PMC Usability Meeting 20100409
Date: 2010-04-09
Time: 2PM GMT
Venue: irc #adempiere-team
Support Spreadsheet: Adempiere PMC Usability
Chat times in GMT-5
(09:00:17) CarlosRuiz: Hi
(09:02:22) red1: Hola CarlosRuiz
(09:02:47) red1: u know.. the Client Accounting debugging.. its so much faster than waiting 30 mins
(09:03:08) red1: once u told me about it, i could confirm the patch right away
(09:03:13) CarlosRuiz: yes - and there is now a way to make it post immediately in a safe way
(09:03:28) red1: now i test everything with it.. and its instant results if i code right or wrong
(09:04:00) red1: so it can work without apps server for live system?
(09:04:25) kai7: Hi
(09:04:33) CarlosRuiz: Hi kai7
(09:04:42) CarlosRuiz: yes Redhuan, it can work totally standalone
(09:05:12) CarlosRuiz: as a test I tried to set up an adempiere in a 256MB instance without server - but it needs minimum 512MB to work
(09:07:13) CarlosRuiz: so, Kai, the forum thread was posted, a couple of new suggestions were added
(09:07:48) kai7: Yes
(09:08:13) CarlosRuiz: I'm adding the last two that are collected in thread
(09:08:29) red1: title of thread? i want to check
(09:08:45) kai7: Call for usability improvements
(09:09:11) CarlosRuiz: https://sourceforge.net/projects/adempiere/forums/forum/610546/topic/3642406
(09:09:26) kai7: Did you transfer the ideas to the wishlist?
(09:10:10) CarlosRuiz: yes - and doing at this moment for the latest not transferred
(09:10:21) kai7: Great
(09:10:58) red1: i want to ask if this is useful http://www.adempiere.com/index.php/Usability
(09:11:25) red1: Carlos made good suggestion that some of them are specific and not general and so i am putting them as non trunk
(09:12:00) red1: but it caters for usability issues seen by an Oracle team in Thailand
(09:12:12) red1: they insist that users dont behave rationally
(09:12:30) red1: or shuld be allowed to be creative in accounting
(09:13:14) CarlosRuiz: done, finished the transfer
(09:14:16) kai7: OK
(09:14:18) CarlosRuiz: Redhuan, I agree that Adempiere can be more helpful to avoid users committing mistakes - and I agree that's a usability problem
(09:15:10) kai7: I just looked to the tasklist
(09:15:33) kai7: I guess 4 & 6 could be the next steps
(09:16:36) red1: i am now realising that usability means different from what i tot
(09:16:55) red1: usable as in utilities or functions?
(09:18:35) CarlosRuiz: kai7: agree
(09:18:35) CarlosRuiz: red1: usability is about making the user experience easier, safer, richer - some of your suggestions on usability page share such goal
(09:19:58) kai7: How to start?
(09:20:09) kai7: With another thread in forum
(09:20:23) kai7: We know some improvments
(09:20:29) kai7: I guess we should collect them
(09:20:35) kai7: How could this be done?
(09:20:48) red1: can find the wishlist.. in the idea torrent is only one idea about auto filter
(09:21:06) CarlosRuiz: red1: in usability google spreadsheet
(09:21:42) kai7: red1 :
(09:21:42) kai7: http://spreadsheets.google.com/pub?key=tQ2kb6ztizwZQmIrxFOTqUA&gid=1&hl=en
(09:21:50) CarlosRuiz: kai7: here comes the same problem again :-) how to make implementors contribute? how to ease the contribution path? how to encourage contributors over non-contributors?
(09:21:55) red1: ok danke schon
(09:22:06) kai7: bitte sch?n :-)
(09:22:40) CarlosRuiz: what are the stoppers? I refuse to think is just selfishness - there are others
(09:23:10) red1: 20. Implement User Preference table is interesting.. i think it cna house alot of the Doc control i am focusing on
(09:23:36) kai7: I guess we should collect them activly
(09:23:45) kai7: There are not so many:
(09:23:47) red1: CarlosRuiz: is that the issue? Contributing?
(09:23:51) CarlosRuiz: I had an interesting chat with Allan yesterday about on Functional meeting
(09:23:57) kai7: We have 3 metas
(09:24:00) kai7: 1 Schaeffer
(09:24:12) kai7: 1 e-evolution
(09:24:14) red1: fortunately only 1 red
(09:24:26) kai7: :-)
(09:24:43) kai7: The bigger problem: We have to review it.
