=== jalcine is now known as webjadmin_ === jalcine is now known as webjadmin_ === rye_ is now known as rye [09:29] Good morning, all! :D [11:03] good morning! [11:26] gatox, morning! [11:26] mandel, hi [11:26] gatox, I added the test to heck if the app name is correct [11:26] gatox, can you approve it now ;) [11:27] mandel, yes! can you give me the link [11:27] gatox, https://code.launchpad.net/~mandel/ubuntuone-control-panel/add-app-name/+merge/97071 [11:56] gatox, what is the lib we should be using instead of xdg? [12:00] mandel, mmmm not sure about the lib, what i know if that info we should get it from sso [12:01] gatox, ok, I'll look there then === jalcine is now known as webjadmin_ [12:52] gatox, I'm off to have lunch, will be back in 30 mins mor or less (depends on the waiter) [12:52] mandel, ok..... i see you have fanzy lunchs [12:52] jejeje [12:53] gatox, on tuesdays I go with the people at the office :P [12:53] ahhhh [12:58] good morning! [12:59] ralsina, hi! === webjadmin_ is now known as jalcine [12:59] hello gatox! === m_conley_away is now known as m_conley|dinner [14:02] me [14:06] me [14:06] oh, I think we're an hour off [14:06] stupid DST [14:06] BACK! [14:06] welcome back, briancurtin [14:06] hey urbanape [14:07] progress was not exactly made in your absence. [14:07] * mandel back! [14:08] briancurtin, urbanape you are hehehe [14:08] so the standup is in one hour, or it already happened (i'm horrible at time changes) [14:08] briancurtin, in an hour :) [14:08] briancurtin, it is at 3 pm utc [14:09] nice. even though it's only ever +1 or -1, it becomes incredibly hard math for some reason [14:09] briancurtin, the trick is to set the time indicator to have both and if you use the company google calendar to set the alarm to use utc [14:09] ralsina, ping? [14:10] i need a sun dial [14:11] urbanape: no biggie. where we left off with the FS stuff and what its capabilities are was not too encouraging of it being a nice easy effort [14:12] briancurtin: the trick is to just be a slacker and wait for nessita to ping you about the standup :) [14:12] hello all! [14:13] alecu, hi! [14:13] hola alecu [14:14] alecu: why should the webclient calls go through the proxy tunnel process, instead of just direct through the proxy? [14:15] dobey, only the webclient calls made from syncdaemon go thru the tunnel. [14:16] dobey, otherwise we would need to use either qtnetwork or libsoup from syncdaemon. [14:16] dobey, she is not working today [14:16] alecu, dobey what happened with the openssl issue? [14:17] mandel: pong [14:18] mandel: ralsina mailed legal afaik [14:18] ralsina, I was going to ask you about openssl :) [14:18] dobey: I mailed cristian and chipaca who are near legal [14:18] I will bring it up in today's mgmt talk and then it will probably go to legal [14:20] welcome back briancurtin! [14:20] alecu, dobey any idea on how to use openssl to get the hash used to sign a cert? [14:20] hello all :) [14:21] briancurtin: in about 2 hours we need to talk a lot about prepping a windows release for next week [14:21] ralsina: sounds good, can mumble if you want [14:21] briancurtin: cool, so in 2:09 save me a spot [14:21] briancurtin, ralsina: make sure we get lots of testing on the proxy stuff on windows [14:22] briancurtin, ralsina: and also, I think we should make a new .exe with the tunnel process used for proxying syncdaemon. [14:23] mandel, what ssl "hash" do you need? [14:23] alecu, second point from here: https://developer.mozilla.org/En/Cert_override.txt [14:23] alecu: ack [14:23] alecu, or I can simply ignore it [14:23] mandel, "hash algorithm OID" ? [14:24] mandel: i don't know [14:24] mandel, perhaps you need point 3? point 2 is the ID of an algorithm [14:24] alecu, yes, although I can get the OID from that list, I'm more interested in the hash algorithm SHA1 etc.. [14:24] alecu, yes, I wanted to first make sure it was hard to get that so that I had a reason to ignore point 2 [14:25] ralsina: about ssl? [14:25] ralsina: or about what, re legal? [14:25] briancurtin, ralsina the integration tests for squid3 do not work in windows.. in theory everything should work ok.. [14:25] Chipaca: yes