[00:41] godbyk: Typically a screeshot is on the top of the page and then the text follows. Is there any problem if I move some text before screenshot (actually put the screenshot after a paragraph that is slit by it) [00:45] split* [01:03] godbyk: When you find some time, send me an e-mail about what can we do about the "movement" of screenshots. Some screenshots have to be placed after some paragraphs or even on different pages... (I have sent you 1-2 mails on your gmail so I guess you have my e-mail address :D) [01:03] long day .... night all [02:54] Finally home. [02:55] * flan hax test code. [03:02] Bleh. I forgot to push my Quickshot commits from work. [04:19] Website leaf naming convention: nouns or verbs? [04:20] (index/list_projects, screenshot/upload) [04:20] Anyone with substantial development experience have a preference? [04:21] Or is it generally fine to just use whatever makes sense? [04:57] * flan went with nouns. [04:59] Next design thing: construct archives in-memory or on-disk? [05:00] Actually, I can't do it in-memory, so that's a non-issue. [05:03] Ooh. I can combine the tempfile module and the tarfile module. Yay for Python. [05:04] lol [05:05] I'm enjoying just watching you talk to yourself [05:06] I think better when I've convinced people care about my musings. [05:06] I'm* [05:35] URL mapping rules defined. Now for sleep. [05:35] (I can actually start playing with templating and data tomorrow! Yay!) [08:19] * humphreybc is super tired [08:29] humphreybc: ping [08:32] czajkowski: ello [15:02] ubuntujenkins, after a bit of reflecting, I kinda think it might make more sense to go with reStructuredText for documentation, after all. We're not building a library, so the main benefits of epytext would be lost, and Pylons, which I'm using for the server, auto-generates reST stubs everywhere, in addition to using it as the project-encompasing document template standard. [15:04] flan: that fine by me can you mail the list and update this page please [15:04] ?style [15:04] The Quickshot code base tries to follow some conventions a list of them is here https://wiki.ubuntu.com/ubuntu-manual/quickshot/style . [15:04] I will. I was kinda hoping you'd link to that so I wouldn't have to go looking. :) [15:05] no probelm the bot currently has these commands [15:05] ?help [15:05] Useage: + Examples: +website , +blueprint , +qssource , +involved , +question , +style , +use , +ppa , +quickshot , +code , +launchpad , +bug [15:06] except its not a + any more i need to change that its now ? people +1 ing was confusing it [15:06] Launchpad is an OpenID provider? Weird... [15:11] flan: so Pylons is to your liking? found any stumbling blocks? [15:20] I accidentally sent my first response directly to you, Luke. Please ignore my momentary stupid. [15:20] I'm liking it more as I read through its reference documentation, but I've yet to actually start gluing its pieces together, thorwil. [15:20] I'll be able to answer you this weekend. [15:20] Or maybe today. [15:21] cool [15:21] Since I plan to make the first connection between controller, template, and data-model soon. [15:21] It's looking pretty awesome, though. [15:21] As far as providing a flexible framework with enough standards to allow for collaboarative development goes, anyway. [15:24] i wonder if https://fedorahosted.org/moksha/, which adds to turbogears (which adds to pylons) will be useful. it seems it's too early, no adoption yet at all [15:25] This is a pretty simple system, really. Pylons, by itself, seems more than capable of handling it. [15:25] (I also want to keep dependencies light) [15:25] (Since this needs to run on one of our webhosts) [15:26] (While mine is super-Python-friendly, godby-k's is a little more conservative) [15:26] (I can easily handle the traffic for 10.10, though, if it comes to that) [15:29] it's quite sad how virtually every hoster supports php, but python? i found some who just list it as checkmark item somewhere, with no version and no word aboust wsgi. if at all [15:30] thorwil: I agree. [15:30] my host doesn't have good wsgi support, it seems. [15:30] (at least from what I've gathered) [15:32] It has wsgi support as a side-effect of a Rails module. [15:32] Which is good enough. [15:32] But it's not very happy-making. [15:32] Mine used to have the name "python-hosting.com", though. [15:32] Which is the main reason I looked at them. [15:33] Hi all [15:33] * flan makes decisions based on poor information. [15:33] Hi, you. [15:33] hello rickspencer3 [15:34] flan: true, but I don't know how good it actually is. [15:35] I'm working on the application developer manual blueprint [15:35] I read into it a bit before deciding to actually use Pylons, after thorwil's initial recommendation. [15:35] It looks like it works reliably and reasonably efficiently. [15:35] Nobody was complaining and there were a number of guides on setting it up. [15:36] I'm going to try to be as aggressive as possible with memory-management, though. [15:36] (Whether that means using disk for caching stuff or tuning the GC in some way) [15:37] But, again, worst-case, I can host this thing easily. And the public site will be written in PHP, since it'll be all-static, with the exception of its bulletin-board thing. [15:38] So that can easily live on your host. [15:39] can you guys help me out with the bluerprint? [15:39] I'm