[19:15] <Tm_T> AlanBell: choo choo says night train
[19:21] <AlanBell> :)
[19:24] <AlanBell> so Tm_T shall we go through the agenda?
[19:24] <IdleOne> o/
[19:25] <IdleOne> let's get to it, we are burning daylight.
[19:25] <AlanBell> #startmeeting IRC team
[19:25] <meetingology> Meeting started Sun May 26 19:25:57 2013 UTC.  The chair is AlanBell. Information about MeetBot at http://wiki.ubuntu.com/meetingology.
[19:25] <meetingology> 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
[19:26] <AlanBell> https://wiki.ubuntu.com/IRC/IrcCouncil/MeetingAgenda
[19:26] <AlanBell> we have missed a couple of meetings, and this one isn't really getting off to a great start :(
[19:27] <Tm_T> hmmm
[19:27]  * knome blows to a vuvuzela to lift up the mood
[19:27] <Tm_T> if there's something that require broader participation from IRCC I suppose we can continue via mail with those subject
[19:28] <IdleOne> +1
[19:28] <IdleOne> m4v still hasn't been made a member :/
[19:29] <AlanBell> yes, we can and m4v membership application is something I would really like to do by mail, that has been hanging for far too long
[19:29] <Tm_T> agreed
[19:29] <AlanBell> #topic Review last meetings action items
[19:30]  * AlanBell hunts back for the last meeting that wasn't just postponing stuff to the next meeting
[19:33] <AlanBell> gosh, it really has been some time
[19:34] <AlanBell> I will have a further hunt later for undone action items
[19:35] <AlanBell> #topic Open items in the IRCC tracker
[19:35] <AlanBell> there has been no activity in the tracker
[19:35] <AlanBell> #topic Review Bugs related to the Ubuntu IRC Council
[19:35] <AlanBell> we have no bugs
[19:35] <AlanBell> #topic Membership applications
[19:36] <AlanBell> we do have a membership application from m4v
[19:36] <AlanBell> https://wiki.ubuntu.com/m4v
[19:36] <AlanBell> https://launchpad.net/~m4v
[19:37] <AlanBell> m4v has been a consistent long term contributor to Ubuntu in particular in the IRC area, and especially on the bot infrastructure
[19:37] <bazhang> that should be an insta-pass
[19:37] <Unit193> Is he here?
[19:38] <AlanBell> I am entirely comfortable that the IRCC is an appropriate membership body for this application
[19:38] <bazhang> m4v, not in this channel
[19:38] <AlanBell> the wiki page looks good and there are strong testimonials
[19:39] <IdleOne> m4v gets my +1
[19:39] <IdleOne> please to pass that along on the mailing list :)
[19:39] <AlanBell> we will vote on it by email, anyone else got any thoughts to add?
[19:40] <Tm_T> +1 for mailinglist
[19:40] <lderan> Hello
[19:40] <knome> lderan, o/
[19:42] <Tm_T> let's move to next topic
[19:43] <AlanBell> yup
[19:43] <AlanBell> #topic Replace floodbot kick+ban behaviour for webchat users who flood - tsimpson
[19:44] <AlanBell> https://wiki.ubuntu.com/IRC/IrcCouncil/IRCteamproposal has further details
[19:44] <AlanBell> Currently when a webchat user floods the bots will set a ban and kick the user, then forget about them. I think this behaviour is overly defensive and discriminates against webchat users unfairly. It would make more sense if the bots simply set -e for that user (as to apply the existing quiet on webchat users) instead of the +q they normally set.
[19:44] <AlanBell> This would require some modification to the floodbot script but, as already treats webchat users who flood differently, it shouldn't be too major.
[19:44] <Tm_T> +1 from mew
[19:44] <knome> AlanBell, what's -e ?
[19:44] <AlanBell> removing the exempt
[19:45] <AlanBell> so as I understand it, webchat is +q by default
[19:45] <Tm_T> knome: webchat is banned entirely, individuals have +e exempt that allows them to talk at all
[19:45] <knome> okay
[19:45] <Tm_T> banned, quieted, same effect in this case
[19:45] <IdleOne> knome: currently we +q all webchat, then floodbots set +e so they can speak in the channel.
[19:45] <knome> sounds fair
[19:46] <IdleOne> setting -e would allow them to remain in the channel, but the floodbots would need to let the user know somehow how to get the +e reset.
[19:47] <IdleOne> which would mean that when there are no ops around, the ops channel could end up with users waiting a long time and getting frustrated
[19:47] <knome> wouldn't the webchat users get +e if they just rejoined?