siree, I did [14:25] Chipaca: about having to add an exception to our licenses to allow linking openssl [14:25] right [14:25] yep [14:26] ok, pre-meeting break for me [14:26] mandel: what are we doing with cert_override.txt? [14:26] dobey, pinned certs until there is support for that on thekeyring, that is a ref of a text files that does the same [14:26] dobey, so that I don't have to reinvent the wheel [14:27] mandel: so you're implementing the same file format? [14:27] dobey, yes [14:27] mandel, why do you need the exact same format? [14:28] alecu, I don't need it, I wanted to make it as close as possible to just say, look we do it like that [14:28] mandel: i don't think openssl knows about that format. you'd have to handle converting to that OID and back [14:28] alecu, nothing else, I can always ignore the has algo and just keep the fingerprint which is more than enough [14:29] mandel, so, what class of objects do you get from qtnetwork or libsoup? [14:30] alecu, I get the ssl pem which I can convert to a openssl object [14:31] mandel, alecu do you guys know if launchpad is disabled in this moment or something?? i can't create a new branch [14:31] sigh, nessita. [14:31] lol [14:31] she claimed a review on my branch, and hasn't reviewed it, and is off today [14:31] dobey, half day [14:32] dobey, she is reading the emails [14:32] dobey, do you know if there is any problem or something with launchpad? [14:32] mandel, like this, right? QSslCertificate::toPem() [14:33] gatox, afaik there is not [14:33] i'm trying to submit a new branch and i can't [14:33] :S [14:33] alecu, yes and raw_cert.props.certificate_pem [14:33] gatox: not afaik [14:33] mandel, great. [14:33] gatox: submit as in push, or as in propose? [14:34] alecu, I'm using pem to centralize the code that gets the details rather than using each of the libs [14:34] mandel, and what are you using to turn those into pyssl objects? [14:34] mandel, that sounds right [14:34] dobey, as in trying to create the branch from the web page with "submit new code" [14:34] alecu, from OpenSSL.crypto import load_certificate, FILETYPE_PEM [14:34] alecu, then load_certificate(cert_pem, FILETYPE_PEM) [14:34] gatox: oh i think something changed on lp recently and the "create a branch" link on the web site doesn't work [14:35] dobey, yep..... i see [14:35] mandel, the args are reversed, right? [14:35] dobey, i'll do it manually [14:35] mandel, load_certificate(type, buffer) [14:35] gatox: if you just make a local branch and push it, it works [14:35] ralsina: can we move the windows call before or after you were planning it? at about 16:40 UTC i need to go to a doctor's appt [14:35] alecu, yes, sorry :P [14:35] mandel, great. [14:35] gatox: manually? using the web site is "manual." why would you ever use the link on the web site to create a branch? [14:35] alecu, said it from the top of my head [14:35] briancurtin: after is ok, just ping me when you have a slot [14:35] i hope it goes away for good [14:36] dobey, i was used to that [14:36] mandel, and that returns a X509 object? [14:36] alecu, yep, which has all the info (O, CN, L, ST etc..) plus the fingerprint [14:37] alecu, CN is usually the domain name (Common Name) [14:37] mandel, the fingerprint should be X509.digest("sha1"), right? [14:38] I'm looking at: http://packages.python.org/pyOpenSSL/openssl-x509.html [14:39] alecu, hmm let me double check, I think you are right and I just had a major brain fuck.. [14:40] alecu, yes, you are right.. oh my what an embarrassment, I'll take it form here :) [14:44] ralsina, when you have a moment please: https://code.launchpad.net/~diegosarmentero/ubuntu-sso-client/check-cut-off/+merge/97231 (qt thing) [14:44] gatox: sure, but I am on call for a while [14:45] ralsina, no problem, is not urgent [14:45] gatox: and alecu's branches have priority (if they still need reviews) [14:55] gatox, reviews are veeeery welcome! let me find the link. [14:56] alecu, i have both of your branches opened already [14:56] gatox, cool [14:56] alecu, i'll do both now [14:56] gatox, the first one to review is the ussoc one [14:56] alecu, ack [14:56] gatox, since the u1-client branch