| *** dangtrinhnt_ has quit IRC | 00:04 | |
| *** jamesmcarthur has quit IRC | 00:08 | |
| *** dangtrinhnt has joined #openstack-tc | 00:10 | |
| *** jamesmcarthur has joined #openstack-tc | 01:14 | |
| *** jamesmcarthur has quit IRC | 01:24 | |
| *** jamesmcarthur has joined #openstack-tc | 04:24 | |
| *** jamesmcarthur has quit IRC | 04:29 | |
| *** jaosorior has joined #openstack-tc | 05:37 | |
| *** tosky__ has joined #openstack-tc | 06:54 | |
| *** ricolin has joined #openstack-tc | 07:13 | |
| *** jpich has joined #openstack-tc | 07:53 | |
| *** tosky has joined #openstack-tc | 08:43 | |
| *** tosky__ has quit IRC | 08:44 | |
| *** e0ne has joined #openstack-tc | 08:59 | |
| *** dtantsur|afk is now known as dtantsur | 09:48 | |
| *** jamesmcarthur has joined #openstack-tc | 10:25 | |
| *** jamesmcarthur has quit IRC | 10:29 | |
| *** jaosorior has quit IRC | 11:27 | |
| *** e0ne has quit IRC | 12:01 | |
| *** e0ne has joined #openstack-tc | 12:08 | |
| *** jaosorior has joined #openstack-tc | 12:10 | |
| *** jamesmcarthur has joined #openstack-tc | 12:27 | |
| *** jamesmcarthur has quit IRC | 12:35 | |
| *** lbragstad has joined #openstack-tc | 12:39 | |
| *** lbragstad has quit IRC | 12:40 | |
| dhellmann | those of you interested in our communication tools might find the thread about python-dev moving to discourse an interesting read, both in how the decision is/was being made and in the results: https://mail.python.org/pipermail/python-committers/2018-September/006187.html | 12:54 |
|---|---|---|
| *** jroll has quit IRC | 12:54 | |
| *** jroll has joined #openstack-tc | 12:55 | |
| jroll | interesting | 12:58 |
| *** jamesmcarthur has joined #openstack-tc | 12:59 | |
| TheJulia | hmmm | 13:02 |
| cmurphy | it seems like one of the key motivators is burnout from policing the mailing lists for CoC violations, I don't really feel like we have that problem | 13:08 |
| *** jamesmcarthur has quit IRC | 13:15 | |
| smcginnis | I haven't used Discource, but at first look, I kind of like that it shows view counts and (surprisingly to me) it has likes. | 13:18 |
| smcginnis | One thing with the ML is there are times when you post something and you get little response, and you wonder if no one actually read it, or they don't care. | 13:19 |
| fungi | i'm not really a fan of the like/dislike mechanism. it's one of the things i have noted to look into disabling in hyperkitty for mailman3 | 13:27 |
| dhellmann | cmurphy : yes, we've been lucky in that regard | 13:27 |
| *** cdent has joined #openstack-tc | 13:28 | |
| cdent | o/ | 13:28 |
| *** annabelleB has joined #openstack-tc | 13:34 | |
| *** jamesmcarthur has joined #openstack-tc | 13:36 | |
| *** jamesmcarthur has quit IRC | 13:37 | |
| *** jamesmcarthur has joined #openstack-tc | 13:38 | |
| *** dklyle has joined #openstack-tc | 13:47 | |
| *** jaosorior has quit IRC | 13:48 | |
| mnaser | i kinda like the like/dislike because its kindof a soft +1 feedback | 13:56 |
| mnaser | sometimes i see something and i just want to say "i like this idea" | 13:56 |
| mnaser | but not necessarily write out a whole email | 13:56 |
| persia | It is subject to entirely different gaming mechanism. While convenient, it ends up having weak semantic value: some folk consider it "I like this idea", others "I read this", and worse, some "this is important (although I disagree)". | 13:59 |
| mnaser | i do think that the forums approach is so much easier | 14:04 |
| mnaser | its more approachable to users too | 14:05 |
| mnaser | and maybe it can get rid of the whole ask.o.o thing | 14:05 |
| smcginnis | fungi: I normally do not like the like/dislike, but in this case I do not dislike the like/dislike because I see some benefits. | 14:07 |
| mnaser | would this be something we'd propose inside infra-specs? | 14:08 |
| cmurphy | corvus made a good point at one of the infra ptg sessions and one of the emails in that thread mentioned it as well, which is that while $platform may be more approachable/easier to newcomers, those of us who have spent years honing our workflows with certain tools have achieved remarkable efficiency with them, so switching to a new tool is not just about getting old stubborn people to adjust to the new | 14:09 |
