****** Meeting logging now! ****** lol gmgbot` has no regard for schendje's feelings [11:02] schendje: Error: "has" is not a valid command. we'll just have to use the copy-pasta route since I don't know what I'm doing gmgbot`: help nyergler: (help [] []) -- This command gives a useful description of what does. is only necessary if the command is in more than one plugin. gmgbot`: commands awesome paroneayea: Error: "commands" is not a valid command. haha ok ok, faking it so first of all [11:03] nice, or should I say *amazing* job everyone on this release I honestly do think it was our best month yet. It was certainly the month with the most fun screenshots in the press release at least * schendje claps so everyone pat themselves on the bat and feel good about themselves :) [11:04] and then stop so we can continue with the meeting :) so! this next month I'm going to ramble for a second then open the floor [11:05] basically we intend to try and get 0.1.0 out this month, and http://mediagoblin.com as a real example site next month, so we need to wrap things up, though I feel like I've said "wrap things up" the last couple of meetings :) I wrote out a vague outline of things that I knew I wanted to discuss to the list: - Licensing support! [11:06] - Wrap up 0.0.5 stragglers - RDFa integration - Caching integration - More user experience testing and feedback integration! - A new and updated website look and feel! - Starting to work on multi-media support in the processing backend, if there's time? - What else?! - Surely there are things. - ... http://mediagoblin.org/ redesign oh documentation is surely a thing so I don't know where we want to start specifically so I say let's do the website first :) [11:07] schendje: ping! Presumably you're here http://bugs.foocorp.net/issues/543 I just wrote this up, which includes ascii art mockups of the site based on conversations in here last night so I guess the big thing is, how do we intend to execute on this :) [11:08] I'm opening the floor here btw I'm also going to be traveling a bunch for work this month, so I think I'll need a lot of help directing this one [11:09] no comments? :) are the bills paid? :p [11:10] wctype_t: the bills are paid :) for mediagoblin.org at least that's to speak nothing of mediagoblin.com ;) [11:11] okay, so no serious comments, so I guess I hope we'll iterate on mockups on the ticket and jsut start pushing it forward in general my preference is for simpler, easier to read websites that aren't fussy. I think it's probably better to start with a list of what needs to be on mg.org that isn't there... and I'll put this in the ticket. Csrf needs attention tychoish: ok great, that feedback is appreciated [11:12] gullydwarf: yes, that's true though that csrf ticket has turned from being "CSRF ticket" to "everything and the kitchen sink security" ticket without any action items Don't want to get diaspora*-ed is the .org site static HTML? it's own python app? (sorry for the n00b question) nyergler: not a n00b question it's pyblosxom :) rendering to static HTML [11:13] ok, cool for ultimate laziness I actually wrote up a wiki page on how to mess with it oh, sorry, my client doesn't ping me :( http://wiki.mediagoblin.org/Update_the_website np schendje :) (i assume we'll loop back around to CSRF, since that sounds important for an 0.1 ) nyergler: yes let's get to that in a second :) [11:14] schendje: so, transforming my ascii art shittiness into real mockups iirc school's starting for you so you're getting busy but paroneayea: I think I can handle that :) cool, I've been thinking maybe the design could actually be more like your original mediagoblin mockups still dark themed, similar to what we have, but more assertive as it's marketing materials [11:15] cool well we've certainly got a lot more freedom, so that's fun look forward to seeing those then :) yeah great will it be translated, btw? or English only? schendje: unfortunately likely not we might handle that at a future time for now static html that we barely have to maintain is best paroneayea: well for the design that's actually a lot easier ;) yeah ok :) schendje: I'm assigning the ticket to you then sounds good! [11:16] comments in the ticket from people generally are welcome yeah so I think I'll just keep updating that one great and everyone can comment on mockups when they want assigned :) sounds perfect ok! so CSRF :) gullydwarf: ^^ Present so honestly I think we should probably split out a new ticket that's actually CSRF only since the old one is confusing maybe? [11:17] regardless, someone should take the lead on actually acting on *csrf* this month. what's the existing ticket? volunteers? I'd love to see something like the lazy-csrf stuff django has found it, 361 yes good question oh it's not as bad as I remember the ticket being :) [11:18] still a lot of non-csrf-activity :) so I haven't used it since it was after I stopped