BugZappers/Meetings/Minutes-2008-Jul-15

From FedoraProject

Jump to: navigation, search
poelcat who is in the house? 10:00
* stickster here to audit 10:00
* DemonJester 10:00
jds2001 f13: I've been reading that thread with great interest, but your idea got lost i guess :) 10:01
jds2001 i'll look again i guess :) 10:01
poelcat stickster: i thought you weren't auditing any more :) 10:01
stickster Well, I'm half-auditing :-D 10:01
jds2001 stickster is a full-on participant now :) 10:01
stickster Yeah, you caught me 10:01
poelcat hello DemonJester jds2001 10:01
* stickster does a crappy job of being a ninja 10:01
-!- giarc [i=hidden-u@gnat.asiscan.com] has joined #fedora-meeting 10:01
stickster "The Ninja Who Wore White" 10:01
* SMParrish Here 10:02
f13 jds2001: essentially triage folks who go to SIG meetings and spend a small time running through any 'easyfix' bugs that are within the SIG domain, letting the sig discuss/decide, and then doing whatever to the bug is necessary 10:02
poelcat mcepl: around? 10:02
poelcat f13: there is also the corresponding Keyword "patch" i've been putting on bugs 10:03
f13 nod 10:03
mcepl poelcat: yes 10:03
-!- fcrippa [n=fcrippa@83.225.231.234] has quit ["Leaving"] 10:03
poelcat people really do include patches 10:03
poelcat  :) 10:03
-!- comphappy [n=comphapp@c-71-227-212-6.hsd1.wa.comcast.net] has joined #fedora-meeting 10:04
-!- zydoon [n=zydoon@41.225.153.246] has joined #fedora-meeting 10:04
jds2001 and it's a real shame if they include patches and they dont at least get an ack/nack from the maintainer :/ 10:04
-!- zydoon [n=zydoon@41.225.153.246] has left #fedora-meeting [] 10:04
poelcat Should we wait for anyone else? 10:05
comphappy woke up 40 sec ago heh 10:05
jds2001 btw, ajax mentioned (and rightly so) that if he read all of his bugmail, he'd get nothing else accomplished. How to handle special cases like that? mcepl probably has the best ideas here :) 10:05
poelcat comphappy: :) 10:05
poelcat thanks for rising early 10:06
jds2001 that's right, it's 7AM for you left-coasters :) 10:06
poelcat did anyone have any additional agenda topics? 10:06
* jds2001 thought that you were taking care of the agenda :) 10:06
poelcat i did :) 10:06
comphappy I would like some input on what people want in the metrics stuff I am writing 10:07
f13 jds2001: that was my triage people to sig meeting thought to solve the 'I can't read bugmail' problem. 10:07
poelcat let's start w/ f13's topic 10:07
poelcat my only thought so far was sending people URL queries of: "help us with these bugs" 10:08
* poelcat is seeing the same problem w/ bugs set to NEEDINFO for maintainers 10:08
-!- gvarisco [n=gvarisco@66.187.233.202] has left #fedora-meeting [] 10:08
mcepl jds2001: there is a query https://bugzilla.redhat.com/buglist.cgi?cmdtype=runnamed&namedcmd=XGL%20having%20patch&namedowner=mcepl%40redhat.com giving all Xorg bugs with EasyFix or Patch keywords and then I set up whinning to make a weekly run of this query and send email which si then redirected to ajax. Except ajax discards all bugmail from bugzilla, including this one, so I had to remind him about the existence of this weekly mail ;-) 10:08
poelcat comphappy: okay we'll cover that too 10:08
comphappy so are we are taking about the mail generated by BZ? 10:09
-!- jhutar [n=jhutar@nat/redhat/x-9a33b7033b874905] has joined #fedora-meeting 10:09
poelcat comphappy: correct... there was a recent thread on f-d-l about maintainers drowning in bz emails 10:09
mcepl jds2001: not sure, how to expand it to the bugzilla-wide level (I had to split my original one query into two -- one Xorg and other Gecko -- because people complained about getting a list of bugs which doesn't concern them) 10:09
mcepl poelcat: my experience is that NEEDINFO set to maintainer is one of the best ways how to get bug lost. 10:10
poelcat oh no :( 10:10
mcepl with some people (ehm, ajax) is works like 100% 10:11
* poelcat has been going through the oldest NEW rawhide bugs and finding a lot of "do these still apply any more" bugs 10:11
mcepl (note that he is very good in going through bugs which are ASSIGNED to him) 10:11
poelcat could someone research this topic a little more, maybe post to the list with ideas and we can revisit next week? 10:12
poelcat f13: did you have any ideas on how to fix the situation? 10:13
poelcat mcepl: does the nag mail have any special mail headers? 10:13
poelcat mcepl: or wonder if we could have some added? 10:13
-!- covex [n=covex@sandbox.cz] has joined #fedora-meeting 10:13
mcepl I think it has special Subject 10:13
mcepl (you can set the Subject for your whines) 10:14
poelcat it is kind of a hard problem to solve people really don't want to receive the mail 10:14
poelcat would RSS help at all? 10:14
f13 poelcat: I think the only real solution is to A) get more qualified people interested in the software, and B) in person discussion. Email just gets lost 10:15
f13 poelcat: people like ajax are more than welcome of qualified comaintainers 10:15
poelcat f13: so we're coming up against a shortage of package maintainers? 10:16
