[00:22] I would love to have an RSS feed of the status page so that I don't have to keep looking for new information about the couchdb issues. [00:38] Have a great night pplz! [08:38] morning all! [08:43] Morningses! [09:01] fougout [09:01] LOGOUT === zyga-afk is now known as zyga [10:40] * mandel afaik for 30 min === mandel is now known as mandel_bbl [10:41] * fagan needs some tea === teknico is now known as teknico_away [10:56] :/ they delivered my package to somewhere else [10:56] Just not my week so far [11:34] hello everyone! [11:34] hey nessita [11:35] hi fagan [11:35] mandel_bbl: when you come back, let me know [11:37] ralsina: you around? [11:37] nessita: he didnt say morning so id say not yet [11:37] fagan: thanks [11:39] nessita: ping === mandel_bbl is now known as mandel [11:40] mandel: pong [11:40] mandel: hola! [11:40] fagan: did say morning, but you were not here ;) [11:40] nessita: buenas! [11:40] mandel: I meant ralsina didnt say morning :) [11:40] fagan: oh, hehe [11:40] mandel: had to move my computer and shout at some people about losing my package [11:41] * fagan angry [11:41] fagan: was out of context, sorry :P [11:41] mandel: so, I have a couple of questions re the fix-os-helper branch. Did you re-propose that already? [11:41] mandel: its cool :) [11:41] mandel: so I can base my question in a LP diff (if not, no problem at all) [11:41] nessita: no, is basically the one you have, so shall we use yours or mine? [11:42] nessita: I can propose it if you want… is no problem waht so ever [11:42] mandel: mine, I added some aesthetic changes as well, fixed some typos, and the like [11:42] mandel: but I (maybe) wanna request some changes based on some questions [11:43] nessita: I used yours and recreated all branches based on that, so all your changes should be there [11:43] nessita: did you get my morning report? [11:43] mandel: I did, I read it, and that's why I'm asking: [11:44] did you propose fix-os-helpers? I mean, that branch has a good size now to propose [11:44] nessita: no, lets do it now [11:45] nessita: if you have comments I can fix them there and pump the changes up to the other branches [11:46] mandel: yeah, I have some comments. I'll be using my branch to base comments on, is that ok? [11:47] nessita: lp just gave me this error: 'There is already a branch merge proposal registered for branch lp:~nataliabidart/ubuntuone-client/fix-os-helper to land on lp:ubuntuone-client that is still activ' [11:47] mandel: that is your proposal, since my branch was built on top of yours [11:48] o no? [11:48] I never proposed that branch for review [11:49] mandel: you proposed my branch for review! :-) [11:49] nessita: no idea, it looks like I propose it for you… can you do that in lp? [11:49] that is weird... [11:49] mandel: seems like it [11:49] let's remove that proposal [11:49] mandel: so. you branch from my branch, push to a location of yours, and re-propose [11:50] nessita: on it [11:50] mandel: you removed the old proposal??? [11:50] old == https://code.launchpad.net/~mandel/ubuntuone-client/fix-os-helper/+merge/68227 [11:51] yes [11:51] mandel: never remove proposal with comments in it, just mark it as rejected [11:51] nessita: ok, did not know that... [11:51] nessita: the new is here: https://code.launchpad.net/~mandel/ubuntuone-client/fix-os-helper/+merge/68530 [11:52] great, looking [11:54] mandel: ok, my first concern is the name of the method remove_utf8_chars [11:54] not correct? [11:54] mandel: the name is misleading, if I understood correctly, it should be something like remove_unicode or map_unicode_to_utf8_chars [11:55] mandel: since the method grabs unicode objects and replace it with bytes encoded with utf8 [11:55] nessita: what about, readd_illegla_chars? [11:55] illegal* [11:55] restore_illegal_chars can work [11:56] nessita: and the other one would be, remove_illegal_chars [11:56] exactly [11:56] so we know one those the opposite of the other [11:56] yes [11:57] nessita: ok, np [11:57] mandel: so, next question is in the file filesystem_notifications.py. Why the ex-longpath decorator was removed from _add_single_watch and rm_path? [11:57] and from add_watch as well? [11:58] mandel: is making me 'noise' that the decorator was there before, and now, after the rename to windowspath, is not longer used