[19:47] <AlanBell> but at the moment they just get banned altogether
[19:47] <knome> or would that have to be done manually
[19:48] <IdleOne> knome: floodbots track which users have been banned so we would need some sort of mechanism to let floobots know that $user is ok to let back in the channel and be able to speak
[19:48] <knome> mhm
[19:50] <IdleOne> I don't think it will be a major issue. +1 to setting -e instead of kick/banning
[19:50] <AlanBell> yeah, I think the policy suggestion is fine, it just needs someone to tangle with the floodbot code and implement it
[19:51] <AlanBell> LJL knows that code, dunno if tsimpson was proposing to do it or not
[19:51] <IdleOne> How is the open sourcing of that code coming, if at all?
[19:51] <AlanBell> it would be good to get other people familiar with the floodbot code
[19:52] <Unit193> Could be just like a quiet, -e and if it continues +b kick, and if not, +e.
[19:52] <AlanBell> the code is on launchpad, not much progress since then, LjL wasn't around for some time and it became less of a priority once it was on launchpad (albeit in a private project)
[19:54] <AlanBell> https://launchpad.net/floodbot is the project
[19:54] <IdleOne> Would it be easier to use supybot with the functionality of the floodbots added?
[19:55] <AlanBell> floodbot is written in PHP rather than python, it is 1807 lines of code
[19:56] <AlanBell> and does some funky stuff with multiple instances of the bot talking to each other through control channels and sharing updates via a pastebin and other surprising and interesting features
[19:57] <IdleOne> short answer: no.
[19:57] <LjL> if the floodbots aren't going to set +b on kicked webchat users any longer, are we confident the ops will know how to actually ban them?
[19:59] <IdleOne> most of us use scripts to do that
[20:01] <LjL> IdleOne: which are webchat-aware?
[20:01] <IdleOne> chanserv.py currently defaults to *591c6114@* for webchat
[20:02] <LjL> uh?
[20:02] <LjL> you mean it bans the ident?
[20:02] <IdleOne> correct
[20:02] <LjL> well that won't let the floodbots know the user is banned
[20:03] <IdleOne> I know
[20:03] <IdleOne> which is why those of us who use chanserv.py try not to ban webchat and instead we remove/kick and let floodbot set the ban
[20:03] <LjL> then again these days the IP is in the webchat mask, so it's less bad than it used to be. but banning the ident or other funny things one might do don't play nice with the system
[20:04] <bazhang> yep
[20:04] <bazhang> just a remove
[20:04] <LjL> IdleOne: and what do the others do, just ban *!*@gateway/whatever?
[20:04] <IdleOne> not sure what auto_bleh does
[20:04] <IdleOne> I haven't used it in a long time
[20:07] <LjL> anyway to answer your question IdleOne, i don't think anyone has worked on the floodbots to bring them to what could be a releasable state. i know i haven't, and at this point, i really don't think i'm ever going to
[20:07] <Tm_T> I have to go, but I support any changes toward equal treatment
[20:07] <AlanBell> ok Tm_T
[20:07] <IdleOne> LjL: understandable.
[20:10] <AlanBell> so, do we think this is still a good idea, a bad idea, or an idea that requires further thought, and a plan for someone to implement it?
[20:10] <IdleOne> currently the only one who could implement it is LjL right?
[20:12] <IdleOne> I think maybe a better idea would be leave things the way they are but have floodbot send a notice to the channel asking for a live op to set a comment/duration on the ban
[20:12] <Unit193> (Autobleh doesn't ban nick or ident, just host.  At least, that's how I set it up.)
[20:12] <IdleOne> s/to the channel/to the control channel/
[20:12] <LjL> i can comment out the line the says "ban it", but not sure whether it'll have other side effects i'm currently not thinking about. i guess i can just do it and we'll see. i'm not really in the mood to look at it in detail
[20:13] <LjL> IdleOne: well it already sends a "-WARNING" that people are supposed to highlight on
[20:13] <IdleOne> I would prefer if a live op decides whether to remove the ban or set a time limit on it.
[20:13] <IdleOne> LjL: it doesn't tell us the banID though
[20:14] <LjL> it doesn't have a clue about the "banID"
[20:14] <IdleOne> doesn't ubottu ask floodbot to comment on the ban?
[20:14] <LjL> i don't think so
[20:14] <IdleOne> ubottu: sens me a PM every time I set a ban
[20:14] <IdleOne> sends*
[20:15] <LjL> IdleOne, pretty sure the floodbots are exempted from them
[20:15] <LjL> that*
[20:16] <IdleOne> Well, fair treatment is all well and good. IF it means possibly breaking the floodbots other behaviors we might want to hold off.