jds2001 yeah that "qualified" part is the problem. He maintains what I think is probably the most difficult codebase there is to get someone new into. 10:16
f13 poelcat: to some extent yes. 10:16
-!- fraggle_laptop [n=fraggle@bea13-2-82-239-143-199.fbx.proxad.net] has joined #fedora-meeting 10:16
mcepl poelcat: you have to specifically contact maintainers and ask them if they want to get these emails. We are back at the square one -- we are here to help developers as they want to be served. Otherwise it doesn't work. 10:16
f13 poelcat: there is an abundance of people who can massage specs 10:16
poelcat i think this came up too when we started the moving a lot of bugs around in Feb and some raised the issue of bugs that had been open for > 1 year never looked at and then closed 10:17
poelcat but I suppose that is getting off topic for our meeting 10:17
poelcat however... kind of touches on what comphappy is working on 10:17
poelcat comphappy: could we help with any stats in this area? 10:17
comphappy Well I think we need to generate some kind of "wish list" for it 10:18
-!- fraggle_laptop [n=fraggle@bea13-2-82-239-143-199.fbx.proxad.net] has quit [Client Quit] 10:18
-!- fraggle_laptop [n=fraggle@bea13-2-82-239-143-199.fbx.proxad.net] has joined #fedora-meeting 10:18
comphappy I have been sifting through the Bugzilla API the last few days trying to understand the best way to approach the history portion 10:18
poelcat stickster f13 is there any data we can help provide that would help increase participation? 10:19
-!- techbugs [n=siddhart@60.254.79.60] has quit ["Leaving."] 10:19
poelcat kind of a reach, but i thought worth at least asking 10:20
comphappy Are people wanting stats that show the distribution of the age of the bugs? 10:20
poelcat comphappy: that is what I'm wondering 10:20
poelcat would it help if we could show that ___ % bugs never get touched and closed EOL, etc. 10:20
-!- techbugs [n=siddhart@60.254.79.153] has joined #fedora-meeting 10:21
mcepl comphappy: actually, I don't if it is worthy for you, but I have here Python scripts which (using XML-RPC) download all bugs from a set of components to MySQL. Would you be interested? 10:21
poelcat or would that just be demoralizing data w/o really helping anyone 10:21
mcepl *I don't know 10:21
f13 increase participation in what? 10:21
comphappy bug triage 10:21
f13 oh, doing the metrics ? 10:21
poelcat f13: i was referring package maintainers 10:21
poelcat IOW while we are working on creating better metrics on reporting on how we are doing can we help other people too? 10:22
mcepl I think that what mostly increases participation if we make bug triaging more sexy and that is done from the results, so let's make some results! 10:22
comphappy mcepl: I will take it, but right now I am working off of wwoods python-bugzilla XMLRPC work 10:22
stickster poelcat: How much additional documentation has been {warranted, scoped, written} since we got a new Docs person involved here? 10:22
stickster poelcat: (This goes to your question about what we can provide to increase participation) 10:23
mcepl comphappy: of course it uses that 10:23
jds2001 stickster: not much :( 10:23
mcepl comphappy: http://mcepl.fedorapeople.org/repo/dumpBugzilla.git/ 10:23
jds2001 i think that's partly our fault too :( 10:23
wwoods comphappy: yeah, let me know if you have questions about the API 10:23
wwoods or wishlist stuff. still need to check with dkl about a couple things.. 10:24
-!- omar_ [n=mohammed@121.246.224.40] has quit ["Leaving"] 10:24
comphappy dkl? 10:24
-!- marechal123 [n=mumuki@adsl-75-61-120-114.dsl.pltn13.sbcglobal.net] has joined #fedora-meeting 10:24
jds2001 rhbz maintainer :) 10:24
jds2001 dave lawrence 10:24
wwoods comphappy: Dave Lawrence, main RHBZ maintainer (and all-around awesome guy) 10:24
stickster super-awesome 10:25
poelcat okay so what types of things should we add to our "reporting wishlist" ? 10:25
poelcat then we'll start talking about "doing stuff" mcepl :) 10:25
comphappy he is going to have to be involved to some extent if we want to drag in the user stats as they are not assessable through the API as far as i can tel 10:26
stickster To whom are we reporting these wishlist items? Triage group? 10:26
poelcat stickster: wishlist of triage activity (or lack thereof) reports 10:27
jds2001 that's another item - there is no such thing as the "triage" group 10:27
stickster So fedora-test-list 10:27
poelcat well we have http://fedoraproject.org/wiki/BugZappers/ActiveTriagers 10:27
jds2001 there's f-t-l, but the scope of that is much wider 10:27
poelcat jds2001: ever time i bring up a separate mailing list most people disagree :) 10:28
jds2001 yeah i know :) 10:28
stickster OK, I don't want to bring that up either right now :-) 10:28
poelcat comphappy: so for starters... i think we need the "triager stats" 10:29
mcepl poelcat: I don't 10:29
mcepl (sorry, I don't disagree) 10:29
jds2001 mcepl: why not? 10:29
comphappy oh an that is going to require some access the the raw BZ database or a dump of the data 10:29
jds2001 oh, OK :) 10:29
