someone (the one who proposed it or another participant) accepts it
    5:26:10mister__m and assign the issue to themselves
    5:26:13mister__m and update the wiki
    5:26:18mister__m so wiki would have:
    5:26:22mister__m proposed
    5:26:25mister__m accepted
    5:26:29mister__m assigned
    5:26:33mister__m patched
    5:26:36Jiri Kovalsky netfix_candidate is okay in my opinion or simply netfix. netfix_patched would mean one of you provided patch. Status of the issue would say the rest.
    5:26:48mister__m fixed
    5:27:04mister__m what do u think jiri?
    5:27:27mister__m @jiri: module owners need a way to say they agree with the issue being assigned to netfix
    5:27:35Jiri Kovalsky I don't have a problem with more detailed status info on the Wiki page. I just don't want to polute IssueZilla with too many new keywords.
    5:27:36hmichel as my experience with nb teach me, we always need the bless from module owner for many things, and this time it will be necessary
    5:27:50mister__m so jiri, in iz we would have:
    5:27:57mister__m netfix-candidate and netfix
    5:28:07Jiri Kovalsky @hmichel: you are absolutely right.
    5:28:08mister__m netfix-candidate: we want to take it over
    5:28:18mister__m netfix: it's on our hands
    5:28:36mister__m wiki and issue status would tell the rest
    5:29:03hmichel and issue owner should be good
    5:29:20hmichel keeping the module owner as CC
    5:29:58Jiri Kovalsky I don't think we need the second "it's in our hands" - that's why every issue has "Assigned to" field.
    5:30:17mister__m it's in our hands mean
    5:30:33Jiri Kovalsky In my opinion we only need to distinguish issues which are "free" for NetFIX.
    5:30:35Jean Morissette Or we could have only netfix keyword and just let hmchel /Jiri know what issue we want to work, which will make the link with the module owner?
    5:30:36mister__m someone from netfix can assign it to himself
    5:30:43mister__m exactly
    5:30:47mister__m that's what i mean
    5:30:53mister__m candidate: asking for permission
    5:30:54hmichel I understand the jiri point and agree
    5:30:58mister__m netfix: permission granted
    5:31:32hmichel jmori +1
    5:31:45hmichel just put us as CC and add the keyword
    5:31:57Jiri Kovalsky Yes, that's it.
    5:32:39Jiri Kovalsky Module owner will get an e-mail about it and if he agrees he will assign ownership to the requestor.
    5:33:18Jean Morissette +1
    5:33:41amelongo @Jiri: I think it's the right solution.
    5:33:57mister__m requestor doesn't always seem right
    5:34:05mister__m since we need a pool of issues
    5:34:09Jiri Kovalsky So there will be basically two ways: Wiki page with already "free to fix" issues and if one of you wants to work on a particular issue s/he would add "netfix" keyword and in a comment ask module owner for taking over the ownership.
    5:34:15tusharvjoshi people, I also have to leave, will catch up with what ever process is decided, and I am ok with the process discussed till now
    5:34:21mister__m for those willing to work on whatever shows up that they think they are up to
    5:34:31Jiri Kovalsky Great, thanks for joining us Tushar!
    5:34:38mister__m ok, same idea, different methodology
    5:34:43tusharvjoshi left the conversation
    5:34:57Jiri Kovalsky So do you agree with my proposal Michael?
    5:35:06mister__m the main idea yes
    5:35:15mister__m the question is:
    5:35:23mister__m what is more practical?
    5:35:35mister__m if we add a candidate keyword to issues
    5:35:44mister__m michel could simply run a query to update the wiki
    5:35:51mister__m see?
    5:36:17mister__m (and yes, michel, i am implying the wiki page is your responsibility :-P)
    5:36:52Jiri Kovalsky 4 keywords does not guarantee that it will work. We must make sure that communication between NB developers & NetFIX developers will work.
    5:36:55hmichel why not run a query in background and update wiki automatically?
    5:36:58mister__m not 4
    5:37:00mister__m just 2
    5:37:08mister__m netfix-candidate and netfix
    5:37:17mister__m candidate - waiting for module owner approval
    5:37:22mister__m netfix - managed by us
    5:37:37Jiri Kovalsky Wiki page will contain more information that IZ query will provide.