| cmurphy | thing but also throwing out years worth of workflow refinement | 14:09 |
| smcginnis | cmurphy: That's a very good thing to keep in mind. | 14:10 |
| mnaser | cmurphy: yeah, that's solid, but i feel like we gotta compromise a little bit. we would probably be irked about it for a week and probably (maybe) like it after | 14:11 |
| mnaser | a new user will just say: mailing list? that's too complicated | 14:11 |
| cmurphy | i think i would be irked for longer than a week | 14:12 |
| mnaser | i think a lot of these software have 'email' integration | 14:12 |
| mnaser | so you can still use them as a mailing list | 14:12 |
| mnaser | but i worry if our behaviour of not wanting to change might be slowly hurting us and isolating us | 14:13 |
| *** leakypipes is now known as jaypipes | 14:15 | |
| cdent | zaneb: don't make me bring infinite resource classes back into the discussion... | 14:20 |
| zaneb | cdent: I'm guessing I should consider myself fortunate that I have no idea what that means ;) | 14:23 |
| cdent | it's a 2016 model for traits that I had, using resource classes: https://anticdent.org/simple-resource-provision.html | 14:24 |
| cdent | the points in the very beginning of that continue to inform my position on most things, but the actual discussion of using infinity is more noodly | 14:26 |
| dhellmann | I think the placement discussion should probably move over to the nova channel | 14:27 |
| dhellmann | it's not going to do any good to fragment that conversation | 14:27 |
| * zaneb joins #openstack-placement | 14:28 | |
| cdent | dhellmann: this is mostly idle social chatter in reference to being somewhat chided by zaneb in his response | 14:28 |
| dhellmann | still, talking about it in this specific channel seems like a bad idea | 14:29 |
| * cdent sighs and flips a table | 14:30 | |
| * cdent joins openstack-social | 14:30 | |
| zaneb | cdent: not chiding, merely pointing out which areas could still benefit from more elucidation :) | 14:30 |
| clarkb | one thing I see a lot is the assumption these new/different tools are easier. Is there any data for that? | 14:34 |
| clarkb | email is the one tool I feel most computer users are actually familiar with and able to use effectively | 14:36 |
| persia | clarkb: Yes. Far too much. Most with specific userbases, mostly *much* smaller than OpenStack. Sadly, I don't know of any large/diverse userbase studies. | 14:36 |
| cmurphy | i'd also be interested in the data showing that the current tools are a problem | 14:36 |
| clarkb | (I dont have data for that either) | 14:36 |
| * persia very much dislikes email, to the point of using an MUA often less than once a week, but still believes forums to be less useful for most sorts of communications (due to lack of data analysis tools, mostly) | 14:37 | |
| zaneb | certainly signing up for the firehose in your inbox is a big commitment in a way that browsing a forum isn't | 14:37 |
| dhellmann | yeah, I think the usability characteristics for the tools depend on your level of engagement | 14:37 |
| * smcginnis fondly remembers the nightmare of first signing up for openstack-dev | 14:37 | |
| zaneb | OTOH having to create a login for a forum is also an obstacle | 14:37 |
| persia | zaneb: Use a different inbox, or just check lists.openstack.org on a regular basis. | 14:38 |
| zaneb | projects that just use GitHub issues for everything can piggy-back on the fact that ~everyone already has a GitHub login | 14:38 |
| zaneb | persia: creating a new email address is also a big commitment for casual users. browsing lists.o.o is not a great interface, although I gather it will be better with mailman3 | 14:41 |
| clarkb | it adda the forum like experience too | 14:43 |
| fungi | there is a https://lists-dev.openstack.org/mailman3/ up since a couple months ago with a couple of test mailing lists you're free to test out, though i'll be tearing it down and rebuilding it soon | 14:44 |
| fungi | the lists on there were mostly just so i could confirm mm3 handles multi-site listserv functionality out of the box withuot the ugly hacks we did to make that possible in mm2 | 14:45 |