mcepl I am not sure, whether it should be a place of all discussion, but it would be nice to have one email address as a contact point for all things bug-triage 10:30
poelcat comphappy: do you think that python script i sent is workable or something better? 10:30
-!- kms__ [n=kms@mailgate.passback.co.uk] has quit [Read error: 104 (Connection reset by peer)] 10:30
-!- kms__ [n=kms@mailgate.passback.co.uk] has joined #fedora-meeting 10:31
comphappy i looked at it a while ago, what I am doing goes off the xmlrpc so it is live data 10:31
poelcat comphappy: i'll take an action to talk to the RHT bz guys about best way we can proceed 10:31
jds2001 comphappy: i can setup a bz 2.18 instance for ya if you want 10:31
jds2001 it would be upstream 2.18, which is what rhbz is based on (severely hacked up though) 10:32
comphappy we are shifting to 3 in a week or so no? 10:32
stickster Let's get back to the topic of metrics. What are we pulling and providing right now? And where is that list provided? 10:32
mcepl comphappy: and actually, my experience with dkl is that if you need something (and you can explain why you need it), you can have some modification of bugzilla done (probably it helps to have @redhat.com in email address, but I can help) 10:32
poelcat comphappy: i'm glad to help with that part 10:32
jds2001 stickster: i have some queries, ran em once (was intending to every month) sent em to some folks and heard silence 10:33
comphappy we still dont have a list of what we want this thing to do... 10:33
mcepl stickster: I am afraid nowhere, and everybody is reinventing the whell all over again :-( 10:33
jds2001 but i dunno that those reports are worthwhile 10:33
stickster jds2001: Where could people find the queries you ran? (Not results, but the queries) 10:33
jds2001 on my workstation :) 10:34
stickster Here's what I'm getting at --let's get those somewhere public 10:34
stickster That's step 1 10:34
poelcat what were people's thoughts on this https://www.redhat.com/archives/fedora-test-list/2008-June/msg00033.html 10:34
poelcat besides the fact that "metrics are dangerous" ;-) 10:34
poelcat that was what I was putting forth 10:34
poelcat show activity by person 10:34
stickster Then 2. see about automating it 10:34
-!- knurd is now known as knurd_afk 10:35
-!- sdziallas [n=sebastia@p57A2C821.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 10:35
-!- mdomsch_ [n=Matt_Dom@cpe-70-124-62-55.austin.res.rr.com] has quit ["Leaving"] 10:35
mcepl jds2001: are you aware of the existence of the list of all your saved queries (https://bugzilla.redhat.com/userprefs.cgi?tab=saved-searches)? 10:36
comphappy part of my idea of running it as a tg app is that it would be live and we would all have one condensed set of tools 10:36
* stickster sure jds2001 is, even now, pulling those queries up and posting them somewhere :-) 10:36
-!- mdomsch [n=mdomsch@cpe-70-124-62-55.austin.res.rr.com] has quit ["Leaving"] 10:37
* poelcat has another meeting at the top of th hour... should we come back to this next week? 10:37
comphappy I will open up a wiki page for this? 10:37
jds2001 darn i forgot - the report URL's have my bugzilla username and password in em since that was the only way to make em work 10:38
jds2001 i'll take that out and make em public (they'll work in a browser, just not the scripted way I was using em) 10:38
-!- marechal123 [n=mumuki@adsl-75-61-120-114.dsl.pltn13.sbcglobal.net] has quit [] 10:38
stickster jds2001: understood 10:38
poelcat jds2001: we could set those up w/ a special ID that is RO 10:38
poelcat to bugzilla 10:39
comphappy on a side note, those greasemonkey tools are not on the tools list 10:39
stickster poelcat: +1 on making the next action "post this stuff to wiki" and moving on 10:39
poelcat okay mcepl has raise the important point a few times about "what are our goals"? 10:40
poelcat let's talk about that before we run out of time 10:40
poelcat what do we all want to be focusing on in terms of actual bug triage? 10:41
* poelcat proposes all rawhide bugs which are in NEW 10:41
poelcat keeping in mind this is our current focus culminating with GA of F10 10:41
stickster 0. Make it easier for developers to concentrate on current and compelling bugs in current releases of Fedora. 1. Bring entries in Bugzilla in line with reality of bugs in Fedora. 3. Produce helpful information about bug trends. 10:43
-!- bpepple [n=bpepple|@adsl-75-60-192-246.dsl.wotnoh.sbcglobal.net] has quit ["Ex-Chat"] 10:43
stickster That was more of a suggestion for goals, less about focus. 10:43
* stickster feels half-cocked here. 10:44
mcepl stickster: sorry, to oppose you, but that's exactly an example, what I don't think should be our goals. It should be something clear, and measurable. So, poelcat's all Rawhide NEW bugs are much close to what I think it the right answer. 10:45
jds2001 mcepl: that's a focus, not goals :) 10:45
mcepl your set is so general, that we cannot every see any tangible results from tyhat 10:45
jds2001 mcepl: goals, but their nature, aren't really measurable 10:45
comphappy fyi https://fedoraproject.org/wiki/BugZappers/Metrics 10:46