depends on that ussoc branch [14:57] gatox, also, for the u1-client branch it would be right to do some IRL testing. Let me know when you are reviewing it and I can give you some tips on how to test it IRL. [14:59] alecu, great, thanks [15:00] me [15:00] me [15:01] me [15:03] me [15:03] alecu, dobey ? [15:03] me\ [15:04] mandel, go.... dobey is last, and i think ralsina is in a cal [15:04] call [15:04] DONE: Added missing tests for appname branch in control panel. Refactored the webclients to certralize where the cert info is found by passing the pem to the base class. [15:04] TODO: Finish pinned cert file implementation (had a stupid brain fuck moment). Propose move to next bug. [15:04] BLOCKED: no [15:04] * mandel passes the rugby ball to briancurtin [15:04] oh oops [15:04] yes, I skip [15:04] DONE: pycon [15:04] TODO: pick up where i left off, catch up on email, mumble with ralsina, work towards windows release [15:04] BLOCKED: none [15:04] NEXT: gatox [15:04] DONE: [15:04] Review for nessita, Proposed a branch for the enhanced check box being cut off at the right border. [15:04] TODO: [15:04] alecu's reviews and keep fixing more ui bugs. [15:04] BLOCKED: [15:04] No [15:04] urbanape, go [15:04] DONE: Got the release for U1 Music iOS out. [15:04] TODO: Back in the saddle. [15:04] BLOCK: My brain. [15:04] alecu: you're next [15:04] DONE: finished with bug #929212, branches up for review [15:04] TODO: use authentication in tunnel [15:04] BLOCKED: no [15:04] NEXT: dobey [15:04] Launchpad bug 929212 in Ubuntu One Client "[FFE] Tunnel webservice calls if proxy is enabled" [High,Confirmed] https://launchpad.net/bugs/929212 [15:05] λ DONE: poked more about installer image, bug #953062, bug #942025 [15:05] λ TODO: finish installer, bug #951425, bug #934206 [15:05] λ BLCK: none. [15:05] Launchpad bug 953062 in ubuntuone-control-panel (Ubuntu) "Font and sizes are hard-coded" [Undecided,New] https://launchpad.net/bugs/953062 [15:05] Launchpad bug 942025 in Ubuntu One Control Panel "The email address font is too small" [Undecided,Confirmed] https://launchpad.net/bugs/942025 [15:05] Launchpad bug 951425 in Ubuntu One for Rhythmbox "Ubuntu One plugin has problems when being enable/disabled" [Undecided,Confirmed] https://launchpad.net/bugs/951425 [15:05] Launchpad bug 934206 in rhythmbox-ubuntuone (Ubuntu Precise) "track duplication from RB-U1" [High,Confirmed] https://launchpad.net/bugs/934206 [15:07] urbanape, briancurtin: I've been doing some reading on FSevents and /dev/fsevents [15:07] urbanape, briancurtin: did you end up deciding which one to use? [15:08] briancurtin, small question about open on windows, does it make any diff between a and w when opening a file? [15:08] I'm still leaning towards watchdog, but I'd like to see how it does with file events in large directories. [15:09] urbanape, by watchdog you mean this one? http://support.apple.com/kb/TA20622?viewlocale=en_US [15:09] nope [15:10] mandel: you mean when opening an existing file or opening a new file with 'a' or 'w'? [15:10] alecu: http://packages.python.org/watchdog/ [15:10] abstracts FSEvents, inotify, and the Windows stuff. [15:10] briancurtin, both [15:10] (but we'd only use it for OS X right now) [15:10] urbanape, ok. [15:10] mandel: i'm not aware of anything in that area that operates differently on windows than it should on *nix [15:11] just the typical append vs. write behavior [15:11] urbanape, it seems to use FSEvents as the backend, so it will surely have the same drawbacks regarding changes in files... right? [15:11] briancurtin, ok, then I'll toss a b there just in case [15:12] alright, i should get some lunch. bbiab [15:13] urbanape, and I don't think kqueue looks useful, since it would need to open for monitoring each and every file in the Ubuntu One folder plus the UDFs... it does not scale. [15:14] alecu, urbanape I would not even get close to kqueue [15:15] altough it make the port to bsd closer! [15:16] alecu: yes (about FSEvents). [15:16] mandel, "open eeeeeevery file in the folders watched by SD" [15:16] mandel, it makes no sense for us! [15:16] alecu, I know, that is why I would not get close to it :) [15:16] in