IRC meeting decision about "discussion"
Hi all, (Koha general mailing list reader: this mail is also for you, please read it's an opportunity to participate to Koha, even if you don't have technical skills !) Tomas was faster than me to start a discussion, I planned to send a mail about the results of our discussion during the last IRC session. The question was : "how to deal with patches that are 'in discussion'". I made a proposal, that has been a little bit amended during the discussion, here is the final result: * when a patch is rejected or is conflicting with another one (not a technical conflict, but a "strategic" conflict), the bug status can be set to "In Discussion" * Someone (I volunteered to do it, but feel free to do it yourself if you're concerned= - start a wiki page with describing the problem - send a mail to koha-devel and koha pointing to the wiki page, and calling for comments. * the delay for discussion is "until next IRC meeting, not less than 1 week" * at the next IRC meeting, - if the discussion (on the wiki or koha-devel) result in a general agreement, nothing specific need to be made (no vote needed) - if the discussion result in a balanced situation, organize a vote. About the vote (this has not been discussed on last IRC meeting, thinking of it and proposing it now): as some of us are always sleeping during IRC meetings, it's unfair to vote only on the IRC channel during the meeting. I propose that ppl could also express their preference/vote on the wiki itself (and have a specific section for that on the page). Those votes would be added to the IRC vote. I've started the general discussion wiki page: http://wiki.koha-community.org/wiki/Bug_and_Enhancement_Discussion -- Paul POULAIN http://www.biblibre.com Expert en Logiciels Libres pour l'info-doc Tel : (33) 4 91 81 35 08
Salvete!
(Koha general mailing list reader: this mail is also for you, please read it's an opportunity to participate to Koha, even if you don't have technical skills !)
+1
* the delay for discussion is "until next IRC meeting, not less than 1 week"
I think that should probably read "window" in the stead of "delay". If you have something to talk about, don't delay. :D
About the vote (this has not been discussed on last IRC meeting, thinking of it and proposing it now): as some of us are always sleeping during IRC meetings, it's unfair to vote only on the IRC channel during the meeting. I propose that ppl could also express their preference/vote on the wiki itself (and have a specific section for that on the page). Those votes would be added to the IRC vote.
This is very yucky on its face. For a long time now, the bar of involvement has been if you really are that keen to change things, show up at the General Meeting. This used to suck more before we rotated meeting times and folks could be screwed out of a meeting for literally months at a time. While it's nice to have spam protection on the wiki, there are roadblocks to signing up. Do you really want a maths test when you're in a hurry to vote on something before work? How about a long wait for authorisation that might cost you the ability to vote on something you care about? I think this erodes a lot of the impetus for using IRC. I've noticed that there are less folks on there in general, though. Separate problem, but how do we get to know one another now and make new folks feel welcomed, preferably in a realish time way. Identica mebbe? Even if everyone wanted to do it this way, and I turn out to be very much in the minority, how are you going to accurately tabulate this many proxy votes? Deduplicating this as chair gets to be a bit of a bear, but it would be manageable at one issue per meeting. I certainly don't mind it as much for stuff like Conference. However this just doesn't seem right for deeper issues, that there are probably many of since this whole thing started to address a bug backlog. I'm willing to give over on this if we can get 50 replies to this message that say "I wanna do it Paul's way." AND you have a bot or summat that will crawl the wiki for a given issue, tabulate proxy votes and then check against active IRC meeting participants. (But that seems like programming time that could be better spent on fixing barcodes or performance or whatnot.) Cheers, Brooke
For a long time now, the bar of involvement has been if you really are that keen to change things, show up at the General Meeting.