mcepl OK, I stand corrected. 10:46
jds2001 s/but/by 10:46
jds2001 a focus is very measurable :) 10:46
stickster mcepl: I see your point, and I agree that my suggestion wasn't good for measurement. And I don't want to reinvent a wheel here. 10:46
stickster For the more tangible objectives... 10:46
comphappy perhaps a # goal for the amount of open bugs at shift from rawhide to F10 10:47
stickster What about something like "Move X% of all NEW Rawhide bugs to a more accurate status"? 10:47
poelcat i propose all NEW rawhide bugs so we are in better shape when a lot of them become F10 and to be able to clearly see new rawhide bugs coming once alpha, beta, pr start 10:48
stickster 100% is a clear objective :-) 10:48
stickster poelcat: Do you feel it's also worthwhile to kick the tires of all NEEDINFO bugs at least once as well? 10:49
jds2001 NEEDINFO older than 30 days, yeah 10:49
jds2001 we actually have a policy that those can be closed - but I exercise discretion in that area. 10:49
mcepl stickster: actually, this is my question -- I tend to struggle between two rules -- either "don't leave bug until it is in the state you want it to be" and "just go through simple bugs first, to have time for the more complicated ones" (http://www.troubleshooters.com/tpromag/200008/200008.htm) 10:50
stickster Well, if there is a clear policy, you don't want to spin wheels revisiting a lot of bugs 10:50
-!- bpepple [n=bpepple|@adsl-75-60-192-246.dsl.wotnoh.sbcglobal.net] has joined #fedora-meeting 10:50
poelcat jds2001: i say leave NEEDINFO for now and just focus on NEW 10:51
comphappy this all goes back to what Metrics will show us 10:51
poelcat really knock those down... right now we're sitting fairly consistently at 1400 10:51
poelcat if 14 people triaged 100 bugs ... all gone 10:51
stickster And I think there should be an additional objective, regarding better startup documentation. 10:51
poelcat well sort of :) 10:51
-!- rvokal [n=radek@nat/redhat/x-6b95de316e4266af] has quit [Remote closed the connection] 10:52
comphappy yes 10:52
comphappy stickster: I was planing on trying to clean some of the wiki up for that 10:52
poelcat i've found that a lot of them are pretty easy even w/o knowing the component (which is most of the time for me :) 10:52
stickster comphappy: What do you need to get started? 10:52
comphappy the issue is some of our policies are not even documented there, you have to go to fedora-qa and ask and wait 10:53
-!- jhutar [n=jhutar@nat/redhat/x-9a33b7033b874905] has left #fedora-meeting [] 10:53
poelcat comphappy: that is bad 10:53
stickster comphappy: Because I would call this objective more important than a TG app rightnow 10:53
stickster And any time those answers are found, they ought to be posted in a clear policy area. 10:53
comphappy Well i will make it a goal that by next week to have some of it sorted out 10:54
comphappy then feedback? 10:54
stickster comphappy: That is AWESOME. 10:54
comphappy looks like we have about 5min left any trailing ends? 10:55
* stickster shuts the heck up. 10:55
stickster OK, I lied. 10:55
comphappy a log of this will be posted correct? 10:55
jds2001 yep 10:56
jds2001 you want me to take care of that poelcat? 10:56
poelcat mcepl: are you cool with "the goal"? 10:56
-!- fab [n=bellet@monkey.creatis.insa-lyon.fr] has quit ["Leaving"] 10:56
poelcat jds2001: please 10:56
mcepl yes 10:56
comphappy any other goals to have done by next week 10:56
poelcat note that irc2html.py does mediawiki now 10:56
-!- J5 [n=quintice@nat/redhat-us/x-695a0faa3ef8a9f4] has joined #fedora-meeting 10:56
stickster I think that as we go forward, we have to be constantly thinking, "How do I make it as easy as possible for Joe Poweruser to understand (1) where we need help, (2) how to help, and (3) what results or information he needs to pass on for others to do the same. 10:57
poelcat i say we shoot for rawhide bugs: NEW == 1200 next week 10:57
jds2001 oh, I was just gonna post the raw text :) irclog2html would be better 10:57
stickster Awesome 10:57
comphappy I will create a Goal page as well in my cleanup 10:57
* jds2001 gets busy :) 10:57
* poelcat commits to sending meeting agenda for next week's meeting by Friday 10:57
comphappy I may not be able to make it next week, but I will pass on any q and work i have done to poelcat 10:58
* stickster will commit to helping several hours this week with some triage, after work hours 10:58
poelcat comphappy: excellent! 10:58
poelcat stickster: if there is a special time you want to do it 10:59
jds2001 stickster: excellent, I'm mostly around after work hours if you need help :) 10:59
poelcat we could all try to gather at a certain time/day 10:59
* poelcat has to run 10:59
poelcat thanks everyone! 10:59
stickster Thanks poelcat 10:59
stickster Thanks everybody 10:59
* stickster hopes he didn't cause meeting suck. 10:59
comphappy this was more productive then the meeting i am off to is going to be 10:59
jds2001 nope, you caused meeting productive :) 10:59