| fungi | though https://mail.python.org/mm3/mailman3/lists/ is probably more interesting if you want to check out archives on a relatively active deployment | 14:47 |
| fungi | for example, https://mail.python.org/mm3/archives/list/distutils-sig@python.org/ | 14:47 |
| clarkb | fungu that us the one they are deleting? | 14:47 |
| fungi | i only heard they were moving python-dev | 14:48 |
| clarkb | oh just the one list got it | 14:48 |
| persia | zaneb: I'd argue that creating a new address is a smaller commitment than sending openstack-dev@ to an email address that doesn't have tooling prepared to handle it :) | 14:48 |
| fungi | the python-committers list, apparently | 14:48 |
| clarkb | that seems worse than moving everyrhing imo | 14:48 |
| jroll | sounds like they're trying to move everything by EOY | 14:49 |
| jroll | > As long as we don't identify any deal breakers in October, I will send an e-mail like this to python-dev on November 1st, and to python-list and python-ideas on December 1st | 14:49 |
| clarkb | also discourse is not foss? | 14:50 |
| clarkb | oh it us | 14:50 |
| clarkb | they also have hosted version got it | 14:50 |
| clarkb | it renders on mobile such that $100/month is way bigger and above the "we are open source" | 14:51 |
| fungi | "everything" as in the 210 mm2 lists and 45 mm3 lists they're running at mail.python.org? | 14:51 |
| zaneb | persia: I'd argue that it's academic because they're both too big for someone with a casual interest :) | 14:52 |
| persia | Fair :) | 14:52 |
| fungi | "Other mailing lists are welcome to port over to Discourse too. | 14:54 |
| fungi | " | 14:54 |
| fungi | makes it sound like they're not going to just forklift >250 other lists en masse | 14:54 |
| jroll | fungi: sorry, I meant "all of the python lists" when I said everything. didn't realize how much they hosted :) | 14:54 |
| fungi | well, those 255 lists on mail.python.org could be said to all be "python lists" (otherwise why would they be there instead of somewhere else?) | 14:55 |
| jroll | "all of the lists that I believed to be the core python mailing lists, before looking at mail.python.org" | 14:56 |
| jroll | you are correct | 14:56 |
| *** ricolin has quit IRC | 14:58 | |
| fungi | https://mail.python.org/pipermail/python-committers/2018-September/006209.html echoes corvus's comments from the ptg cmurphy cited | 15:01 |
| fungi | personally, i'd rather we moved our lists to nntp-based newsgroups | 15:01 |
| fungi | mail-to-news gateways are still very much a possibility for people who cling to inefficient e-mail based group discussion | 15:02 |
| fungi | for the rest of us, there's a long-established solution to that problem people tried to (poorly) reimplement over smtp | 15:02 |
| fungi | so if we're talking about replacing current discussion platforms, i'd like us to very much consider usenet as the far superior alternative | 15:03 |
| *** mriedem has joined #openstack-tc | 15:07 | |
| persia | Does NNTP still require all hosts to be actively responsive, or are intermittent relays still permitted? The only reason I like email is that I have access to text archives for post-processing (which I didn't have with NNTP without jumping through all sorts of hoops, but I haven't used NNTP at all in the 21st century) | 15:08 |
| zaneb | fungi: I think I'd like to live in a world where that was an option (honestly too young to have tried it properly :P), but in reality I suspect that Google Groups killed NNTP client development dead some time before it effectively finished off Usenet | 15:10 |
| fungi | in truth, spam is really what (mostly) killed unmoderated usenet, much as it has crippled if not nearly killed e-mail years later | 15:11 |
| persia | zaneb: While you are mostly true, lots of folk use Thunderbird, which can do NNTP with an interface very similar to how it does SMTP. | 15:11 |
| zaneb | persia: huh. I use Thunderbird and I did not know that. thanks | 15:12 |
| fungi | lots of muas are also newsreaders | 15:12 |
