#title #ubuntu-meeting Meeting Meeting started by jono at 18:00:20 UTC. The full logs are available at http://ubottu.com/meetingology/logs/ubuntu-meeting/2012/ubuntu-meeting.2012-06-28-18.00.log.html . == Meeting summary == ''LINK:'' https://wiki.ubuntu.com/Accomplishments/Specs/Mobile (jono, 18:51:24) Meeting ended at 18:55:26 UTC. == Votes == == Action items == * (none) == People present (lines said) == * jono (178) * cielak (66) * gepatino (13) * janosTheHun (7) * philipballew (5) * meetingology (3) * notgary (2) * hallino1 (1) * highvoltage (1) == Full Log == 18:00:20 #startmeeting 18:00:20 Meeting started Thu Jun 28 18:00:20 2012 UTC. The chair is jono. Information about MeetBot at http://wiki.ubuntu.com/meetingology. 18:00:20 18:00:20 Available commands: #accept #accepted #action #agree #agreed #chair #commands #endmeeting #endvote #halp #help #idea #info #link #lurk #meetingname #meetingtopic #nick #progress #rejected #replay #restrictlogs #save #startmeeting #subtopic #topic #unchair #undo #unlurk #vote #voters #votesrequired 18:00:36 hey folks, and welcome to the very first Ubuntu Accomplishments meeting :-) 18:00:43 who is here for the meeting? 18:00:55 I am o/ 18:01:09 * cielak is here 18:01:12 "You've unlocked an accomplishment! Attending your first Ubuntu Accomplishments meeting!" 18:01:18 highvoltage, LOL! 18:01:32 well thanks folks for joining us 18:01:47 there is not really a fixed agenda here, but I have a few topics I want to cover: 18:01:49 0/ 18:02:05 * getting more accomplishments written 18:02:12 * getting more translation involvement 18:02:14 * testing 18:02:20 does anyone else have any topics? 18:03:09 ok then :-) 18:03:30 maybe we can cover translations first 18:03:52 so we have four projects -daemon -viewer -community-accomplishments and -desktop-accomplishments 18:03:57 and then also the web gallery 18:04:08 I remember it was suggested that we bound some official translation groups to our project 18:04:12 right now our translation coverage in the desktop app and daemon is pretty good 18:04:20 but the accomplishments themselves is limited 18:04:23 cielak, indeed 18:04:36 especially desktop ones, I wonder if they are available in more than 2 langs 18:04:37 Me :) 18:04:42 any thoughts on how we get more translators involved? 18:04:57 it would be great if we had a translations leader who can help grow this community 18:05:01 cielak, yeah 18:05:42 is anyone interested in helping out with this? 18:05:55 I wonder whether we can't simply ask one of these translator organisations that translate everything in LP 18:06:17 like the Launchpad Translators 18:06:21 cielak, like the Ubuntu translations groups? 18:06:27 so maybe the next step is reaching out there 18:06:37 I know some translators express concern about open permissions too 18:06:42 so we might need to resolve that 18:07:07 we've had recently a case of translation mistake of griefing that coused a critical bug in viewer 18:07:37 right 18:07:54 ok, I will reach out to the translations groups to see if they can help us 18:08:05 if anyone is passionate about this, do let us know 18:08:15 we could definitely use some help here :-) 18:08:18 * gepatino is here 18:08:29 hey gepatino, just the person 18:08:56 gepatino, do you think it would be viable to provide a link in the web viewer so people can be linked to where they can translate opportunities? 18:09:04 this might be a good way of getting more folks involved 18:09:05 How about writing some translation accomplishments -"You have make your first translation", "You have made 10 translations", "You have made 20 translations", etc. I have actually been thinking it would be cool to track Launchpad activity for acheivements, such as "You have merge X patches into Ubuntu". Perhaps this could be done for translations 18:09:25 notgary, we would love to have that, but translations doesn't have an API in LP afaik 18:09:35 notgary: the problem with translations is that Launchpad API does not expose access to them 18:10:00 jono, adding a link shouldn't be an issue 18:10:17 should we check some rules? redirect so some specific language, etc? 18:10:24 gepatino, cool - we would need to figure out how to generate a link to the correct strings, but we can look into that 18:10:32 notgary: and the problem with number-based accomplishments is that they often encourage pointless traffic 18:10:35 gepatino, I am not sure of the details 18:11:04 ok, so lets move on 18:11:10 next I wanted to discuss testing 18:11:17 cielak and I have been discussing this recently 18:11:30 i was catching up... I