Generated by irclog2html.py 2.6 by Marius Gedminas - find it at mg.pov.lt! [jstanley@bigbrother ~]$ cat triage-0715.log.html

poelcat who is in the house? 10:00
* stickster here to audit 10:00
* DemonJester 10:00
jds2001 f13: I've been reading that thread with great interest, but your idea got lost i guess :) 10:01
jds2001 i'll look again i guess :) 10:01
poelcat stickster: i thought you weren't auditing any more :) 10:01
stickster Well, I'm half-auditing :-D 10:01
jds2001 stickster is a full-on participant now :) 10:01
stickster Yeah, you caught me 10:01
poelcat hello DemonJester jds2001 10:01
* stickster does a crappy job of being a ninja 10:01
-!- giarc [i=hidden-u@gnat.asiscan.com] has joined #fedora-meeting 10:01
stickster "The Ninja Who Wore White" 10:01
* SMParrish Here 10:02
f13 jds2001: essentially triage folks who go to SIG meetings and spend a small time running through any 'easyfix' bugs that are within the SIG domain, letting the sig discuss/decide, and then doing whatever to the bug is necessary 10:02
poelcat mcepl: around? 10:02
poelcat f13: there is also the corresponding Keyword "patch" i've been putting on bugs 10:03
f13 nod 10:03
mcepl poelcat: yes 10:03
-!- fcrippa [n=fcrippa@83.225.231.234] has quit ["Leaving"] 10:03
poelcat people really do include patches 10:03
poelcat  :) 10:03
-!- comphappy [n=comphapp@c-71-227-212-6.hsd1.wa.comcast.net] has joined #fedora-meeting 10:04
-!- zydoon [n=zydoon@41.225.153.246] has joined #fedora-meeting 10:04
jds2001 and it's a real shame if they include patches and they dont at least get an ack/nack from the maintainer :/ 10:04
-!- zydoon [n=zydoon@41.225.153.246] has left #fedora-meeting [] 10:04
poelcat Should we wait for anyone else? 10:05
comphappy woke up 40 sec ago heh 10:05
jds2001 btw, ajax mentioned (and rightly so) that if he read all of his bugmail, he'd get nothing else accomplished. How to handle special cases like that? mcepl probably has the best ideas here :) 10:05
poelcat comphappy: :) 10:05
poelcat thanks for rising early 10:06
jds2001 that's right, it's 7AM for you left-coasters :) 10:06
poelcat did anyone have any additional agenda topics? 10:06
* jds2001 thought that you were taking care of the agenda :) 10:06
poelcat i did :) 10:06
comphappy I would like some input on what people want in the metrics stuff I am writing 10:07
f13 jds2001: that was my triage people to sig meeting thought to solve the 'I can't read bugmail' problem. 10:07
poelcat let's start w/ f13's topic 10:07
poelcat my only thought so far was sending people URL queries of: "help us with these bugs" 10:08
* poelcat is seeing the same problem w/ bugs set to NEEDINFO for maintainers 10:08
-!- gvarisco [n=gvarisco@66.187.233.202] has left #fedora-meeting [] 10:08
mcepl jds2001: there is a query https://bugzilla.redhat.com/buglist.cgi?cmdtype=runnamed&namedcmd=XGL%20having%20patch&namedowner=mcepl%40redhat.com giving all Xorg bugs with EasyFix or Patch keywords and then I set up whinning to make a weekly run of this query and send email which si then redirected to ajax. Except ajax discards all bugmail from bugzilla, including this one, so I had to remind him about the existence of this weekly mail ;-) 10:08
poelcat comphappy: okay we'll cover that too 10:08
comphappy so are we are taking about the mail generated by BZ? 10:09
-!- jhutar [n=jhutar@nat/redhat/x-9a33b7033b874905] has joined #fedora-meeting 10:09
poelcat comphappy: correct... there was a recent thread on f-d-l about maintainers drowning in bz emails 10:09
mcepl jds2001: not sure, how to expand it to the bugzilla-wide level (I had to split my original one query into two -- one Xorg and other Gecko -- because people complained about getting a list of bugs which doesn't concern them) 10:09
mcepl poelcat: my experience is that NEEDINFO set to maintainer is one of the best ways how to get bug lost. 10:10
poelcat oh no :( 10:10
mcepl with some people (ehm, ajax) is works like 100% 10:11
* poelcat has been going through the oldest NEW rawhide bugs and finding a lot of "do these still apply any more" bugs 10:11
mcepl (note that he is very good in going through bugs which are ASSIGNED to him) 10:11
poelcat could someone research this topic a little more, maybe post to the list with ideas and we can revisit next week? 10:12
poelcat f13: did you have any ideas on how to fix the situation? 10:13
poelcat mcepl: does the nag mail have any special mail headers? 10:13
poelcat mcepl: or wonder if we could have some added? 10:13
-!- covex [n=covex@sandbox.cz] has joined #fedora-meeting 10:13
mcepl I think it has special Subject 10:13
mcepl (you can set the Subject for your whines) 10:14
poelcat it is kind of a hard problem to solve people really don't want to receive the mail 10:14
poelcat would RSS help at all? 10:14
f13 poelcat: I think the only real solution is to A) get more qualified people interested in the software, and B) in person discussion. Email just gets lost 10:15
f13 poelcat: people like ajax are more than welcome of qualified comaintainers 10:15
poelcat f13: so we're coming up against a shortage of package maintainers? 10:16