| fungi | persia: usenet supported nntp over uucp at the dawn of the internet, so i'm not sure what suggests a need for actively responsive server peers | 15:12 |
| fungi | i remember being on a newsserver which got batch feed updates each time its uucp cron kicked off and it dialled up its peers | 15:13 |
| persia | fungi: At some point around when I started moving away from NNTP, there were rumours of some groups not being cross-posted to "inactive" hosts. I do remember being on a newsserver that didn't have active TCP relay to the "Internet", so I know it used to be possible. | 15:14 |
| * cdent misses running innd | 15:14 | |
| persia | cdent: I don't believe you. I believe you miss having a running innd that was useful. | 15:14 |
| cdent | persia: try again. I liked it it so much I've got code in it. | 15:14 |
| persia | Wow! | 15:14 |
| cdent | I didn't use it much. It's wrong to get high on one's own supply | 15:15 |
| mnaser | fwiw: i like vanilla, they have a hosted platform (on openstack none the less!) but also an oss version | 15:15 |
| mnaser | https://github.com/vanilla/vanilla | 15:15 |
| cdent | <- porn king of indiana 1997 | 15:15 |
| fungi | following the python-committers thread about the supposed move to discourse, guido's followup at https://mail.python.org/pipermail/python-committers/2018-September/006233.html suggests the original message was overzealous and it's only an experiment, not a done deal | 15:30 |
| *** annabelleB has quit IRC | 15:31 | |
| *** annabelleB has joined #openstack-tc | 15:47 | |
| *** jamesmcarthur has quit IRC | 15:49 | |
| *** jamesmcarthur has joined #openstack-tc | 15:51 | |
| *** e0ne has quit IRC | 15:58 | |
| TheJulia | cdent: that is quite a title... | 16:00 |
| *** jpich has quit IRC | 16:29 | |
| *** jamesmcarthur has quit IRC | 16:34 | |
| *** annabelleB has quit IRC | 16:45 | |
| *** annabelleB has joined #openstack-tc | 16:59 | |
| *** dtantsur is now known as dtantsur|afk | 17:03 | |
| *** jamesmcarthur has joined #openstack-tc | 17:04 | |
| *** mriedem has quit IRC | 17:04 | |
| zaneb | I suppose somebody has to get the 4pm Thursday speaking slots at the summit, but I was really hoping it would be somebody else | 17:08 |
| cmurphy | better than the 5pm Thursday slot | 17:11 |
| *** e0ne has joined #openstack-tc | 17:24 | |
| *** tosky has quit IRC | 17:33 | |
| *** e0ne has quit IRC | 17:37 | |
| *** e0ne has joined #openstack-tc | 17:40 | |
| *** e0ne has quit IRC | 17:41 | |
| notmyname | is the published summit schedule final? eg https://www.openstack.org/summit/berlin-2018/summit-schedule/global-search?t=project+update is interesting (or rather, what's *not* there is interesting) | 18:01 |
| *** e0ne has joined #openstack-tc | 18:14 | |
| *** e0ne has quit IRC | 18:14 | |
| *** mriedem has joined #openstack-tc | 18:20 | |
| *** annabelleB has quit IRC | 18:23 | |
| dhellmann | notmyname : I don't think the deadline for asking for those has passed, has it? | 18:25 |
| dhellmann | we just talked about asking for one in the oslo meeting this morning | 18:26 |
| *** annabelleB has joined #openstack-tc | 18:26 | |
| dhellmann | tc-members: I'm not likely to be able to participate in the next board meeting on 25 Oct because it falls during my vacation plans. Are any of you planning to attend and willing to take notes for the rest of us? | 18:31 |
| *** diablo_rojo has joined #openstack-tc | 18:34 | |
| *** dangtrinhnt has quit IRC | 18:35 | |
| *** e0ne has joined #openstack-tc | 18:37 | |
| TheJulia | Attend as in call-in? | 18:44 |
| TheJulia | If so, I can | 18:44 |
| dhellmann | yeah, it's a phone meeting | 18:46 |
| dhellmann | thanks, TheJulia | 18:46 |
| *** jamesmcarthur has quit IRC | 18:58 | |
| *** diablo_rojo has quit IRC | 19:08 | |
| *** tosky has joined #openstack-tc | 19:09 | |
| *** diablo_rojo has joined #openstack-tc | 19:14 | |
| *** dangtrinhnt has joined #openstack-tc | 19:36 | |
| *** jamesmcarthur has joined #openstack-tc | 19:38 | |
| openstackgerrit | Merged openstack/governance master: Retire fuxi projects https://review.openstack.org/604527 | 19:41 |
| fungi | confirmed, i plan to dial in though the second half does overlap with our thursday office hour and the security sig weekly meeting | 19:42 |