    5:37:44mister__m such as?
    5:38:14Jean Morissette A good example of such a wiki is SwingX project
    5:38:26mister__m @jean: link?
    5:38:28Jiri Kovalsky Well, I can discuss this internally but I already know that we prefer having as little new keywords as possible.
    5:38:39mister__m is two too much?
    5:39:04Jean Morissette I will send a link, just a second
    5:39:27Jiri Kovalsky I had problems getting even 1 keyword for VOC issues, Michael!
    5:39:38mister__m ouch
    5:39:40mister__m k
    5:40:20Jiri Kovalsky Anyway, I understand it and will have a talk about it with some decision makers, ok?
    5:40:34Jean Morissette I still prefer having only one keyword and just using email to request issue ownership. It's simple.
    5:41:05mister__m the question jean is: does the module owner agrees? how can we tell?
    5:41:38Jiri Kovalsky Yes, me too. It has only one disadvantage: you can't find out how many issues are really in NetFIX hands by a simple query to IssueZilla.
    5:41:39amelongo @Jiri: are you going to send us an email with the list of modules so that each of us  can decide which issue s/he wants to be owner of?
    5:41:47Jean Morissette He just set the issue owner to netfix. No?
    5:42:21mister__m jiri kind of said "no netfix iz user"... or have i missed sth?
    5:42:51Jiri Kovalsky Yes, that would be also possible.
    5:42:58Jiri Kovalsky I mean Jean's idea.
    5:43:20Jiri Kovalsky netfix would mean it's in NetFIX team's hands.
    5:43:32mister__m i like it
    5:43:38Jean Morissette Adn if you want to know who in particular has which issue, you look on wiki
    5:43:42mister__m was my original idea
    5:43:44Jiri Kovalsky Wiki would contain "free to netfix" bugs without netfix keywords.
    5:43:53mister__m k
    5:43:55mister__m great
    5:44:07Jiri Kovalsky Great, so we end up with only 1 keyword.
    5:44:14hmichel good
    5:44:33Jiri Kovalsky Should not be hard to convince my peers about it.
    5:44:49Jiri Kovalsky OK, so in my opinion we are done.
    5:44:58mister__m +1
    5:44:59Jiri Kovalsky Does anybody have a question about anything?
    5:44:59Jean Morissette What next?
    5:45:01mister__m lunch time :-P
    5:45:16amelongo @Jiri: you haven't answer my question...
    5:45:18Jiri Kovalsky Next steps: I will create the mailing list.
    5:45:22mister__m (it's 1:45 pm here in brazil, meeting started @ 12 pm)
    5:45:33Jiri Kovalsky I will create the Wiki page.
    5:46:17Jiri Kovalsky @melongo: Oh yes, you will see in the Wiki which bugs are available, together with name of module they belong to.
    5:46:26hmichel @amelongo: we will have a wiki with issues available to be fixed
    5:46:34Jiri Kovalsky Yep.
    5:46:54amelongo and when are we going to choose ownership?
    5:47:14Jiri Kovalsky I will also summarize output of our discussion into another Wiki describing the intended process.
    5:47:38Jiri Kovalsky Once it's created and the issue is available you can take it.
    5:47:52Jiri Kovalsky I expect this to be done later next week.
    5:47:59amelongo I say module's ownership
    5:48:12Jean Morissette Can we pick any issue from issuezilla instead? Do we really need to have netfix fixable list of issues?
    5:48:20mister__m there will be module ownership
    5:48:31Jiri Kovalsky Module's ownership?
    5:48:33mister__m @jean: we say we want to fix a issue
    5:48:36Jiri Kovalsky Not yet.
    5:48:39mister__m module owner approves it
    5:48:45mister__m then we can fix it
    5:48:58mister__m netfix list is basically a pool
    5:48:59Jean Morissette alright, this is what I wanted too
    5:49:11hmichel @jmori: just to track this
    5:49:12mister__m for those willing to work on anything that meet their skills
    5:49:18amelongo We decided there would be a module owner who will grant request. My question is how would I be a module owner?