really using django but I hear https://docs.djangoproject.com/en/dev/ref/contrib/csrf/ is really great and easy to use i think it'll take a bit for me to dig through this, but i'm happy to work on CSRF-ness nyergler: great! Sweet! i've done some of it for ye-olde-daye-jobe and we're dealing with both Django and non-Django there [11:19] I obviously don't want to use django's but just an idea of doing something similarly easy Even better got it great nyergler: could you register on the tracker and I'll mark you in the hacker group then I can assign it to you nyergler: thanks :) i suspect i'll have stupid questions, but i think paroneayea is probably used to answering those from me :) i just registered (nyergler, naturally) nyergler: it'll be payback for ~2 years of dumb questions from me when you were my boss [11:20] i was thinking i was the one asking dumb questions then ;) we'll make it even, we were both dumb sweet K, driving home now ok, assigned to you [11:21] everyone's dumb. the world goes on. gullydwarf: don't IRC an drive! :) gullydwarf: we can talk more later if there are things you want to follow up on okay, so other things docs! that's a big one tychoish: ^^^^ :) unfortunately willkg is not here :( I really wanted him to be part of this meeting for that, but oh well [11:22] Evening! tychoish: a pretty clear first action item for the docs is homogenizing all the "hacking howto" type pages or at leas splitting them up and making them clear tychoish: presumably I should write up a ticket on that and assign to you? hiya Elrond :) [11:23] everyone's favorite elvenlord has joined the party. those are the pages in the wiki about how to get started with the hacking? yes ok. yeah. tychoish: there's even an old .rst version of it back in the git history :) even a version still un-deleted from the site ;) paroneayea - :-))) Except only few people bow. ;o) [11:24] http://docs.mediagoblin.org/hackinghowto.html lightweight markup is fungable tychoish: great, you want to write up the ticket for that or should I? tychoish: I also wonder about other things that should be done docs-wise. Are there things you need from me to plow ahead? is the version in rst different from the ones in the wiki? tychoish: the wiki ones have been updated a bit [11:25] ok, OSX and distro instructions improvements, possibly some other stuff tangential question: if i'm getting started with dev, and find things that are out of date, should i edit the wiki? email the devel list? depends on which things, I guess. nyergler: editing is good, and pinging on IRC is great great, thanks usually we can coordinate... I always seem to not be far away from my desktop :P [11:26] if people can tell me what's up and what needs to be changed, I'll make tickets. I'm not always here, but logging happens, email's good too. my main questions are: 1) how does the docs.mediagoblin get published tychoish: cronjob :) [11:27] the poor man's daemon every half hour or so, git pull and sphinx make awesome. yup it runs whatever's in master presently 2. are we still comfortable with docs/ being in the source code repo? tychoish: I'm comfortable with it, yes [11:28] sweet. we can consider moving it, but if so let's do it at a later time when willkg is also here I think it's fine there for now though of course, I don't mind, but I don't want it to irritate you/other people so just tracking that. I'm happy with it where it is. is there a 3) ? :) yeah, just everyone should open tickets and assign them to me for documentation bugs, or things that they think needs to be documented or better documented. [11:29] Is the meeting already over or are we currently at the "docs" topic? Elrond: on "docs" tychoish: ok, I'll try to review things and send more issues your way later today [11:30] and you write up the hacking howto ticket sounds good? until I get more up to speed I kinda need to depend on folks to set my agenda :) yep sounds good. Okay, I'll wait for the next topic then. :) ok i'm done. :) next topic! :) caching well this one's on my plate Elrond says I should just merge it now without tests ;) though honestly we're not far away from that it should be done this weekend I'm not sure there's more to say about it [11:31] unless people have questions, let's move on - More user experience testing and feedback integration! paroneayea - IF it doesn't break anything. That's on the "doesn't hurt, might even help. Maybe someone else will write tests. Or someone else will start using it and find issues". Elrond: it doesn't break anything it just sits on the side [11:32] minding its own business twiddling its thumbs merge it. leave ticket open for tests. * nyergler wonders what the test coverage is like right now as a whole next topic. nyergler: good question ;) nyergler: probably not great ok, fair enough nyergler: we have tests for submission (including evil submissions) and auth stuff and most of the components config