in those methods [11:58] nessita: let me look at the context to give you a decent answer [11:58] thanks [12:03] the issue is that rm_watch is called by the watch manager rm_watch, that has two problems, one, that we are mangling the path twice, two that our asserts will compain because in the second pass of the decorator we are passing unicode and not bytes [12:04] * fagan brea k [12:04] the assert complaining is not a problem but a relief, since mangling the path twice is bad [12:05] mandel: I don't understand why rm_watch is called twice [12:06] nessita: so we have to rm_watch, one from FileSystemMonitor and the other from WhatchManager, FileSystemMonitor is the public interface used by sd, therefore passes get byte, the other is internal [12:07] which is the watchmanager and has no decorator, before the asserts there was no problem, now when we have the decorator in both the asserts fail [12:07] hey folks, is it possible to publish photos synced from my phone and some how grab all the public url's? So I can chuck a simple html file in the front to show a gallery of my photos [12:07] mandel: ok, but are we calling rm_watch twice for the same path? [12:08] j0nr: hi there, let me ping someone that knows a lot about that [12:08] karni: you around? [12:08] nessita: always! [12:08] ;) [12:08] karni: hola! can you please give j0nr a hand? [12:08] sure [12:08] thanks nessita [12:09] lemme see [12:09] j0nr: you're welcome! [12:09] mandel: so, not sure if you understand my question [12:09] nessita: no, we are calling FileSystemMonitor.rm_watch which calls WatchManager.rm_watch [12:09] nessita: yes, why we removed the decorator, answer: because the decorator is applied on FileSystemMonitor method [12:09] I may be thinking outside the 'storage' box here, but I have done simple php scripts to fetch all images in a directory and create a simple gallery. Just wondering if I could somehow do the same with U1 so my gallery is always showing latest pics synced [12:10] whic follows the FileSystemMonitor interface from linux, which has more than just calling the wathcmanager [12:10] mandel: so, you said that FileSystemMonitor receives bytes, right? [12:10] rye: Could you give j0nr the link to your U1 gallery script please? [12:11] j0nr: ↑ I think this is closed to what you have in mind [12:11] karni: but the random url of published photos seems to make this difficult... its not like I can loop throu photos ammending their names to www.ubuntuone.com/j0nr/photos/... [12:11] nessita: yes, and passes unicode to WatchManager [12:11] j0nr: People complained about easily guessable urls. We can't have it that easy. [12:11] so WatchManager.rm_path should not get byte but unicode [12:11] mandel: where is that code? [12:11] karni: oh ok cool. so can I run php scripts on U1? [12:11] j0nr: That's not what I meant :) [12:12] mandel: I mean filename and approx line [12:12] one moment [12:12] nessita: ubuntuone.platform.windows.filesystem_notifications line 806 [12:12] that would be the FileSystemMonitor method [12:13] mandel: the conversion to unicode is in the decorator? [12:13] j0nr: rye will help you when he's around. He has a script that you can run locally, to generate a nice gallery. [12:13] I think ralsina also had one! :) [12:13] nessita: yes [12:14] nessita: so that when we move to unicode, we remove the decorator [12:14] mandel: ok, I see. SO why the add_watch right next to rm_watch does not have the decorator? [12:14] line 822 [12:14] mandel: the inconsistency is what is killing me :-) [12:14] (in order to process this0 [12:16] karni: cheers... does it involve having to have the photos synced to a local machine? [12:16] nessita: hehe, I know you… and you are right, there the problem is that Watch in the __init__ is performing the same operation, lets make the init do not care and add a decorator there, does that sound better? [12:16] j0nr: It's one-time run thing. It's not umm.. dynamic. [12:16] j0nr: And I'm not sure about the answer to your question. [12:17] j0nr: I think it's possible. But once you generate it, you can simply put the page on a server [12:17] karni: ok thanks :) [12:17] and it will access your files via public urls [12:17] So just hunt these people ;) [12:18] karni: so I take it there is a way to interogate U1 files to discover if