jds2001 yeah that "qualified" part is the problem. He maintains what I think is probably the most difficult codebase there is to get someone new into. 10:16
f13 poelcat: to some extent yes. 10:16
-!- fraggle_laptop [n=fraggle@bea13-2-82-239-143-199.fbx.proxad.net] has joined #fedora-meeting 10:16
mcepl poelcat: you have to specifically contact maintainers and ask them if they want to get these emails. We are back at the square one -- we are here to help developers as they want to be served. Otherwise it doesn't work. 10:16
f13 poelcat: there is an abundance of people who can massage specs 10:16
poelcat i think this came up too when we started the moving a lot of bugs around in Feb and some raised the issue of bugs that had been open for > 1 year never looked at and then closed 10:17
poelcat but I suppose that is getting off topic for our meeting 10:17
poelcat however... kind of touches on what comphappy is working on 10:17
poelcat comphappy: could we help with any stats in this area? 10:17
comphappy Well I think we need to generate some kind of "wish list" for it 10:18
-!- fraggle_laptop [n=fraggle@bea13-2-82-239-143-199.fbx.proxad.net] has quit [Client Quit] 10:18
-!- fraggle_laptop [n=fraggle@bea13-2-82-239-143-199.fbx.proxad.net] has joined #fedora-meeting 10:18
comphappy I have been sifting through the Bugzilla API the last few days trying to understand the best way to approach the history portion 10:18
poelcat stickster f13 is there any data we can help provide that would help increase participation? 10:19
-!- techbugs [n=siddhart@60.254.79.60] has quit ["Leaving."] 10:19
poelcat kind of a reach, but i thought worth at least asking 10:20
comphappy Are people wanting stats that show the distribution of the age of the bugs? 10:20
poelcat comphappy: that is what I'm wondering 10:20
poelcat would it help if we could show that ___ % bugs never get touched and closed EOL, etc. 10:20
-!- techbugs [n=siddhart@60.254.79.153] has joined #fedora-meeting 10:21
mcepl comphappy: actually, I don't if it is worthy for you, but I have here Python scripts which (using XML-RPC) download all bugs from a set of components to MySQL. Would you be interested? 10:21
poelcat or would that just be demoralizing data w/o really helping anyone 10:21
mcepl *I don't know 10:21
f13 increase participation in what? 10:21
comphappy bug triage 10:21
f13 oh, doing the metrics ? 10:21
poelcat f13: i was referring package maintainers 10:21
poelcat IOW while we are working on creating better metrics on reporting on how we are doing can we help other people too? 10:22
mcepl I think that what mostly increases participation if we make bug triaging more sexy and that is done from the results, so let's make some results! 10:22
comphappy mcepl: I will take it, but right now I am working off of wwoods python-bugzilla XMLRPC work 10:22
stickster poelcat: How much additional documentation has been {warranted, scoped, written} since we got a new Docs person involved here? 10:22
stickster poelcat: (This goes to your question about what we can provide to increase participation) 10:23
mcepl comphappy: of course it uses that 10:23
jds2001 stickster: not much :( 10:23
mcepl comphappy: http://mcepl.fedorapeople.org/repo/dumpBugzilla.git/ 10:23
jds2001 i think that's partly our fault too :( 10:23
wwoods comphappy: yeah, let me know if you have questions about the API 10:23
wwoods or wishlist stuff. still need to check with dkl about a couple things.. 10:24
-!- omar_ [n=mohammed@121.246.224.40] has quit ["Leaving"] 10:24
comphappy dkl? 10:24
-!- marechal123 [n=mumuki@adsl-75-61-120-114.dsl.pltn13.sbcglobal.net] has joined #fedora-meeting 10:24
jds2001 rhbz maintainer :) 10:24
jds2001 dave lawrence 10:24
wwoods comphappy: Dave Lawrence, main RHBZ maintainer (and all-around awesome guy) 10:24
stickster super-awesome 10:25
poelcat okay so what types of things should we add to our "reporting wishlist" ? 10:25
poelcat then we'll start talking about "doing stuff" mcepl :) 10:25
comphappy he is going to have to be involved to some extent if we want to drag in the user stats as they are not assessable through the API as far as i can tel 10:26
stickster To whom are we reporting these wishlist items? Triage group? 10:26
poelcat stickster: wishlist of triage activity (or lack thereof) reports 10:27
jds2001 that's another item - there is no such thing as the "triage" group 10:27
stickster So fedora-test-list 10:27
poelcat well we have http://fedoraproject.org/wiki/BugZappers/ActiveTriagers 10:27
jds2001 there's f-t-l, but the scope of that is much wider 10:27
poelcat jds2001: ever time i bring up a separate mailing list most people disagree :) 10:28
jds2001 yeah i know :) 10:28
stickster OK, I don't want to bring that up either right now :-) 10:28
poelcat comphappy: so for starters... i think we need the "triager stats" 10:29
mcepl poelcat: I don't 10:29
mcepl (sorry, I don't disagree) 10:29
jds2001 mcepl: why not? 10:29
comphappy oh an that is going to require some access the the raw BZ database or a dump of the data 10:29
jds2001 oh, OK :) 10:29