my opinion, that it abstracts away inotify and the Windows system means it could *potentially* be used on all platforms (and become something we don't have to own) [15:17] * gatox lunch...... brb === gatox is now known as gatox_lunch [15:20] urbanape, that sounds reasonable, but I doubt that having to rehash every file in a directory every time one file changes would perform reasonably well. [15:20] urbanape, from what I've read it might make sense to have a small daemon reading everything from /dev/fsevents and filtering what's interesting to us [15:20] even though it has to run as root :-( [15:23] I wonder if Apple ditched folders in iCloud because move events are typically handled so badly. [15:23] urbanape, briancurtin: you guys have probably read this ars article that talks about FSEvents and /dev/fsevents: [15:23] http://arstechnica.com/apple/reviews/2007/10/mac-os-x-10-5.ars/7 [15:23] alecu: havent seen it, thanks for the link [15:25] briancurtin, that's the article that convinced me that we should be using /dev/fsevents if we want to have decent performance for this. [15:25] ooh, mentions of NEXTSTEP already - i used to use GNUStep as a (graphical) reminder of NeXT days :) [15:46] just read: Yo no digo que esa chica sea un poco suelta, solo digo que si su entrepierna tuviera contraseña, sería '1234' [15:46] sorry, I'm to lazy to translate it :P [15:51] alecu: you still needing reviews? === gatox_lunch is now known as gatox [15:51] ralsina, looks like I still do! for the two branches "ready for review" here: https://bugs.launchpad.net/ubuntu-sso-client/+bug/929207 [15:51] Ubuntu bug 929207 in Ubuntu One Client "[FFE] Proxy "tunnel" for syncdaemon" [High,In progress] [15:52] alecu: ok, doing one [15:53] ralsina, btw: I've just got a call from the state sponsored kinder right around the corner, that there's an opening, so we'll be changing Amelia's kinder. That means that my schedule will be hell for the next couple of weeks... :P [15:54] alecu: well, it's probably for the best long term [15:54] having kinder around the corner is niiiice ;-) [15:54] ralsina, for the shorter term too, since we'll skip paying the private kinder :-) [15:54] ralsina, also: the -sso branch goes first, since the u1-client branch depends on it. [15:55] alecu, this one gave me a lint issue: https://code.launchpad.net/~alecu/ubuntu-sso-client/updated-txweb/+merge/97119 [15:55] gatox, can you please paste it? [15:55] alecu, just this: [15:55] == Python Lint Notices == [15:55] ubuntu_sso/utils/webclient/common.py: [15:55] 247: [W0511] TODO [15:56] alecu, except for that..... it's a +1 from me [15:56] gatox, I'll fix that, and thanks! [15:56] alecu, let me know, when it's fixed so i can approve it [15:56] gatox, are you brave enough to take on the following branch? [15:57] alecu, of course..... i'll start the review of the other one right now [15:58] gatox, the funny thing is that my branch does not touch that file :P [15:58] gatox, can you check if that lint notice also shows up on trunk? [15:58] alecu, yes..... just a minute [15:59] alecu, yep..... it's in trunnk [15:59] alecu, approving [16:01] gatox, that TODO is probably being fixed in one of mandel's current branches. [16:01] gatox, alecu yes, is begin fixed, and TODOs do not brake tests, they are just printed out [16:02] aka u1lint returns 0 [16:02] alecu, no problem..... i already approve your branch, now i'm branching the other one, run the test, review the codem and i'll ping you if i need to test it IRL [16:04] alecu, i haven't done it in a while.... for u1-client with an specific u1-sso..... is it ./autogen.sh --uses=path/to/sso = [16:04] gatox, I'm finishing with the review of your branch, and I also got lint warnings! [16:04] alecu, my branch? [16:05] ./autogen.sh --with-sso /home/alecu/canonical/ubuntu-sso-client/branch_name [16:05] gatox, ^ [16:05] alecu, thanks!...... which branch? [16:05] gatox, make sure to include the whole path (it sometimes not work with ~/) [16:05] alecu, yes [16:05] gatox, check-cut-off [16:05] gatox, http://pastebin.ubuntu.com/882006/ [16:05] alecu, weird.... it didn't show anything when i run the tests here [16:05] ralsina: i'm a day off - doc appt is tomorrow. i'm free for the call when you are [16:06] briancurtin: ok, I need to wrap up a few things first [16:06] no problem [16:06] gatox, that's on precise, updated yesterday (I think) [16:06] alecu, i'll update again [16:06] alecu, thanks === beuno is now known as beuno-lunch === beuno-lunch is now known as beuno [16:23] briancurtin: want to sync up after your windows release call? [16:23] urbanape: that sounds good [16:23] cool [16:29] hmm [16:43] briancurtin: mumble? [16:43] ralsina: logging in [16:50] urbanape: i'm in mumble whenever you have a few minutes [16:50] cool, brt [16:51] k, I'm in Desktop Plus === yofel_ is now known as yofel [17:02] alecu, which would be the proper way to test your u1-client branch IRL? [17:03] gatox: in a manner not totally unlike that of advancing the story in Monkey Island [17:03] dobey, what? [17:03] :P [17:03] heh [17:04] gatox, look behind you! A three headed monkey! [17:05] dobey, how appropriate, you fight like a cow. [17:05] heh [17:05] jejee [17:05] gatox, so: first you need to install a squid [17:05] That's the second biggest stone monkey head I ever saw! [17:05] gatox, then you need to use syncdaemon in the branch, both thru the proxy and without using the proxy. [17:05] alecu, ah...... it's not going to be easy..... [17:06] gatox: use my squid! [17:06] ralsina, ah! right! [17:06] gatox, a default squid install should be fine. [17:06] ralsina, how can i do that? [17:06] gatox: configure proxy ip address 184.82.108.14 [17:06] gatox, if you install the squid3 package it starts a proxy on localhost:3128 [17:06] gatox: port 8888 [17:06] first you ask, then you find out alecu is sad, and then you do a bunch of stuff and find out he needs a beer, so you buy him a beer, then you find out you need squid, but squid is unhappy, so you have to reconfigure it [17:06] gatox: and let me set a user/password for you [17:07] alecu, and then you ask why it's always difficult to get reviewers!! [17:07] jejejeje [17:10] alecu, ok, i'm using ralsina proxy now [17:12] gatox, ok, now you need to start Syncdaemon like this: [17:12] U1_DEBUG=True PYTHONPATH=.:~/canonical/ubuntu-sso-client/updated-txweb/ bin/ubuntuone-syncdaemon --debug [17:13] gatox, (before that you should configure the proxies in the system settings, of course) [17:14] alecu, done [17:15] gatox, once it's running, try doing "u1sdtool -c" so it connects. Then "u1sdtool -s" to see that it has reached the QUEUE_MANAGER state [17:16] alecu, yep! [17:16] alecu: we support http_proxy=http://user:pass@1.2.3.4/ right? [17:17] ralsina, env vars? probably not yet. [17:17] gatox, u1sdtool --publish-file=~/Ubuntu\ One/folder/file.txt [17:17] gatox, and check the web to see if the file was published. [17:18] gatox, u1sdtool --offer-share ~/Ubuntu\ One/folder/ gatox@gmail.com shared_folder view [17:18] gatox, and check the web or your email to see if the folder was offered. [17:19] alecu: ack [17:19] alecu, yes! [17:21] alecu, it's working! [17:21] \o/ [17:21] yay! I have now gatox's credentials! Nah, just kidding ;-) [17:21] * gatox faints === 17WAAYI1Q is now known as zyga [17:22] gatox, finally.... we need to check get_public_files, but it does not have an u1sdtool option [17:22] ok, I need to go EOD here, catch you all tom! [17:22] ralsina: what the heck is setFontWeight() in Qt? [17:22] mandel, bye [17:22] alecu, so...... what we do that? [17:23] alecu, if you need any reviews tom CET morning let me know [17:23] via email [17:23] dobey: allegedly, thing to set the font's weight [17:23] mandel: you moved one hour earlier? [17:23] ralsina: i don't think we're using that right [17:23] dobey: we should not be using that [17:23] gatox, I usually do it with d-feet [17:23] ralsina, ein? I always go at my 6:30 on tuesdays.. [17:23] dobey: without a VERY good reason [17:23] ralsina, isn't that something you use to set the font from normall to bold or something likke that? [17:23] ralsina, i don't remember exactly [17:24] gatox, do you have d-feet installed? [17:24] alecu, yes [17:24] mandel: I was wondering if you aremoving to DST or something [17:24] ralsina: i am seeing a bunch of it