they have a public url. Can this be done via command line? [12:18] mandel: indeed [12:18] can I sync U1 to a server? I.e. no GUI [12:18] nessita: ok, lets add that in the comments and lets take care of it! [12:19] mandel: same question for freeze_being vs freeze_rollback and freeze_commit [12:19] j0nr: run u1sdtool in termnal and read up on --info=PATH switch, maybe it contains the public url :) [12:19] mandel: I'll start adding the comments to the MP, but I have more things I want to talk with you from this branch [12:19] j0nr: you can certainly publish from command line, I never grabbed public link sthat way [12:20] nessita: ok, lets talk about everything before I start typing like a mad man [12:20] mandel: yes. So, whta about freezes? [12:20] "same question for freeze_being vs freeze_rollback and freeze_commit" [12:21] karni, nice one :) [12:22] nessita: events should have paths in byte, so commit which takes events should not have a decorator, agreed? [12:23] ok [12:23] same for rollback and begin, but begin has the deco [12:24] yes, and that should be wrong… since its calling the shared class by windows and linux which is in sd and should be using byte... [12:24] mandel: ^ === teknico_away is now known as teknico [12:25] nessita: ^^ [12:25] nessita: two up, or which ever got first [12:25] nessita: we have not been bitten by a bug there because the notification from windows are a little smarter regarding moves, which is what the freeze is for [12:25] so, as I'd say in spanish, pura suerte [12:26] mandel: ok, so I think this class needs a careful review of decorators, I'll add that to the MP, yes? next question: [12:26] os_helper.py [12:26] nessita: completely agree [12:26] _get_syncdaemon_valid_path, I would expect this class, that assert over the param to be a unicode, to return bytes [12:26] nessita: let do this, rule of thum, Watch and Watch manager do nothing with paths, FileSystemNotifications take care of it [12:27] mandel: makes sense, can we add tests? I see that in this branch we removed some decorator and tests did not change [12:28] nessita: yes, we should, this is probably the most important branch we have, if we get it right lots of bugs will be fix, lets take our time and add as many tests as possible if needed [12:29] nessita: I'm 110% sure all the encoding and \\?\ things are the main root of issues in the port [12:29] visteme despacio... [12:29] right [12:30] mandel: so, I'm starting again with the next question to avoid confusions: in os_helper.py: _get_syncdaemon_valid_path, I would expect this class, that assert over the param to be a unicode, to return bytes [12:31] can i install u1sdtool on a server? [12:32] mandel: ex-remove_utf8_chars only replaces unicode chars, but does not transform to bytes [12:33] mandel: that method (_get_syncdaemon_valid_path) should do the .encode('utf8') [12:34] nessita: it is asserting that it gets unicode, it is returning unicode, I just did a grep and this is only used in the filesystem notifications to get the path of the even (not to be confused with the pathname) which is never used, yet it should be using bytes, issue here is that wath should be using unicode in self._path, so we have to remove the \\?\ and the unicode chars and be able to perform the correct operation in line 212 in filesystem_notificati [12:34] nessita: sorry for the terrible spelling, I was looking at the vm while I typed... [12:35] mandel: the thing is the name of the method. A method with that name, *should* return bytes since is meant to be used by syncdaemon [12:35] nessita: I'd remove that method and use reinstantiate_illegla_chars instead [12:35] mandel: restore_illegal_chars? :-) [12:35] eso :P [12:35] ok [12:36] mandel: also, what do you think about changing _validate_windows_path to _assert_path_is_bytes? [12:37] nessita: yes, sounds good, is more direct [12:38] mandel: and then you can use that in the implementation of _valiate_output_path, reaplacing the call to [12:38] assert isinstance(path, str), 'All paths returned to sd should be bytes' [12:38] mandel: yes? [12:38] morning! [12:38] nessita: yes [12:39] ralsina: hey there! you ok? :-) [12:39] mandel: and I think that's all I have :-) [12:39] nessita: great review :) [12:39] mandel: :-) [12:40] nessita: had another bout of stomach ache