    5:49:28mister__m there is a module owner already
    5:49:30hmichel all moduesl already had a owner
    5:49:31mister__m the nb module owner
    5:49:35mister__m that was the intent
    5:49:35hmichel an nb developer
    5:49:57Jiri Kovalsky Is it clear Melongo?
    5:49:59mister__m Jiri, u forgot to mention creating the netfix iz user
    5:50:08amelongo yes, it's clear now
    5:50:16Jiri Kovalsky No, I will not create the netfix user.
    5:50:19Jean Morissette You can only be owner of an issue, which the Sun developer module owner will grant you.
    5:50:23mister__m ?
    5:50:28Jiri Kovalsky Jean, did you mean to create this user?
    5:50:38mister__m why not?
    5:50:49Jean Morissette This is what I had in mind, yes
    5:51:01mister__m i guess we said the netfix user would mean it's in the pool
    5:52:07Jean Morissette This pool is not clear for me
    5:52:31Jiri Kovalsky Once again: you want issue #423, you add a comment saying: "Hey, I know how to fix this and there is a strong demand in community. Do you grant me right to take over this issue?" Module owner will think it over and if he replies "Yes" than you or s/he assigns this issue to you and adds "netfix" keyword. That's all.
    5:52:53Jean Morissette yes
    5:53:02amelongo @jmori: the pool means NetFix has taken ownership...
    5:53:02mister__m but didn't we agree to form a pool of issues for those willing to work on anything?
    5:53:03Jean Morissette This is how I see it too
    5:53:07mister__m the candidate stuff?
    5:53:10Jiri Kovalsky Why do we need "netfix" user?
    5:53:16mister__m to say:
    5:53:19mister__m hey
    5:53:24mister__m this needs fixing
    5:53:37Jean Morissette I dont think taht we need that
    5:53:39mister__m i might not know how to do it, but guess it is a good idea we do
    5:53:50Jiri Kovalsky @Jean: yes, exactly!
    5:54:01mister__m if we only rely on comments
    5:54:13mister__m issues pending module owner approval are not trackable
    5:54:17Jiri Kovalsky That's why we have the mailing list.
    5:54:42Jiri Kovalsky You can't simply put "netfix_candidate" to mark an issue and somebody else please fix it.
    5:54:56mister__m k, ic and agree
    5:55:04Jiri Kovalsky Great!
    5:55:12amelongo @Jiri: what if the module owner is on vacation or something...what should the team do then?
    5:55:15Jiri Kovalsky Anything else guys?
    5:55:29Jiri Kovalsky Escalate it to Michel or directly to me.
    5:55:46Jiri Kovalsky And I will make sure you get or not the permission.
    5:55:49amelongo Ok. No issue on my part.
    5:55:56Jiri Kovalsky Fine, others?
    5:56:00mister__m fine
    5:56:03hmichel no more for today
    5:56:09hmichel everything is fine to me
    5:56:11Jean Morissette This is good for me. Our process sounds simple, which I like.
    5:56:33Jiri Kovalsky Cool, now you see Michael why meetings are faster? :-)
    5:56:50mister__m no, it took 2h of my time and i'm starving
    5:57:04Jiri Kovalsky OK, sorry for delaying you then.
    5:57:11Jean Morissette It's normal for the first meeting
    5:57:25mister__m just kidding guys (to a degree...)
    5:57:31hmichel ok, lunch time... see you guys.
    5:57:41Jiri Kovalsky Great, so thank you very much for joining us today and your active input in today's kick off brainstorming!
    5:57:42Jean Morissette Thanks guys. Talk to you more on teh mailing list =)
    5:57:53mister__m cya @ the ml
    5:58:00amelongo Thanks. Bye
    5:58:03Jiri Kovalsky OK, so see you later on the mailing list. Bye!
    5:58:42Jean Morissette left the conversation
    5:58:53mister__m left the conversation

    $Date: 2009/10/29 16:51:07 $ GMT

    Project Features

    About this Project

    QA was started in November 2009, is owned by Jiří Kovalský, and has 104 members.
    By use of this website, you agree to the NetBeans Policies and Terms of Use (revision 20160708.bf2ac18). © 2014, Oracle Corporation and/or its affiliates. Sponsored by Oracle logo
    Please Confirm