mcepl I am not sure, whether it should be a place of all discussion, but it would be nice to have one email address as a contact point for all things bug-triage 10:30
poelcat comphappy: do you think that python script i sent is workable or something better? 10:30
-!- kms__ [n=kms@mailgate.passback.co.uk] has quit [Read error: 104 (Connection reset by peer)] 10:30
-!- kms__ [n=kms@mailgate.passback.co.uk] has joined #fedora-meeting 10:31
comphappy i looked at it a while ago, what I am doing goes off the xmlrpc so it is live data 10:31
poelcat comphappy: i'll take an action to talk to the RHT bz guys about best way we can proceed 10:31
jds2001 comphappy: i can setup a bz 2.18 instance for ya if you want 10:31
jds2001 it would be upstream 2.18, which is what rhbz is based on (severely hacked up though) 10:32
comphappy we are shifting to 3 in a week or so no? 10:32
stickster Let's get back to the topic of metrics. What are we pulling and providing right now? And where is that list provided? 10:32
mcepl comphappy: and actually, my experience with dkl is that if you need something (and you can explain why you need it), you can have some modification of bugzilla done (probably it helps to have @redhat.com in email address, but I can help) 10:32
poelcat comphappy: i'm glad to help with that part 10:32
jds2001 stickster: i have some queries, ran em once (was intending to every month) sent em to some folks and heard silence 10:33
comphappy we still dont have a list of what we want this thing to do... 10:33
mcepl stickster: I am afraid nowhere, and everybody is reinventing the whell all over again :-( 10:33
jds2001 but i dunno that those reports are worthwhile 10:33
stickster jds2001: Where could people find the queries you ran? (Not results, but the queries) 10:33
jds2001 on my workstation :) 10:34
stickster Here's what I'm getting at --let's get those somewhere public 10:34
stickster That's step 1 10:34
poelcat what were people's thoughts on this https://www.redhat.com/archives/fedora-test-list/2008-June/msg00033.html 10:34
poelcat besides the fact that "metrics are dangerous" ;-) 10:34
poelcat that was what I was putting forth 10:34
poelcat show activity by person 10:34
stickster Then 2. see about automating it 10:34
-!- knurd is now known as knurd_afk 10:35
-!- sdziallas [n=sebastia@p57A2C821.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 10:35
-!- mdomsch_ [n=Matt_Dom@cpe-70-124-62-55.austin.res.rr.com] has quit ["Leaving"] 10:35
mcepl jds2001: are you aware of the existence of the list of all your saved queries (https://bugzilla.redhat.com/userprefs.cgi?tab=saved-searches)? 10:36
comphappy part of my idea of running it as a tg app is that it would be live and we would all have one condensed set of tools 10:36
* stickster sure jds2001 is, even now, pulling those queries up and posting them somewhere :-) 10:36
-!- mdomsch [n=mdomsch@cpe-70-124-62-55.austin.res.rr.com] has quit ["Leaving"] 10:37
* poelcat has another meeting at the top of th hour... should we come back to this next week? 10:37
comphappy I will open up a wiki page for this? 10:37
jds2001 darn i forgot - the report URL's have my bugzilla username and password in em since that was the only way to make em work 10:38
jds2001 i'll take that out and make em public (they'll work in a browser, just not the scripted way I was using em) 10:38
-!- marechal123 [n=mumuki@adsl-75-61-120-114.dsl.pltn13.sbcglobal.net] has quit [] 10:38
stickster jds2001: understood 10:38
poelcat jds2001: we could set those up w/ a special ID that is RO 10:38
poelcat to bugzilla 10:39
comphappy on a side note, those greasemonkey tools are not on the tools list 10:39
stickster poelcat: +1 on making the next action "post this stuff to wiki" and moving on 10:39
poelcat okay mcepl has raise the important point a few times about "what are our goals"? 10:40
poelcat let's talk about that before we run out of time 10:40
poelcat what do we all want to be focusing on in terms of actual bug triage? 10:41
* poelcat proposes all rawhide bugs which are in NEW 10:41
poelcat keeping in mind this is our current focus culminating with GA of F10 10:41
stickster 0. Make it easier for developers to concentrate on current and compelling bugs in current releases of Fedora. 1. Bring entries in Bugzilla in line with reality of bugs in Fedora. 3. Produce helpful information about bug trends. 10:43
-!- bpepple [n=bpepple|@adsl-75-60-192-246.dsl.wotnoh.sbcglobal.net] has quit ["Ex-Chat"] 10:43
stickster That was more of a suggestion for goals, less about focus. 10:43
* stickster feels half-cocked here. 10:44
mcepl stickster: sorry, to oppose you, but that's exactly an example, what I don't think should be our goals. It should be something clear, and measurable. So, poelcat's all Rawhide NEW bugs are much close to what I think it the right answer. 10:45
jds2001 mcepl: that's a focus, not goals :) 10:45
mcepl your set is so general, that we cannot every see any tangible results from tyhat 10:45
jds2001 mcepl: goals, but their nature, aren't really measurable 10:45
comphappy fyi https://fedoraproject.org/wiki/BugZappers/Metrics 10:46