trying to figure out what tasks I'll need to do as editor of the manual [15:39] https://blueprints.edge.launchpad.net/ubuntu/+spec/desktop-m-opportunistic-developer-manual [15:51] A critical sense of judgment about what constitutes a best-practice approach and what tools are best-suited to accomplising tasks would be valuable. [15:52] i.e., not recommending Perl for cases where readability and maintainability are important (things that would likely be better-served by Python) and not recommending building an application framework from scratch when well-established frameworks exist that already do all of the common work. [15:54] Probably just knowing enough about available tools and technologies to correct the writers when they provide outdated information or repeat myths. [15:55] Is that the sort of thing you're wondering about? [16:28] rickspencer3, Dive Into Python has fallen out of favour (though I'd still recommend it to some people). [16:28] flan http://www.omgubuntu.co.uk/2010/05/daily-digest-windicator-mock-ups.html the e-mail is out :) [16:29] I can't remember what's currently popular, but you can find it in the topic of #python. [16:29] (I'm not in there because I get distracted far too easily) [16:30] Ooh. Windicator looks like a good use of the new whitepsace. [16:30] Lawl, me making typos. [16:30] Told you I'd miss something. [16:39] flan, yeah, I'm working on the work items now [17:10] Hey all. How are you. [17:11] hello, I am good you? [17:11] Ok. [17:12] I feel I have not really been on irc much recently [17:13] I haven't really paid much attention to what hasbeen going on. [17:14] I've seen your namy many times, without actually knowing who you are. [17:14] (I recently nickchanged from Red_HamsterX; I'm some Quickshot guy) [17:17] Hello [17:17] Hi. [17:18] I guess I am some Translation / whatever I can do kind of guy. [17:19] Why the change from Red_Hamster [17:19] It was awkward and silly. [17:19] (This handle was taken, though) [17:20] I guess it isn't now [17:20] Nope. [17:20] Just telling you who I am, in case you've been lurking. [17:21] Lurking.... That sounds scary [17:23] It just means someone who watches a discussion without participating. [17:26] I do like to participate. I have just been busy recently [17:27] Oh, I'm not trying to imply anything. [17:27] I'm just trying to minimize confusion. [17:27] And failing. [17:27] * flan fails a lot. [17:28] Never mind. I often find confusion occurs a lot while writing. [17:38] Summary: I used to be someone else. You did not know who I was. Therefore, no knowledge needs to be ported from one identity to another. [17:39] Summary: I was aware of you, so knowing about the name change was useful. [17:42] Just out of interest. What happened to the UK translation of the manual again. It appears that the translations have changed again to "NNeeds Review"..... again.... [17:43] iirc there was a reason it happend but i can't remeber what it was [17:43] Unfortunately, I don't know anything about how translations states work. :( [17:43] -s [17:43] flan: Error: "s" is not a valid command. [17:44] * ubuntujenkins gggrrr silly bot [17:45] I remember that godbyk fixed it last time. [17:46] I do also vaguely remember that there is a bug that Humphrybc has going to talk to the launchpad devs about at uds [17:46] I think that bug was addressed. [17:46] Whatever it was. [17:47] Someone announced that some Launchpad issue was fixed following a UDS discussion and then people cheered. [17:47] I thought the tempoary fix to that bug was to disable po editing and not allow manual imports [17:47] That sounds about right [17:48] Or my memory's broken. [17:48] * flan shouldn't comment on things that don't directly affect him. [17:48] It could also be my memory that is broken. [17:48] or mine :) [17:49] Bottom line though. Something has made it happen again [17:49] hhmm [17:50] I guess I will mention it to Godbyk when he turns up [18:09] o/ [18:34] Hi, dutchie. [19:04] thorwil: re he-mark, err, what? [19:16] dutchie: it's an exercise in restraint and seriousness [19:17] thorwil: what's with the darth vader-alike in the background? [19:18] dutchie: like i wrote on the blog, a koala-skeletor didn't work [19:20] ha [19:25] But Skeletor always wins! [20:12] i hate sony music, but it was not very wise of Ben to use some of "their" music for his UDS video [20:14] "their" referring to which label? [20:14] * flan hasn't watched it. [20:16] sony [20:39] godbyk-sagan: http://www.advogato.org/person/raph/diary.html?start=420 [20:43] Ooh. More fonts. [21:37] * ubuntujenkins likes it now the channel has a few more people in again [21:42] flan: as a wish list item for this release if we have time I would quite like to add apport support. This does tie us into 'buntu distros, may be we can add it as a option in the build. I am looking into how hard it is etc. It would make bug reports easier. What is your opinion on this? [21:43] you can always patch it in to the ubuntu client [21:43] lots of already-existing packages do that iirc [21:44] it looks like it is easy to do, only a handful of lines of code. I think it would make life easier. [21:45] When people submit bugs there is a list of things that would help us with it but people don't always read it. [22:03] We're going to be maintaining Ubuntu/UMP-specific patches. Making it part of the .deb build