stuff, storage system, etc [11:33] most of the components have test sounds like a great start but not all plenty of views don't have tests but most of the ones that worry me about evil things happening do I think, test coverage is fine currently. it could always be better ;) [11:34] but yeah not bad for now non user facing things need tests! And the most important ones have. true okay - Licensing support! user facing stuff... is tested by users. ;o) So actually I'm prototyping a tool at work-ish about this I'd actually like to discuss it with nyergler later more I don't know if I'll have time to finish it by this release :( but I'll try this is a tool for license selection? [11:35] nyergler: yes we could do it the cheap way for now :) Does that also include the selection of harmless-ness? Elrond: whaa right -- i was going to ask if we had schema support for license (good for everyone, not good for children, bad for everyone) i suspect the answer is "yes" since we're mongo and the answer is always yes :) Elrond: that seems like a different flag nyergler: yeah we just need to add a field to the MediaEntry thing that has 'license' [11:36] and have it point at URLs. nyergler - It is. I think generally the solution mlinksva and I have been discussing is the value of the license field is actually the URL, and maybe somewhere else in the application is info about that license based on its url of course we could probably get the field into MediaEntry and add static support for the core 6 [11:37] and the tool I'm talking about is just auto-jsonifying it using rdfa profiles but yeah iterate on that for adding support for all variants, etc we don't need that tool for 0.1.0 yet we could just do it manually that makes no sense what I just said without context maybe :) is there a ticket for license support? [11:38] nyergler: I don't think so nyergler: do you want to write one up maybe? :) or I can after the meeting. and you can comment on it happy to do it great I also have as an issue for this meeting - RDFa integration and now that nyergler is here I might be not the only #mediagoblin participant who cares about that anymore ;) [11:39] I would like to add "- database schema (I have some feelings there!)" to the end of the topic list, either for public discussion or private discussion after the meeting. Elrond: is this about non-mongodb support :) paroneayea - No. Elrond: oh, what is it then? I'm curious paroneayea: i definitely care about it, not sure i know enough about GMG to comment at the moment [11:40] paroneayea - Let's postpone this until that topic is there. :-) Don't want to disturb now. okay nyergler: honestly I care about it but it's pretty low priority for 0.1.0 maybe comparatively i suggest we create a ticket for basics nyergler: yeah though honestly title, creator, pub date it's pretty easy. yup right nyergler: could you make that ticket too ;) [11:41] haha this is payback, isn't it sure ;) :) great - More user experience testing and feedback integration! this is a thing I really care about but none of the people who have been working on it are present at this meeting I think. but basically we need to turn those UX testing reports into real bugs [11:42] *** gullydwarf_ (~vdfc@50.15.244.85) has joined channel #mediagoblin heya gullydwarf_ Home :) horray paroneayea - I have some private UX testing going on omgmg.*. I will try to make bugs from this. I just brought up user experience testing sweet home Two gullydwarf? One isn't enough? ;) [11:43] people have started writing things on the wiki, we should turn them into bugs, but I'm not sure if there should be a process for that Elrond: :) Elrond: :-)) we should probably read Jan's paper ;) * paroneayea wonders if joar is lurking :) schendje: yes I think so schendje: I'll talk to him about how to go about transforming those reports into bugs he would know if anyone would. yep * paroneayea files into orgmode [11:44] ok :) will do. also, thank you thank you gullydwarf_ for starting omgmg :) paroneayea - I would at least say: If someone went over one report and turned it into bugs, that someone should put links to the bugs at the end of the report, so that one can follow, what happens, etc. paroneayea: you bet I feel bad about yesterday gullydwarf_: don't. and well compensated for my time, btw :) gullydwarf_: it was a good lesson and experience :) [11:45] it always sucks when someone has to come in and clean up your own mess :/ *** mlinksva (~mlinksva@adsl-76-200-160-216.dsl.pltn13.sbcglobal.net) has joined channel #mediagoblin gullydwarf_: don't apologize, thanks for working with me on it but I am using it as a learning exp, you're correct oh hey mlinksva ! now there are 3 people who care about the rdfa issue :) wow, i'm feeling all nostalgic :) nyergler - Why do you feel nostalgic? [11:46] mlinksva: was nyergler's boss/coworker, is mine paroneayea, mlinksva, and I used to work together at CC yeah