[20:21] <AlanBell> yeah, I am not sure about this
[20:21] <AlanBell> maybe it could be a banforward rather than a ban
[20:21] <AlanBell> but I am not sure how much of a problem this actually is
[20:21] <LjL> honestly IdleOne, if you want to talk about this, you might as well talk about removing the webchat +b entirely and just handling webchat users manually. the floodbots introduced automatic handling for this because it was hard to match people's real hostnames with the webchat hostname they obtained
[20:21] <LjL> but these days, the hostname for webchat users contains their real IP
[20:21] <LjL> so you can just go ban that as needed
[20:23] <AlanBell> ok, well it was a proposal from tsimpson, who was around earlier
[20:23] <IdleOne> LjL: I didn't see the +b as an issue honestly. I would like it if the floodbots could ask us to set comments on the banID (not a big thing if they don't)
[20:23] <IdleOne> just thought it would be a nice feature
[20:24] <AlanBell> I think we need to discuss further, perhaps on the mailing list, I am not sure that there are any changes that really need to be made, and if there are, it might not be the one proposed
[20:24] <IdleOne> IMO leave things as is. I think we are searching for a tech solution where one isn't needed.
[20:25] <LjL> IdleOne: maybe ubottu can do that, the floodbots sure can't as they don't know about banIDs (and if they somehow got ubottu to tell you, it would be annoyingly complicated to deal with it, and there's no reason why ubottu couldn't do it in the first place)
[20:25] <AlanBell> OK, I will send a mail to the list later with minutes of this meeting and highlight the discussion around this topic
[20:25] <IdleOne> sounds good.
[20:26] <LjL> AlanBell: maybe the only problem that should be a real problem that i see is what happens when people got automatically muted for flooding. if the involved person is a webchat user, then they get an automatic ban, and if there's no one around to remove that, they stay banned.
[20:26] <AlanBell> yeah, I just don't know if that actually happens much to real people who are not intentionally flooding
[20:26] <LjL> AlanBell: as to when ops actually kick someone manually, though... i think they should have a look at the modes being set and the floodbot "-WARNING" and realize
[20:28] <LjL> anyway tsimpson may also be able to implement whatever is needed, i'm sure he's among the people most familiar with both the floodbots and ubottu. get his opinion, or just let him work out what seems best
[20:28] <AlanBell> yeah :)
[20:28] <AlanBell> #topic Any Other Business
[20:28] <tsimpson> ubottu does
[20:28] <tsimpson> "[ubottu] Please somebody comment on the ban of *!*@static-mum-59.181.90.147.mtnl.net.in in #ubuntu done by FloodBot3, use: @comment 55195 <comment>" eg
[20:28] <AlanBell> oh, hi tsimpson
[20:28] <tsimpson> \o
[20:28] <LjL> tsimpson: ah that's nice, IdleOne's concern is settled then i think
[20:29] <IdleOne> so the real problem is humans not listening to the bot when it asks for something
[20:29] <LjL> yeah :)
[20:29] <AlanBell> yeah, I think maybe a refresher on what messages the bots do that humans should respond to might be good
[20:29] <Unit193> AlanBell: Review of the ubuntu-bots bugs?  I'd think [Eir] ones could be closed?
[20:31] <AlanBell> #action AlanBell to compile page on bot messages for humans
[20:31] <meetingology> ACTION: AlanBell to compile page on bot messages for humans
[20:31] <tsimpson> there is only bug #899630 left
[20:31] <tsimpson> that should probably be closed
[20:31] <AlanBell> interesting point Unit193
[20:31] <AlanBell> https://bugs.launchpad.net/ubuntu-bots
[20:32] <AlanBell> bug #899630 is now closed
[20:33] <knome> meetingology is tweaked by lderan! \o/
[20:34] <meetingology> knome: Error: "is" is not a valid command.
[20:34] <knome> boo.
[20:34] <lderan> :P
[20:34] <AlanBell> yes, lderan has done some great stuff :) private votes are on the way \o/
[20:34] <knome> and the moin output is so much better, it's pretty much ready-to-copy-paste now
[20:35] <AlanBell> yup
[20:35] <AlanBell> ok, any other stuff before I close this meeting?
[20:35] <lderan> :D
[20:35] <AlanBell> #endmeeting
[20:35] <meetingology> Meeting ended Sun May 26 20:35:51 2013 UTC.
[20:35] <meetingology> Minutes (wiki):        http://ubottu.com/meetingology/logs/ubuntu-meeting/2013/ubuntu-meeting.2013-05-26-19.25.moin.txt
[20:35] <meetingology> Minutes (html):        http://ubottu.com/meetingology/logs/ubuntu-meeting/2013/ubuntu-meeting.2013-05-26-19.25.html
[20:35] <AlanBell> thanks everyone
[20:36] <IdleOne> thank you
[20:36] <knome> you should all look in the wonderful wiki output
[20:38] <lderan> yes do :P
[20:42] <IdleOne> looks beautiful
[20:52] <AlanBell> https://wiki.ubuntu.com/MeetingLogs/IRCC/20130526