last night but much better than the night before [12:40] mandel: I'll add all this to the MP [12:40] nessita: probably will not have another one tonight :-) [12:40] nessita: if we have everything in comments I'll start working on it right now, I really think this is coming very nice, even if we need some many cycles to land it [12:41] ralsina: there is an age where you can only dinner raisins… [12:41] * mandel hides [12:42] mandel: and I hate raisins :-) [12:42] * fagan slaps mandel [12:43] mandel: all comments added [12:43] mandel: let me know as soon as those are fixed [12:44] mandel: and I will start charging you and ralsina a dollar per each needs fixing :-P (j/k) [12:44] nessita, mandel, alecu: possible hotel for the sprint http://www.hoteljacaranda.com.ar [12:45] ralsina: does that work for you? [12:45] nessita: 15 minutes by car [12:45] nessita: it's quite far fr alecu though [12:45] ralsina: have a googlemap references? [12:46] http://is.gd/Pct7ly [12:48] nessita: it should be calculate by the size of the branch.. otherwise is not fair! [12:48] ralsina: a taxi from AEP would be like $100? [12:48] mandel: lol [12:48] nessita: more or less, yes [12:49] ralsina: do taxis go there or shall I book a remis? [12:49] ralsina: I know that taxis from capital charges extra for going into the provincia [12:49] taxis usually go there, but since it's outside CABA they may charge you a bit extra (say, $10) [12:50] or they may not, usually you arrange that before starting the trip [12:50] 10 mins to standup [12:50] ralsina: ok [12:50] mandel OTOH is going to have close to the longest taxi ride possible in buenos aires :-) [12:51] mandel: did you add your travel details to the wiki page? [12:51] ralsina: so no night out… :( [12:51] nessita: no, will do after the stan up [12:51] mandel: nah, it's les than 1 hour http://is.gd/0PbzjH [12:52] ralsina: that in spain is like goign to a diff city... [12:52] mandel: at least on a sunday it's less than 1 hour [12:52] mandel: here too! It's just that the cities are next to each other ;-) [12:52] ralsina: I moved the sprint page to https://wiki.canonical.com/OnlineServices/Sprints/SprintJuly2011 [12:52] ralsina: I'll let marianna know [12:53] ralsina: well, I guess is better for you guys that I'm not too close :) [12:54] mandel: you will be 15 minutes from my home, we'll show you around if you want [12:54] ralsina: good :) [12:55] mandel: and check the satellite view, the hotel is 2 blocks from the private piers. In the "national capital of sailing" [13:00] moi [13:00] me [13:00] me [13:00] ralsina dobey [13:00] me [13:00] me [13:01] ill go [13:01] DONE [13:01] * tested alecu's script [13:01] * set up a usability phone call for tomorrow with matthew revell about lp [13:01] * finished the 3 months of backlog [13:01] TODO [13:01] * More unassigned bugs (if nothing else is needed) [13:01] * reply to a lecturer about something with college should take a bit [13:01] BLOCKED [13:01] * Nope [13:01] mandel: go [13:01] DONE: Rebased all branches in the work done by nessita. One of the branch has a small inconsistent test failing. Talk with nessita about improving fix-os-helper. [13:01] TODO: Perform changes in fix-os-helper according to review. Add travel details to wiki. Talk with verterok about local_rescan failing test. [13:01] BLOCKED: no [13:01] nessita, go go go [13:01] DONE: bug #810662, tons of reviews, meetings, chasing people [13:01] TODO: more of the same! :-) [13:01] BLOCKED: nopes [13:01] NEXT: dobey [13:01] Launchpad bug 810662 in ubuntuone-control-panel "QT UI: "Delete" button in the devices tab does not work (affects: 1) (heat: 15)" [High,Triaged] https://launchpad.net/bugs/810662 [13:01] λ DONE: reorg plan, converted protocol and cp to dh_python2 in ubuntu [13:01] λ TODO: reorg work, shim code [13:01] λ BLCK: None. [13:01] ralsina [13:01] DONE: windows call, mgmt call, reviews, closed bug #810521, worked on bug #811111, published an installer, worked on bug #800376 [13:01] TODO: work on #800376, close #811111 and work a lot of minor ones [13:01] Launchpad bug 810521 in ubuntuone-client "Typo in signal name (affects: 1) (heat: 6)" [High,Fix committed] https://launchpad.net/bugs/810521 [13:01] Launchpad bug 811111 in ubuntuone-windows-installer (and 1 other project) "There is no visual indication