Ahh. and nyergler used to be my boss ;) okay [11:47] - Starting to work on multi-media support in the processing backend, if there's time? I'm not sure I'll get to this [11:48] probably realistically post-0.1.0 i suggest we punt *** gullydwarf (~AndChat@173-147-118-171.pools.spcsdns.net) has quit: Quit: Bye I'd vote for post-0.1 too. but we're getting close to an infrastructure that we will support that yeah ok :) let's ignore that entirely ignoreeed! I guess all tehre is to talk about then is 0.0.5 stragglers, or anything else people intend to raise! * paroneayea links http://bugs.foocorp.net/projects/mediagoblin/roadmap [11:49] oh, I'd really like to wrap up http://bugs.foocorp.net/issues/296 because i think it's risky with this not working as-is it makes dumb assumptions for example, it would totally do the wrong thing on joar's site. most of the rest of the things on that list I'm not sure I care to comment on [11:50] except that things like tag clouds and etc are maybe not so easy to do until we figure out if we want to outsource those tasks to celery or mapreduce * paroneayea hates using mapreduce unless necessary in this case, probably necessary but I don't think it's a critical 0.0.5 feature [11:51] okay I'm done bringing things up anyone else have things? * gullydwarf_ is looking through the roadmap is willkg working on 296? *NEWS* issues: Feature #544 (New): Basic content license support nyergler: realistically I think no :\ he's swamped with migrating the bug tracker off of foocorp onto redmine and a bunch of other things we don't have a db dump so he wrote an html scraping library [11:52] foocorp isn't running redmine? (can't tell what it is) w.t.f.? paroneayea: what do you think about me claiming http://bugs.foocorp.net/issues/357 for refactoring? can not wait to hear this story :) nyergler: yes we can discuss later ;) gullydwarf_: claim it! please! [11:53] done :) ok so after I wrap up the caching thing, I wonder if I have major things to do this month besides a lot of triaging and management and site redesign and etc :) [11:54] sounds like you have major things to do this month :) - buy mediagoblin.com - Make business plan. ;o) Elrond: already bought it Elrond: for $200 :p someone had it. Oh, yuck. [11:55] Elrond: well anyway, yes, we're working on that also. - Add features to GMG for mediagoblin.com business plan. ;) oh! there is one! I can't believe I forgot captchas. - Gather and sell people's data to advertisers or similar. schendje: :) schendje: hehe [11:56] schendje - Oh, yeah. ;) proper freemium plan. (or whatever that is spelled) well captchas though for serious. I don't know what to do about this paroneayea: captchas for the signup page? yes just for registration we'll likely need them I hate using recaptcha because of the centralization aspect [11:57] but honestly all existing captcha solutions suck :\ for python recaptcha is easy, I think. It's just a very good mind mood testing thing. we could probably integrate a "shitty recaptcha hack" into mediagoblin for now and have an open ticket for non-evil-centralized captchas. and maybe make a library for it. [11:58] Hmm, yeah. right -- just make sure you can yank it out once we know what the decentralized version is :) yeah Just ask people a math question. ;o) Elrond: computers will never solve those! Like "integrate ln(x) from 1 to x" ;o) ok, i need to drop off shortly, anything else i can help with at the moment, paroneayea? +1 math, something you can plug into wolfram alpha nyergler: I think you've volunteered for plenty, and excited about you jumping on board :) [11:59] hrm, maybe not * paulproteus waves to nyergler! gullydwarf_: that would be a great solution if we wanted a bot-only site We could also do chemistry questions. ;o) keep the humans out! * gullydwarf_ bye nyergler! wow, it is old home day on #mediagoblin paroneayea: indeed tty soon later :) okay :) [12:00] looks like we've wrapped up then with some good action items for the month :) :-) paroneayea: oh [12:01] any new thoughts on federation? we seem to be about where we were last month gullydwarf_: my thoughts are "we work out a plan first thing after 0.1.0 is out the door" :) *NEWS* issues: Feature #545 (New): Add RDFa to HTML output || Feature #357: "Lost password?" functionality ("Change password" functionality possibly embedded i... and the month before ... - db schema: I feel MediaEntry gets added fields like mad. "Oh, we need x, it has something vaguely to do with some MediaEntey, add it there." I don't like that. Lately the processing errors. They're in every entry, but only needed *while* processing or for error cases. 0.1.0 is supposed to be "the last pre-federation-work" branch paroneayea: sounds good that's one reason why I said this month's should be 0.0.5 instead of 0.1.0 ok :) [12:02] I'm going to go press my french press and drink some coffee thanks all great meeting! ******* END LOGGING! ********