meant it would be a problem to add link, thinking in linking to launchpad translations 18:11:44 our codebase is growing, and so are our users, and thus the potential for bugs could increase 18:11:44 then I've read about the translation groups not being open... 18:11:46 gepatino, cool 18:12:18 gepatino, well, open in terms of whether contributions are reviewed 18:12:37 I think we should get advice from our translation community about how the translations are best governed 18:12:49 * cielak agrees 18:12:49 ok 18:12:56 so in terms of testing 18:13:10 I would like to build unit test suites for all of our projects 18:13:16 I know the web team are already working on this 18:13:33 and I am planning on putting in place unit tests for the daemon 18:13:45 I suspect the viewer is important but less critical than the daemon 18:13:56 if the daemon gets it wrong, all viewers are screwed :-) 18:14:02 hey janosTheHun 18:14:13 hey jono ! 18:14:13 viewer is just a bunch of GTK+ hacks, the real code is in the daemon ;) 18:14:22 cielak, indeed :-) 18:14:29 the other part of the testing which I wanted to discuss was how we test the server 18:14:35 again, cielak and I discussed this a little earlier 18:14:47 * janosTheHun here now, but won't be long 18:15:12 so while we have seen good traffic on the validation server, which suggests things are generally working well, there are sometimes issues, and these could be either U1 syncing delays or bugs in the server 18:15:13 janosTheHun: semi-final? :) 18:15:20 cielak: yup :) 18:15:53 or generaly any case causing trouble with the validation process 18:15:59 there can be many more factors 18:16:08 in terms of the U1 lag, I have a solution which I think could work - I will set up a U1 user who will generate some files and put the timestamp inside the file - when the file is synced we can compare the timestamps of the server to the file and get an idea of lag 18:16:29 I think having visibility on U1 lag will help us in tracking down some issues 18:16:48 such file would be sent like each 10 minutes, or daily? 18:16:48 I would also like the server testing to dynamically create and register shares and check on the success of that too 18:16:57 cielak, I am thinking every 10 mins or so 18:17:03 and then plot this into a a graph 18:17:09 so we can compare when a bug occurs to the lag time 18:17:21 right 18:17:22 bug X happened on 5th June, and oh look...U1 was lagging :-) 18:17:57 in terms of ensuring the server is actually validating trophies correctly, I think the first step is probably unit tests 18:18:19 well, most our problems with validation server is not just lag, but no signature at all, yet that's a good idea nevertheless 18:18:24 one challenge we have now is that if a user types in the wrong identification it will constantly fail 18:18:34 cielak, right 18:18:45 so a trophy not getting signed means either: 18:19:10 (1) the user is screwing around with .trophy files 18:19:20 (2) there is a bug in the code that uploads a trophy 18:19:37 1) is unlike, if someone intentionally messes things up, they won't report a bug 18:19:54 (3) the user entered extra-information that generated the .trophy and then changed it after it was synced, s when the server validates it the extra-info doesnt work 18:20:14 yeah I think few people, if any are faking trophies 18:20:45 (4) there is a bug in the code that validates the trophy 18:20:45 3) is not valid, daemon will regenerate the trophy with new extrainfo, if it wasn't yet signed 18:21:01 gepatino, oops, yes 18:21:04 faking does not give you that good feeling that earning it does. 18:21:45 cielak, right, but imagine this: the user adds e-a, it gets approved and generates a .trophy, that gets uploaded, they then change their e-i and U1 doesnt sync it yet 18:21:51 so, the real programming bugs seems to be (2) and (4), am I right? 18:21:53 philipballew, indeed 18:22:03 gepatino, yup 18:22:34 jono: this happens just once, the next time the .trophy is sent it will be correct, and the server will sign it then 18:22:37 or will it not? 18:22:59 cielak, agreed, I just mean that it is not inconceivable that there could be a mismatch 18:23:01 if I modify my .trophy and send a new version, will the server re-sign it? 18:23:06 I do think we have some bugs in there somewhere 18:23:25 cielak, if you already have the .asc, the server wont resign 18:23:34 if you don't have the .asc it will try to sign 18:23:43 each time I upload a new version? 18:23:50 or just then the file is created? 