process should be pretty easy. [22:04] (I've added patches against .debs on the Debian side before) [22:04] cool, we also have to do a make file i guess [22:04] (It was generally just a matter of making sure the patches were in the right directories and indexed in the serial application file) [22:05] We'll worry about that when we get to it. :) [22:05] I'm sure there are other packages we can leech from. [22:05] (I've never actually done a Python makefile before, but they aren't so bad for C) [22:05] good, I have never done any patching stuff. I think this will be another steap learning curve :) [22:05] Patching's easy. =P [22:05] generally you use distutils for python [22:05] bzr branch, hack stuff, bzr diff > patch [22:05] rather than make [22:06] Yeah. [22:06] But ubuntujenkins is the only one of us who's built a .deb from scratch. [22:06] As far as I know. [22:06] I've just altered existing ones. [22:06] i think i have. once. a long time ago [22:07] I tend to cheet and use quicklys help there is abit of hacking needed on it. [22:07] Do we want to packages the server? [22:07] I have learnt allot about packaging what with the latex ppa [22:08] I'm not sure how well packaging the server would work. [22:09] I've seen a few PHP-based systems packaged and it's always been awkward to configure them. [22:09] Or, actually, no, that's just been the ones packaged by third parties. [22:09] We could probably use the Wikimedia packages or something as a reference. [22:10] I'll restructure the directories a bit once I have something working, to try to make that a viable idea. [22:10] I don't know it was just an out loud thought. It does rely on people having a "standard" setup [22:10] As for apport itself, I think that'd be a really good idea. [22:10] I hate the thought of bugs going unreported due to users being overwhelmed or too busy. [22:11] And Ubuntu's our main focus, so we might as well make use of its features to make out stuff better. [22:11] yea and it doesn't look like much work aprox 10 - 20 lines if i understand it right [22:11] I'm guessing it's just a top-level exception-grabber. [22:14] basically we have to get it to grab log files so, quickhsot must make a log file in /var/logs [22:14] thats my understanding of it atm [22:14] Got a link? [22:14] I spend 5 mins looking. emm handon ... [22:15] https://wiki.kubuntu.org/Apport/DeveloperHowTo [22:17] I think you misunderstood that part (which is good for us, 'cause it means we don't have to mess with /var/logs unless we have to) [22:17] ok i didn't read it properly i did skim it :) [22:17] The example is just adding the contents of a logfile to the report that'll be submitted. [22:18] But it's cool that it can do things like that, since it means we can add whatever we want to the report. [22:18] Which means we can probably implement some sort of step-tracer in the design. [22:18] Like an in-memory log. [22:19] Every time the user advances through a dialogue box or clicks something, we could log that event, like what we were doing with the --debug flag in 0.0.8. [22:19] thats what i was going to ask about [22:19] If it dies, we could just attach that thing as a big string. [22:20] And then trace every single step. [22:20] I think this would be good in so many ways [22:20] Agreed. [22:20] Excellent idea. [22:20] * ubuntujenkins adds it to the todo list [22:21] Must-have. [22:21] added it to my list [22:22] I am liking tracks for my organising [22:22] I'm still working off of http://pad.ubuntu-uk.org/f0VIdaLXWZ [22:23] I think ther si a page on the wiki .... [22:23] ?plan [22:23] Factoid 'plan' not found [22:23] https://wiki.ubuntu.com/ubuntu-manual/quickshot/next-release-plan [22:25] Oh, but it looks like the hook doesn't share memory with the client... [22:26] (i.e., it's launched externally) [22:26] (Which is kinda good, in that it means we won't need to change our core code to make use of it) [22:27] We'll probably have to write a per-session activity log to .quickshot/debug/log or something, then. [22:27] thats a good place to put it [22:32] this happended to be in my inbox https://wiki.ubuntu.com/MeetingLogs/devweek1001/DetailingBugs [22:35] hello @all [22:35] hey daker [22:37] I've been interested in Apport for a while, but I do most of my development projects with ultra-conservative Debian as a target platform, so... [22:37] this will be a fun to add [22:38] It'll be well worth however long it takes. [22:38] (And, as you said, it looks easy) [22:39] I agree [22:41] Wait, what? SunK's in Calgary, too? [22:47] hey flan [22:47] yeah I'm [22:47] * flan was formerly known as Red_HamsterX. [22:47] (I finally got this name freed up) [22:47] wow two people on the quickshot team in the same place [22:47] Yeah. Kinda strange. [22:48] But Calgary's a big place. [22:48] * flan answers the looming question first. [22:48] Midnapore. [22:51] i'm new to Calgary [22:51] use to live in Vancouver [22:51] used* [22:52] That doesn't tell me what part of the city you live in. :( [22:52] How am I supposed to mock you now? [22:53] haha Panorama Hills [22:53] ...Well, I can't really mock that. [22:53] ubuntujenkins: in fairness, we're not *that* far apart [22:54] true dutchie, I wich i could have gone to ogg camp [22:54] ubuntujenkins: next one may be down south [22:54] I am on placemnet next year so i have a car. So fingers crossed I may make it