(09:24:52) CarlosRuiz: review it for?
(09:25:22) kai7: To see if we could drop the point from the list as done.
(09:25:35) red1: i was talking to my new client about Oracle features.. and they say AD is more flexible and user friendly
(09:25:48) red1: just that its too open in doc control
(09:26:02) kai7: red1 but it doesn't mean we couldn't do better ;-)
(09:26:04) CarlosRuiz: kai7: don't understand, sorry - review it for what?
(09:26:44) kai7: To see if this is already done (and to push it to trunk and/or release)
(09:27:06) red1: and i discovered a new feature in AD.. the processmsg pop up for return false.. i can improve it easily
(09:27:17) red1: with more meaningful msgs.. and dynamic reference
(09:27:22) CarlosRuiz: what if you simply publish it - without any extra work - let others to the effort if they like it
(09:27:54) red1: but i need carlos advice.. if such dyanmic msging can it go to trunk right away?
(09:28:02) kai7: Well, but we have it on our list. :-)
(09:28:09) red1: many error msg are listless
(09:28:51) kai7: The question is: what is the job of this "group"
(09:29:06) red1: facilitate usability and user experience issues?
(09:29:56) kai7: And how should this be done?
(09:30:03) red1: it shuld be more user oriented
(09:30:13) CarlosRuiz: I proposed to adopt these (borrowed from KDE):
(09:30:13) CarlosRuiz: * Educating developers about usability practices
(09:30:13) CarlosRuiz: * Conducting usability activities to help gather feedback from users
(09:30:13) CarlosRuiz: * Working with projects and help design usable user interfaces
(09:30:16) red1: listing of user issues rather than merely developer POV
(09:30:34) red1: GW template and sample data
(09:31:39) red1: we inherited something very complex and presented in lacking manner (still in spite of great GW)
(09:32:44) CarlosRuiz: kai7: not sure if you answered this one (mixed conversations problem) :-)
(09:32:44) CarlosRuiz: what if you simply publish it - without any extra work - let others to the effort if they like it
(09:32:44) CarlosRuiz: (just asking because I'm trying to realize which are the stoppers - and trying to avoid them)
(09:34:02) kai7: Well, I am thinking about what to do next in this group
(09:34:35) kai7: If we say: Publish your contribution and let other do the review work,
(09:34:40) kai7: what is our job then?
(09:35:05) kai7: "Educating developers" sounds great.
(09:35:08) kai7: But how?
(09:35:11) CarlosRuiz: nope, not our job - developer's job
(09:35:31) kai7: Also the review?
(09:35:34) CarlosRuiz: but if it's not published - then there's nothing to review
(09:35:45) kai7: OK, good point. :-)
(09:35:54) kai7: I could take care of this.
(09:36:00) kai7: No problem
(09:36:01) red1: i support the listing of needs and let others refer to it to help
(09:36:23) red1: listing can invite more interest
(09:36:30) CarlosRuiz: the issue is I don't want to add extra effort to implementors to contribute their improvements - release early
(09:36:50) CarlosRuiz: metas example with role inheritance is good
(09:37:00) CarlosRuiz: they published an outdated version - not compatible with trunk
(09:37:17) red1: the sub-role?
(09:37:18) CarlosRuiz: I found the improvement so good that I made the work to make it compatible with trunk
(09:37:22) CarlosRuiz: and it was voted
(09:37:27) CarlosRuiz: and yesterday was integrated into trunk
(09:37:34) CarlosRuiz: metas and teo already did the big work
(09:37:40) red1: ok u are faster
(09:37:42) CarlosRuiz: I just did the minor
(09:38:07) kai7: OK, I guess you are right.
(09:38:09) CarlosRuiz: even there was a little point not following best practices - and I make it complain
(09:38:38) CarlosRuiz: complain is not the word ...