I think this is a valid point. The meeting agendas are posted well in advance, so it should be possible to arrange to be present for a vote. We've already demonstrated that for larger issues (like the Koha non-profit question) we can use alternate methods to vote. I say we should keep the smaller stuff (especially developer-centric stuff) to the IRC meetings. -- Owen -- Web Developer Athens County Public Libraries http://www.myacpl.org
Le 10/02/2012 14:24, Owen Leonard a écrit : >> For a long time now, the bar of involvement has been if you really are that keen to change things, >> show up at the General Meeting. > > I think this is a valid point. The meeting agendas are posted well in > advance, so it should be possible to arrange to be present for a vote. Well, with time shifting, don't expect ppl from the "it's 2AM" timezone to be here, so I strongly think we should adress this issue. > We've already demonstrated that for larger issues (like the Koha > non-profit question) we can use alternate methods to vote. I say we > should keep the smaller stuff (especially developer-centric stuff) to > the IRC meetings. So maybe we should say: * discussion votes are made on the mailing list / on the wiki, and not on IRC if you think we should not have 2 places to vote ? Note that i've nothing against more than one method to vote. In France, if you're not present the day of the vote, you can do a "vote par procuration" (proxy vote says gg translate). It would be the same kind of voting. To answer Brooke concern: I also think the "you must register on the wiki" is not a big deal. If you want to be involved in Koha, registering on the wiki & bugzilla will be hard to avoid... -- Paul POULAIN http://www.biblibre.com Expert en Logiciels Libres pour l'info-doc Tel : (33) 4 91 81 35 08
Salvete! >> I think this is a valid point. The meeting agendas are posted well in >> advance, so it should be possible to arrange to be present for a vote. > Well, with time shifting, don't expect ppl from the "it's 2AM" > timezone > to be here, so I strongly think we should adress this issue. > How? I don't think anyone expects people from the crummy timezone to be present. The best we can do is slate a time that works for a good chunk of the geographic world, as we do now. It's always going to be a reality of life for real time meetings when we have a global project that part of the project is disenfranchised part of the time. If we come together as we have and say that "Okay, this is a large issue, let's invoke a special procedure" that's fine. It's basically a motion to suspend the rules as far as I'm concerned. Those are in order some of the time in special cases. Making a special case a day to day reality is why I'm starting to dig in. I don't want to overpromise myself here. >> We've already demonstrated that for larger issues (like the Koha >> non-profit question) we can use alternate methods to vote. I say we >> should keep the smaller stuff (especially developer-centric stuff) to >> the IRC meetings. > So maybe we should say: > * discussion votes are made on the mailing list / on the wiki, and not > on IRC if you think we should not have 2 places to vote ? Note that i've > nothing against more than one method to vote. In France, if you're not > present the day of the vote, you can do a "vote par procuration" > (proxy > vote says gg translate). It would be the same kind of voting. > All of this is a matter of frequency. Voting in France is not monthly on multiple small issues. Furthermore, there's infrastructure at work that I don't have at my disposal. I am not even an arrondisement, Paul. If we really want this, then we'd need a committee of volunteers that would show to count proxies _every_ month. That's a lot of man hours that I'd frankly prefer to dedicate elsewhere. If we decided on this for roles *maybe*. I'd be inclined to say certainly if we went to a 9 or 12 month release cycle. Conference to me works well: it's a once a year occurrence. We're lucky to have Nicole helping there and tabulating things. Hashing things out on the mailing list first certainly works. I still like formalising it at the meeting so that there's a logical end to discussion. I don't think anyone wants to miss a discussion, which is why the one week window before the meeting for things labeled discussion works. I think it's going to be tricky to ensure that we don't fall back to things discussed over the list and MUNG in IRC. We'll see though. > To answer Brooke concern: > I also think the "you must register on the wiki" is not a big deal. If > you want to be involved in Koha, registering on the wiki & bugzilla will > be hard to avoid... > It's not a big deal with plenty of lead time. If the wiki is down, then it becomes a big deal. Bugzilla is not particularly hard to avoid for non devs. We need a nice participation flow from listserv, IRC, wiki to harder things like sandboxing, bugzilla, and proper developing. Cheers, Brooke
On 11/02/12 00:24, Owen Leonard wrote:
For a long time now, the bar of involvement has been if you really are that keen to change things, show up at the General Meeting. I think this is a valid point. The meeting agendas are posted well in advance, so it should be possible to arrange to be present for a vote. We've already demonstrated that for larger issues (like the Koha non-profit question) we can use alternate methods to vote. I say we should keep the smaller stuff (especially developer-centric stuff) to the IRC meetings.
-- Owen
The usual method of voting where you can't turn up in person is to send a 'proxy' vote. I see no reason why (known and experienced) people shouldn't be able to send a vote to the chair prior to the meeting, or even to ask someone else to exercise their vote at the meeting. This needs trust and good sense which both exist most of the time (whilst formalism is to be avoided if possible). Just a thought. Bob
participants (4)
-
Bob Birchall -
BWS Johnson -
Owen Leonard -
Paul Poulain