in sso. i guess i'll just remove them too [17:24] ralsina, not yet :) [17:24] gatox, great. go to com.ubuntuone.Syncdaemon in the session bus [17:24] gatox, and look for /publicfiles [17:24] alecu, i'm already there [17:24] ralsina, I'm still UTC +1 [17:24] mandel: ok, thanks [17:25] gatox, execute this method: iface: com.ubuntuone.SyncDaemon.PublicFiles method: get_public_files() [17:25] alecu, should i get something? it says: 'This method did not return anything' [17:25] ralsina, there is like 4 hours earlier, right? [17:25] mandel: 2:25PM right now [17:25] mandel: so yes [17:26] gatox, yes. But the sd log would show the public files [17:26] alecu, and the console says: 2012-03-13 14:25:19,306 - twisted - INFO - Stopping factory [17:26] gatox, (they are returned not in the call, but in a separate signal) [17:26] gatox, something like: AQ_PUBLIC_FILES_LIST_OK, kwargs: {'public_files': [{'public_url': 'http://ubuntuone.com/1IpJvcAhD0zV2MjOC3k2sl', 'node_id': '6fbb435c-2650-4d15-9480-cd1b1ba38dae', 'volume_id': ''}, {'public_url': 'http://ubuntuone.com/7kiDLJg2SeW3F17WrQiADx', 'node_id': 'fb71d997-d9ac-439e-8cd5-41dda0a1e6cf', 'volume_id': ''}, {'public_url': 'http://ubuntuone.com/4cpiEexLtKKvf2F6j079VD', 'node_id': '2b102f89-f5dc-4f49-85d4-38131fad3583', 'volume_id [17:26] ': ''}]} [17:26] ralsina, ok, catch you tom, I need to walk the dog, visit the brother and rugby.. [17:27] gatox, if that shows up on SD logs, it means that the webcalls worked ok. [17:27] gatox, and to finish the IRL test, stop SD, disable proxies, and repeat all three calls. [17:28] gatox, I need to *run* to the new kinder. I'll catch up in an hour. [17:28] alecu, that work [17:28] \o/ [17:28] ralsina: ah, for some reason the .ui files had a bunch of 75 in them, which didn't seem right, and the generated _ui.py files thus had setFontWeight(75) which made no sense [17:28] alecu, i'll test again and approve it...... GREAT WORK! [17:28] alecu, again without the proxy i mean [17:28] gatox, graaaaacias! [17:28] COOL [17:33] dobey: hmmmm [17:33] dobey: yes, that weight isnot default [17:34] ralsina: i would presume it is supposed to mean "make it bold" [17:34] dobey: not even [17:34] dobey: bold is 100 IIRC [17:34] ralsina: all the affected cases also had true [17:35] http://qt-project.org/doc/qt-4.8/qfont.html#Weight-enum says 75 is bold [17:37] ralsina: so… https://code.launchpad.net/~dobey/ubuntu-sso-client/system-font/+merge/97264 [17:37] dobey: ok, then [17:37] ralsina: sso also will need a separate qss for windows [17:37] dobey: affect it on the other bug [17:37] yep, am about to [17:37] So far, it's just QApplication.instance.setFont("Ubuntu") [17:39] dobey: will review after lunch [17:39] sure [17:39] all add nessita to it also [17:39] yes please [17:40] you know, if we had done nothing in the way of features this cycle, we at least made a huge step forwarad in managing bugs and releases, i think [17:41] +1 [17:43] also, yet another branch: [17:43] https://code.launchpad.net/~dobey/ubuntuone-client/man-typo/+merge/97259 [17:51] dobey: +1 and I think you can merge it with one review [17:53] aye, it's trivial === m_conley|dinner is now known as m_conley [18:21] ralsina: btw. openssl exception? :) [18:21] dobey: gimme 5' [18:21] sure [18:27] dobey: stupid question, how can I see the UI in your SSO changes? Start that SSO and try to re-login? [18:29] you can run both bits from the tree, and just run ubuntu-sso-login-qt with args for a different app name and such, i guess [18:34] dobey: can't see the changes because with the ubuntu font everything is bold ayway [18:35] dobey: removeing the ubuntu family from the Qss, I get that the page titles are now not bold. So I think you should undo that. [18:36] hrmm [18:36] i didn't un-bold anything that was set to bold [18:36] afaik anyway [18:36] ralsina: i wonder if it was only bold by accident before? [18:37] No, I think that was intentional [18:37] dobey: the titles are now less prioritized than the field labels, let me show you [18:37] dobey: http://ubuntuone.com/1Ckpt7IpnHToCZ0sRGE4Z2 [18:38] ralsina: hrmm. +TITLE_STYLE = u'' should maybe be x-large or xx-large then? [18:39]