18:24:10 cielak, each time the file is updated 18:24:18 okay, that's correct then 18:24:22 if you modify a file in U1 that is in a subscribed folder, it gets synced 18:24:35 so 3) is not really a problematic case 18:25:53 cielak, right, I think the mismatch scenario is pretty rare anyway 18:26:00 indeed 18:26:05 it will result in a logged failure on the server, but then just resolve itself 18:26:11 we better seek for bugs in the code :) 18:26:14 what we need to identify is what is the source of the failures 18:26:18 cielak, totally agree 18:26:41 cielak, we just need to do more testing and find failures 18:26:49 i recommend writing unit tests around this issue, and try to cover all the corner cases you can think of 18:26:54 right now I think there is a lot we can do to get better visibility on failures 18:26:59 janosTheHun, agreed 18:27:24 janosTheHun, I was saying before you joined that I would like us to have unit tests for all of our projects 18:27:33 and before we land code we can run the test suites 18:28:40 jono: yup, good idea 18:28:51 we might also think to include some easy access to debug data 18:28:59 the daemon log file is working great 18:29:02 cielak, what kind of debug data 18:29:08 for bug reports it's really useful 18:29:21 but there are other things we ask all users that report a bug 18:29:44 cielak, I agree we might want to build some better debug tools into our software 18:30:00 like the share data, whether it got accepted, what files are in trophies directory 18:30:03 one thing that could be useful for example is an easy way to see if a share is active 18:30:07 yeah 18:30:30 cielak, maybe we could add this to battery? 18:30:36 add a switch to summarize system info 18:30:47 accomplishments-battery -i for example 18:30:56 you have recently implemented support for getting our share data from the u1syncdaemon, what if we simply printed the result to the daemon log? 18:30:57 and it displays the share status and other reporting 18:31:22 I'd integrate it with either daemon or viewer, so that when one reports a bug, we do not need to ask him to install the battery 18:31:27 cielak, we could do that 18:31:33 makes sense 18:31:39 I am happy to take a look into that 18:31:47 should be a simple addition :-) 18:31:52 actually, there is some bug in determining share ID, found it recently 18:32:00 cielak, oh? 18:32:24 will need to take a closer look, but it makes me unable to publish my trophies :) (the shareid in URL does not match my actual shareid) 18:32:38 cielak, I think I know this bug 18:32:45 this might be because you have two active shares 18:32:50 nope, just one 18:32:54 really? 18:32:55 odd 18:33:04 yeah, will report it & investigate 18:33:07 cool 18:33:09 speaking of which.... 18:33:14 I have another suggestion 18:33:21 I would like to suggest we have two log files 18:33:24 daemon.log 18:33:29 and scriptrunner.log 18:33:38 the daemon is getting spammed with all the checks for accomplishments 18:33:40 aaah, that's wise 18:33:42 which makes it difficult to read 18:33:45 is there a wiki page about running the unit tests in the daemon project? 18:33:55 janosTheHun, not yet, they don't exist 18:34:09 janosTheHun, we have an original set of tests that hasnt been touched since January 18:34:14 I need to go in and update them 18:34:26 when I get a few working we can work together to build out full coverage 18:34:46 cielak, would you be happy to split the daemon log into the two files? 18:35:15 we may even want to have a log file standard info such as the share, share id, if it is active, trophydir etc 18:35:21 so these daemon logs: 18:35:29 * daemon.log - general run time daemon info 18:35:43 * scriptrunner.log - log of when accomplishments are checked 18:36:10 * environment.log - a list of settings in the environment (e.g. share id, name, trophy dir, collections installed etc) 18:36:44 well, daemon.log will contain very little information, we might merge environment.log into it 18:37:09 cielak, right, I was just thinking you would need to hunt it our in daemon.log 18:37:16 whereas environment.log can just summarize 18:37:26 I am happy with whatever you prefer though 18:37:47 maybe I'll try both, and compare which one makes more sense 18:38:07 we can have several instances of some 'logging' object, that would write to a single file 18:38:09 cielak, awesome :-) 18:38:14 indeed 18:38:21 this will give us good visibility when people have issues 18:38:31 cielak, we may want to consider this as a push into 2.1 18:38:40 depending on how invasive it is 18:38:51 by the way: we have not created separate series for 0.2 18:39:16 thus we cannot separate 0.3 additions from 0.2.1 ones 18:39:33 cielak, oops 18:39:36 I will do that 18:40:07 I will create the series, but we should probably only push critical fixes to 0.2.x 18:40:12 so lets maybe do this in 0.3 18:40:18 I will create the series today anyway 18:40:25 yeah, will be more time to test all that 18:40:34 ok i guys i have to go, will try to do something for the daemon and then ping you 18:40:41 cool 18:40:48 laters janosTheHun! :-) 18:40:49 thanks! 