| fungi | happy to also take notes | 19:42 |
| openstackgerrit | Merged openstack/governance master: Retire charm-ceph https://review.openstack.org/604530 | 19:45 |
| *** annabelleB has quit IRC | 19:51 | |
| *** diablo_rojo has quit IRC | 20:24 | |
| *** mriedem has quit IRC | 20:42 | |
| *** e0ne_ has joined #openstack-tc | 20:54 | |
| *** e0ne has quit IRC | 20:54 | |
| *** diablo_rojo has joined #openstack-tc | 21:03 | |
| *** annabelleB has joined #openstack-tc | 21:06 | |
| *** jamesmcarthur has quit IRC | 21:30 | |
| *** cdent has quit IRC | 21:51 | |
| *** jamesmcarthur has joined #openstack-tc | 21:55 | |
| *** jamesmcarthur has quit IRC | 21:59 | |
| *** jamesmcarthur has joined #openstack-tc | 22:03 | |
| *** e0ne_ has quit IRC | 22:04 | |
| *** e0ne has joined #openstack-tc | 22:05 | |
| *** e0ne has quit IRC | 22:06 | |
| *** ianw is now known as ianw_pto | 22:17 | |
| *** mriedem has joined #openstack-tc | 22:36 | |
| *** jamesmcarthur has quit IRC | 22:42 | |
| *** tosky has quit IRC | 22:43 | |
| *** jamesmcarthur has joined #openstack-tc | 22:44 | |
| *** jamesmcarthur has quit IRC | 22:46 | |
| *** jamesmcarthur has joined #openstack-tc | 22:48 | |
| fungi | wow, following up on the rest of the python-committers thread about discourse and then checking out the corresponding forum topic on discuss.python.org, it's... painful https://discuss.python.org/t/is-this-really-better-than-a-mailing-list/57/18 | 22:50 |
| fungi | i can't imagine trying to interact that way | 22:51 |
| fungi | there's no thread tree, everything is serialized | 22:51 |
| mtreinish | fungi: wow, yeah that doesn't look like a good way to do things | 22:53 |
| fungi | not to mention the little icons for people and the lack of display names in quoted context is very distracting and hard to follow | 22:53 |
| *** jamesmcarthur has quit IRC | 22:55 | |
| fungi | apparently it has a "mailing list mode" you can turn on if you only want to interact via e-mail, but no idea if in includes sufficient context to work well (and i would probably have to do that as i really can't abide by the web-browser-as-the-default-client disease infecting so much of the internet these days) | 22:57 |
| *** annabelleB has quit IRC | 23:01 | |
| fungi | anyway, curious to see what the python-committers community chooses to do at the end of this 3-month experiment | 23:06 |
| dhellmann | indeed | 23:06 |
| fungi | anyway, in much the same way that mailing list usage discourages involvement from contributors who find e-mail painful, i expect web forum usage will discourage involvement from contributors who find web forums painful | 23:22 |
| persia | As someone who finds *both* painful, I think email wins by the narrow margin of supporting a wider range of interaction mechanisms. Being able to embed an entire email (or thread) as a MIME attachment allows a subscriber to simultaneously inform a non-subscriber and allow them to subscribe and then participate as if they had always been subscribed. Being able to post-process individual messages for statistics, analytics, or just searches via a | 23:30 |
| persia | fairly well-defined API provides a lot of flexibility beyond what is provided by most forum enginers. Being able to "reply" with a hand-modified distribution list lets one easily take conversations out of the wider community, and bring them back once they are suitable (e.g. someone wants to reply to a position, but needs to get confirmation from their boss first: this is tricky in a forum without publishing all details of all participating | 23:30 |
| persia | organisations to properly configure ACLs). | 23:30 |
| persia | (oh, I also like to be able to download complete mbox archives for prior posts, supported by many email archive solutions, but I suppose that's about the same as being able to replicate a forum content DB, so not a clear advantage) | 23:31 |
| dtroyer | I'd argue that the schema for mbox is pretty well-defined and understood by many existing tools, forum X DB schema probably doesn't match forum Y schema very well… | 23:40 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!