mcepl OK, I stand corrected. 10:46
jds2001 s/but/by 10:46
jds2001 a focus is very measurable :) 10:46
stickster mcepl: I see your point, and I agree that my suggestion wasn't good for measurement. And I don't want to reinvent a wheel here. 10:46
stickster For the more tangible objectives... 10:46
comphappy perhaps a # goal for the amount of open bugs at shift from rawhide to F10 10:47
stickster What about something like "Move X% of all NEW Rawhide bugs to a more accurate status"? 10:47
poelcat i propose all NEW rawhide bugs so we are in better shape when a lot of them become F10 and to be able to clearly see new rawhide bugs coming once alpha, beta, pr start 10:48
stickster 100% is a clear objective :-) 10:48
stickster poelcat: Do you feel it's also worthwhile to kick the tires of all NEEDINFO bugs at least once as well? 10:49
jds2001 NEEDINFO older than 30 days, yeah 10:49
jds2001 we actually have a policy that those can be closed - but I exercise discretion in that area. 10:49
mcepl stickster: actually, this is my question -- I tend to struggle between two rules -- either "don't leave bug until it is in the state you want it to be" and "just go through simple bugs first, to have time for the more complicated ones" (http://www.troubleshooters.com/tpromag/200008/200008.htm) 10:50
stickster Well, if there is a clear policy, you don't want to spin wheels revisiting a lot of bugs 10:50
-!- bpepple [n=bpepple|@adsl-75-60-192-246.dsl.wotnoh.sbcglobal.net] has joined #fedora-meeting 10:50
poelcat jds2001: i say leave NEEDINFO for now and just focus on NEW 10:51
comphappy this all goes back to what Metrics will show us 10:51
poelcat really knock those down... right now we're sitting fairly consistently at 1400 10:51
poelcat if 14 people triaged 100 bugs ... all gone 10:51
stickster And I think there should be an additional objective, regarding better startup documentation. 10:51
poelcat well sort of :) 10:51
-!- rvokal [n=radek@nat/redhat/x-6b95de316e4266af] has quit [Remote closed the connection] 10:52
comphappy yes 10:52
comphappy stickster: I was planing on trying to clean some of the wiki up for that 10:52
poelcat i've found that a lot of them are pretty easy even w/o knowing the component (which is most of the time for me :) 10:52
stickster comphappy: What do you need to get started? 10:52
comphappy the issue is some of our policies are not even documented there, you have to go to fedora-qa and ask and wait 10:53
-!- jhutar [n=jhutar@nat/redhat/x-9a33b7033b874905] has left #fedora-meeting [] 10:53
poelcat comphappy: that is bad 10:53
stickster comphappy: Because I would call this objective more important than a TG app rightnow 10:53
stickster And any time those answers are found, they ought to be posted in a clear policy area. 10:53
comphappy Well i will make it a goal that by next week to have some of it sorted out 10:54
comphappy then feedback? 10:54
stickster comphappy: That is AWESOME. 10:54
comphappy looks like we have about 5min left any trailing ends? 10:55
* stickster shuts the heck up. 10:55
stickster OK, I lied. 10:55
comphappy a log of this will be posted correct? 10:55
jds2001 yep 10:56
jds2001 you want me to take care of that poelcat? 10:56
poelcat mcepl: are you cool with "the goal"? 10:56
-!- fab [n=bellet@monkey.creatis.insa-lyon.fr] has quit ["Leaving"] 10:56
poelcat jds2001: please 10:56
mcepl yes 10:56
comphappy any other goals to have done by next week 10:56
poelcat note that irc2html.py does mediawiki now 10:56
-!- J5 [n=quintice@nat/redhat-us/x-695a0faa3ef8a9f4] has joined #fedora-meeting 10:56
stickster I think that as we go forward, we have to be constantly thinking, "How do I make it as easy as possible for Joe Poweruser to understand (1) where we need help, (2) how to help, and (3) what results or information he needs to pass on for others to do the same. 10:57
poelcat i say we shoot for rawhide bugs: NEW == 1200 next week 10:57
jds2001 oh, I was just gonna post the raw text :) irclog2html would be better 10:57
stickster Awesome 10:57
comphappy I will create a Goal page as well in my cleanup 10:57
* jds2001 gets busy :) 10:57
* poelcat commits to sending meeting agenda for next week's meeting by Friday 10:57
comphappy I may not be able to make it next week, but I will pass on any q and work i have done to poelcat 10:58
* stickster will commit to helping several hours this week with some triage, after work hours 10:58
poelcat comphappy: excellent! 10:58
poelcat stickster: if there is a special time you want to do it 10:59
jds2001 stickster: excellent, I'm mostly around after work hours if you need help :) 10:59
poelcat we could all try to gather at a certain time/day 10:59
* poelcat has to run 10:59
poelcat thanks everyone! 10:59
stickster Thanks poelcat 10:59
stickster Thanks everybody 10:59
* stickster hopes he didn't cause meeting suck. 10:59
comphappy this was more productive then the meeting i am off to is going to be 10:59
jds2001 nope, you caused meeting productive :) 10:59

Generated by irclog2html.py 2.6 by Marius Gedminas - find it at mg.pov.lt!