(09:38:44) red1: reminder
(09:38:52) CarlosRuiz: accomplish
(09:38:59) CarlosRuiz: :-)
(09:39:11) CarlosRuiz: so - my thinking is to encourage everybody to publish
(09:39:18) kai7: I will bother metas to publish
(09:39:23) CarlosRuiz: as a "release early" policy
(09:39:29) kai7: Sorry "encourage" of course. ;-)
(09:39:40) CarlosRuiz: let me check the meeting with Allan yesterday
(09:39:40) kai7: And I will tell Karsten to publish ours
(09:39:43) red1: but they can get 'rejected early' too :D
(09:39:46) CarlosRuiz: and bring a couple of phrases
(09:40:05) CarlosRuiz: red1: not in trunk - publish it in contributions folder
(09:40:25) red1: yes - in progress
(09:40:45) CarlosRuiz: complete chat with Allan is here:
(09:40:45) CarlosRuiz: http://www.adempiere.com/index.php/PMC_Functional_Meeting_20100408
(09:40:45) CarlosRuiz: let me bring some suggestions I realized
(09:41:15) CarlosRuiz: <-- snippet -->
(09:41:43) CarlosRuiz: I would think on something like this:
(09:41:43) CarlosRuiz: minimal sharing -> code + dictionary (this is easy to share)
(09:41:43) CarlosRuiz: if you want to get attention -> consider publishing some screenshots
(09:41:43) CarlosRuiz: if you want to contribute complete things -> consider publishing sample scrambled/trimmed data / also documentation can help a lot
(09:41:43) CarlosRuiz: <-- end of snippet -->
(09:42:17) CarlosRuiz: implementor can do it incrementally
(09:42:23) red1: i did that alot.. but thats half the picture
(09:42:31) red1: ppl want complete solutions
(09:42:34) CarlosRuiz: first publish code + dict
(09:42:34) CarlosRuiz: then publish screenshots
(09:42:34) CarlosRuiz: then publish docs or sample data
(09:43:21) CarlosRuiz: > ppl want complete solutions
(09:43:21) CarlosRuiz: yes, in trunk ppl want complete solutions
(09:43:21) CarlosRuiz: in contributions and branches we must apply "release early" principle
(09:44:01) red1: yes good process.. agreed
(09:44:11) vpj-cd [~e-Evoluti@187.159.117.117] entered the channel.
(09:44:24) kai7: OK
(09:44:26) red1: but it may get lost from others' attention
(09:44:38) vpj-cd: Hi guys
(09:44:47) CarlosRuiz: you can get attention with other means - i.e. screenshots
(09:44:49) red1: i tot trunk can be more relaxed :D
(09:44:51) CarlosRuiz: Hi Victor
(09:44:53) red1: hola vpj-cd
(09:45:04) CarlosRuiz: not so relaxed at the level of allowing broken things
(09:45:11) CarlosRuiz: or very underdeveloped things
(09:45:14) kai7: Hi Victor
(09:45:20) red1: its not anyone's intention :>
(09:45:37) red1: some were never complete
(09:45:41) vpj-cd: Hola Red1, Carlos , Kai , Trifon
(09:45:46) red1: like Fixed Assets
(09:45:50) red1: its still not complete
(09:46:17) red1: i can complete that this year as its part of Thai's roadmap
(09:46:23) red1: POS + Promotions
(09:46:31) red1: Master-Detail-Detail swing
(09:47:11) red1: vpj-cd: there is googledoc for discussion http://spreadsheets.google.com/pub?key=0AhBAFce33rWhdFEya2I2enRpendaUW1JcnhGT1RxVUE&hl=en&gid=0
(09:47:20) CarlosRuiz: so Kai - I guess we have good guidelines to promote tasks 4 & 6 :-)
(09:47:39) kai7: Yes, let the other do the work. :-)
(09:47:46) vpj-cd: thank a lot red1
(09:47:59) kai7: OK, what are our taks now?
(09:48:00) red1: and is the Mecurial SVN going to affect better working culture?
(09:48:53) CarlosRuiz: I feel the pressure on trunk quality made the people shy to contribute
(09:49:36) red1: but from Fer_Luck its different
(09:49:45) CarlosRuiz: we need to reaffirm that doors are open for incomplete work (or even just for draft ideas) on branches or contributions
(09:49:53) red1: i asked him and he stated (if not mistaken) that trunk is unstable to contribute
(09:50:22) CarlosRuiz: so, maybe the next step would be to make a call on forums to contribute those usability things in the "release early" (in branches) principle
(09:50:40) red1: but why not have a combined branch?
(09:50:44) red1: release early branch?
(09:50:53) red1: for experiments and more review
(09:50:55) CarlosRuiz: let others review (but don't push them - committing into trunk is pushing - committing into branch is respectful to allow others do the review if they want)
(09:51:03) red1: the sandbox?
(09:51:04) kai7: I could do this.