18:40:51 thanks janosTheHun, see you! 18:41:20 so anything more to discuss on testing? 18:42:17 * cielak just pushed a tiny fix that significantly cleans up demon's log ;-) 18:42:28 cielak, wow, you are fast, buddy :-) 18:42:48 ok, so the final topic I wanted to discuss was growing our community of accomplishments writers 18:42:53 I am a little behind on this 18:43:06 I am planning on putting together documentation and a video for how to get people involved 18:43:18 it may be worth reminding that we have lots of forum accomplishments waiting 18:43:21 but so far I have been focused on a few other things such as the branches 18:43:31 cielak, oh we should look into that 18:43:43 and s-fox will be surely interested in developing them, but Canonical did not reply for some longer time 18:44:02 yeah Canonical IS has not responded yet since his last email 18:44:07 I will see if I can ping them to respond 18:44:33 I also need to update the docs as battery now checks for missing fields in .accomplishment files 18:44:37 might be worth it, I can't imagine 0.3 release without forum accoms! 18:44:43 indeed :-) 18:44:55 there is one more problem with the docs 18:44:59 oh? 18:45:08 someone had problems with following the guide recently 18:45:18 the problem is with the order of chapters 18:45:26 I guess battery information is too soon 18:45:59 it assumes one has already installed the collection, that is working on a bazaar branch etc 18:46:16 ahhh 18:46:22 so basically it requires knowledge from upcoming chapters 18:46:34 we should ask people to file bugs in ubuntu-accomplishments for docs related issues 18:46:42 cielak, could you ask that person to file a bug? 18:46:50 * cielak wanted to re-read the whole guide to determine causes, but hadn't yet time 18:46:58 good idea 18:47:00 jono: sure, will try to 18:47:09 and as we get more contributors involved it will help us spot other issues 18:47:13 ok cool 18:47:22 well I have covered everything I wanted to discuss today 18:47:26 anything else? 18:47:42 wasn't it philipballew? 18:48:12 philipballew: are you still with us? ;-) 18:48:15 yes! 18:48:55 awesome! would you mind reporting a bug in ubuntu-accomplishments concerning the troubles you had with following the guide on creating accomplishments? 18:49:17 Yes. I will. 18:49:23 great thanks! 18:49:51 ok, I guess we can wrap 18:49:53 maybe someone lurking has any questions concerning accomplishments? 18:49:57 thanks folks for joining our first meeting! 18:50:02 yeah, any questions? 18:50:14 and remember that we live in #ubuntu-accomplishments 18:50:16 is there any roadmap for web gallery to 0.3? 18:50:29 or so far only the specs in the wiki? 18:50:51 I am considering a simple video tutorial on making some for people who are not good at reading and understanding as they would be in they could see. 18:51:10 gepatino, just the wiki right now 18:51:14 but part of it is the Mobile spec 18:51:17 did you see that gepatino? 18:51:24 https://wiki.ubuntu.com/Accomplishments/Specs/Mobile 18:51:34 this is basically a stylesheet for the web gallery 18:51:50 philipballew, that would be great 18:52:03 my main advice is ensure that you have written a few accomplishments first before you do a tutorial 18:52:19 someone else did a tutorial and they had not written an accomplishment and it wasnt so good 18:52:21 i've read that, jono, thanks 18:52:26 gepatino, cool 18:53:08 and of course such new accomplishments are very welcome! :) 18:53:13 and finally, is there any release date for the web gallery? we are progressing but there are a lot of things to fix, like page layouts, check url schemas, etc 18:53:55 gepatino, release date for 0.3 is Sep 5th 18:54:04 so if we could shoot for then, that would be grea 18:54:08 * cielak has to remember it, hehe 18:54:20 it would be cool to have it in place earlier so we can get some further testing 18:54:49 sure, it would be nice to have one or two weeks before 18:54:50 hopefully we'll get trophies.ubuntu.com :) 18:55:18 indeed :-) 18:55:22 alright, lets wrap 18:55:24 thanks, folks! 18:55:26 #endmeeting Generated by MeetBot 0.1.5 (http://wiki.ubuntu.com/meetingology)