(09:51:18) CarlosRuiz: red1: just open it - you can do it without any permission
(09:51:34) vpj-cd: my suggestion is that we define a way to that this can be testing of other via a release aerly
(09:51:38) vpj-cd: early
(09:51:57) kai7: Let me start:
(09:52:04) CarlosRuiz: I think the release-early branch needs a test server
(09:52:16) red1: if something break others... or broken its useful
(09:52:22) kai7: Victor, we have No 24 on the wishlist
(09:52:23) red1: for development mode
(09:52:30) kai7: "Access saved queries directly from find button"
(09:52:56) red1: the auto filter?
(09:53:00) kai7: The "comment" says "This was contributed by e-evolution, but it was reverted, needs some improvements"
(09:53:03) vpj-cd: and the integration continue tools �should allow other test
(09:53:12) vpj-cd: but we need following some rules
(09:53:13) red1: the smart browse?
(09:53:23) red1: but rules can scare ppl
(09:53:45) red1: it shuld be u break it, u can still fix it
(09:53:49) CarlosRuiz: I agree with a release-early with no rules - jungle law :-)
(09:54:02) CarlosRuiz: and I would like a demo or test server for such release-early branch
(09:54:05) red1: its still a law!
(09:54:18) red1: the broken branch
(09:54:22) CarlosRuiz: the issue I foresee is - who is going to fix it when broken
(09:54:29) CarlosRuiz: but we can try
(09:54:30) CarlosRuiz: I like it
(09:54:48) red1: 1) Broken branch (remember where you come from)
(09:54:55) CarlosRuiz: :-D
(09:54:55) red1: 2) Trunk (remember how you got here)
(09:54:58) vpj-cd: the dev should know how following easy step to compile and deploy
(09:55:03) red1: 3) Release (just remember)
(09:55:27) red1: 3) Release (don't you ever forget) :D
(09:55:32) vpj-cd: I think tha tis good work in branchs
(09:55:55) red1: Maintainer of Broken Branch - vpj-cd !
(09:56:26) red1: ok when is the next baby born?
(09:57:54) CarlosRuiz: kai7: we're almost done - 2 minutes to go - is that ok then as next steps?
(09:58:15) kai7: OK, I have one now:
(09:58:19) kai7: Forum post
(09:58:28) kai7: Is there something else I misses now?
(09:58:56) CarlosRuiz: do you think we need previously to try to define better the "release early" process?
(09:59:16) vpj-cd: yes
(09:59:47) red1: i think u shuld not worry
(10:00:00) vpj-cd: can set Adempiere liberary in a maven repository
(10:00:05) CarlosRuiz: yes - meanwhile is stated how to do it - all we chatted today
(10:00:11) red1: i read from another OSS link that some projects have few players
(10:00:16) vpj-cd: and we would use this in the branches to resolve the jats
(10:00:23) vpj-cd: and automatizate the builds
(10:00:30) red1: we have done very well so far
(10:00:32) vpj-cd: and release
(10:00:42) CarlosRuiz: ok kai7, I guess it's ok - forum post - encourage publishing your usability things early in branches ... that's our next step
(10:01:07) kai7: OK
(10:01:17) CarlosRuiz: gtg
(10:01:20) kai7: BTW: I have good news:
(10:01:23) CarlosRuiz: ?
(10:01:35) kai7: We released your Front Panel Designer 4.0
(10:01:36) i-zugizo [~i-zugizo@ppp-124-120-84-249.revip2.asianet.co.th] entered the channel.
(10:01:47) kai7: So I guess I will have more time again shortly.
(10:01:51) red1: ah Schaeffer's Front Panel!
(10:01:55) kai7: Jepp.
(10:01:58) CarlosRuiz: great - congratulations
(10:02:03) kai7: If you want to download it:
(10:02:07) red1: that means we can see Karsten Thiemann
(10:02:20) kai7: http://www.schaeffer-ag.de/en/download/front-panel-designer.html
(10:02:32) kai7: For Windows, Linux and Mac now. :)
(10:02:42) red1: Kai, can we use it to replace ADempiere Swing Panel?
(10:03:10) kai7: If you want it out of aluminium... no problem :-)=
(10:03:24) red1: yes the metallic look and feel
(10:04:23) kai7: OK, have to leave now.
(10:04:28) kai7: See you later guys.
(10:04:30) CarlosRuiz: ok guys - thanks for attending