that the form is being processed (affects: 1) (heat: 6)" [High,In progress] https://launchpad.net/bugs/811111 [13:02] Launchpad bug 800376 in ubuntuone-windows-installer "Add "syncing your computer to the cloud" page to the installer (affects: 1) (heat: 8)" [High,In progress] https://launchpad.net/bugs/800376 [13:02] and I am the last one. Comments? [13:02] did alecu say me? [13:02] Nope my call is around the same time as standup tomorrow but im a good multitasker [13:02] nessita: no [13:02] alecu isnt online [13:03] ah, ok [13:03] no comments here [13:03] * mandel lunch === m_conley_away is now known as m_conley [13:04] alecu: hey, tested that script I can give you the output if you want [13:04] it worked I think it gave out a big long id thingy [13:04] fagan, please do [13:05] alecu: kk [13:05] 5 mins ill have it in a pastebin [13:05] alecu: standup :) [13:06] me [13:06] alecu: it is your turn, we already went :P [13:06] * alecu is writing notes [13:06] http://pastebin.ubuntu.com/648224 [13:07] alecu: ^ [13:07] Thats for my main account do you need another account too [13:07] I tested it and it was something pretty similar [13:07] DONE: started working on #803669, many reviews [13:07] TODO: finish branch, work on #803672 [13:07] BLOCKED: no [13:10] alecu: ok here is the second http://paste.ubuntu.com/648228 [13:14] bug #803669, bug #803672 [13:14] Launchpad bug 803669 in ubuntuone-client (and 1 other project) "Get the port numbers from the users' registry (affects: 1) (heat: 6)" [Low,Triaged] https://launchpad.net/bugs/803669 [13:14] Launchpad bug 803672 in ubuntuone-client (Ubuntu) (and 2 other projects) "Do not start two instances of this process (affects: 1) (heat: 6)" [Undecided,Invalid] https://launchpad.net/bugs/803672 === teknico_ is now known as teknico [13:23] * mandel back [13:23] nessita: I'm starting with the changes [13:24] mandel: great! [13:27] alecu, nessita, mandel, dobey, vds, CardinalFang, thisfred: I just sent you your performance reviews. Sorry about the loooooooong delay, it was my own fault. [13:27] thx, just saw it! [13:27] ralsina: ack [13:28] mail got misfiled without me seing it, and I first expected it to appear, then forgot about it, then assumed it was never sent [13:28] so who sai in my peer review that I'm extremely attractive? [13:28] :P [13:28] mandel: fagan after drinking tequilas? :-) [13:29] mandel: you attract jellyfish! [13:29] can someone confirm a bug for me? If you add a file to a synced folder, then immediately choose to publish the file, then afterwards, it's no longer possible to either publish or unpublish the file from nautilus. That means you have to move the file somewhere else and then back in order to do things with it. [13:29] hehehe [13:29] I think that could be fixed by not providing the menu option to publish a file until it's been synced. [13:29] * mandel goes are reads the report... [13:32] ralsina: update my fligth details in the wiki [13:32] mandel: is that an order? ;-) [13:32] ralsina: does marianna know I'm staying til the 2nd [13:33] ralsina: was not an order, it was a missing ed :P [13:33] mandel: if it's in the wiki, she knows. She may not knw canonical is paying for it though, but I'll tell her that. If she asks you, tell her it's approved by me and Chipaca [13:33] at least that's how I understood Chipaca in the mumble call :-) [13:34] ralsina: ok, if not I can pay, unless is too expensive [13:34] mandel: I have a couch [13:35] mandel: And I have a key for a nice furnished appartment downtown as well ;-) [13:35] ralsina: now, that sounds like a plan… [13:36] fagan, thanks for the pastebins, was just what I expected. [13:37] nessita: hi [13:37] nice alecu [13:38] lisette: hi there! [13:38] nessita: got your mail; working on it :) [13:38] lisette: thanks! [13:38] lisette: alecu added a comment to the bug report, just FYI [13:39] i saw [13:39] lisette: great! [13:40] mandel, alecu, nessita, chipaca: looks like I broke my headset. Can we do the mumble a little later today? [13:40] ralsina: sure by me, what time? [13:41] ralsina: yeah, i'll be on late today [13:41] I would say after lunch so I can go buy a new ne [13:41] * mandel read that as 'I broke my head...' [13:41] maybe 13:30 art? [13:41] mandel, is that too late for you? [13:41] ralsina: is never too late :) [13:42] ralsina: weren't you maybe traveling for banking issues today? [13:43] was a maybe, but I am not :-) [13:43] nessita: found other solutions [13:43] ralsina: oh good! [13:43] ralsina: after lunch sounds ok, but maybe is too late for Chipaca? [13:45] nessita: give me a time [13:45] Chipaca: 2pm ART, 5pm UTC [13:45] Chipaca: 13:30 ART? [13:45] ralsina, alecu^ [13:45] ok, 2pm [13:45] I can do 6pm, yes :) [13:45] great! is a date! (?) [13:47] ack [13:47] Chipaca: we'll keep it short, promise! [13:47] nessita: so, when do we meet? [13:48] mandel: 5pm UTC [13:48] mandel: 7pm your time [13:48] nessita: ok [13:55] good day [13:56] nessita: regarding the os_helper branch, I'm going to add the validate_input_path in the WatchManager and Watch methods to ensure that those just get unicide and not bytes, do you agree on that? [13:56] nessita: that way we force FileSystemNotifications to always pass the paths correctly to the more platform dependent code [13:57] mandel: yes, but you mean the assert_path_is_bytes, right? [13:57] mandel: since we rename it (in our heads) [13:58] nessita: yes.. just did the change, and yes _assert_path_is_bytes is the one [13:58] mandel: yes to the first question then [13:58] nessita: but that is the method name not the decorators, should I rename the decorator too? [13:58] mandel: yeah, I think it makes sense [13:59] nessita: ok, search and replace on the way! [14:01] what happened to Ubuntu One? i can't login. this what i get to when i try to log into Ubuntu One account. http://unknown.lt/screen.png [14:02] vendix: are you on 11.04 or 11.10? [14:02] donno [14:02] dunno [14:02] how to see that? [14:03] vendix: if you are on 11.04 with gnome3 ppa installed, the gnome3 stuff broke various things [14:03] vendix: /etc/lsb-release i think [14:04] nessita: other change validate_output_path assert_output_path_is_unicode [14:05] dobey: give me different command [14:05] mandel: but... validate_output_path returns bytes, no? [14:05] to see the version of ubuntu [14:05] vendix: that isn't a command, it is a file that contains the ubuntu version information [14:06] nessita: sorry yes, s/unicode/bytes [14:06] mandel: thing is that validate_output_* also check the long prefix not there [14:07] mandel: that's why I did not suggested the renaming [14:07] mandel: I will leave the function as is, fixing the typo in def _valiate_output_path(path): [14:07] 11.04 [14:07] how to upgrade to 11.10? [14:08] nessita: lets split it in two, so for when ever we need to use unicode we simply remove that decorator and not the check of the long path prefix [14:08] checking for \\?\ will always be needed [14:08] mandel: ok, sounds good (I hope the branch does not grow that much) [14:09] nessita: yeah, seems to be an accordeon branch [14:14] vendix: if you *really* want to upgrade to the development release, you can run "update-manager -d" and it should offer you the option to upgrade [14:56] I'm having a horrible time concentrating on writing code today :-( [14:56] anyone needs any reviews? [14:57] ralsina, I can use a review here: https://code.launchpad.net/~alecu/ubuntuone-client/unbreak-connect/+merge/68309 [14:57] nessita, you might want to re-review the above branch, too. [14:57] alecu: sure! [14:57] thanks! [14:58] alecu: this needs any sort of IRL testing? [14:58] alecu: quick tweak, can you please add a space after the : in lambda _:reply_handler()? [14:58] ralsina, just try stopping and starting again the syncdaemon from the control panel. [14:58] nessita, sure [14:59] alecu: code looks great! running tests as soon as you push that [15:00] alecu: ack [15:01] nessita, pushed [15:02] alecu: yey [15:04] nessita: replied to your mail about what to do with the installer UI and ussoc. Basically, I'm ok with that, debugging inside ussoc is much harder, and things are just weird :-( [15:06] ralsina: ack, I'll start working on that after alecu's gives his +1, if he agrees, of course :-p [15:07] ok, I need to go to the meat shop otherwise my half orange will starve\ [15:07] * nessita -> brbs [15:11] * mandel small break to get some air in to the brain [15:17] has anybody pinged me while I was half-away [15:17] ? [15:19] * ralsina ---> cook === ralsina is now known as ralsina_cooking [15:26] alecu: all green, approving [15:27] nessita, great. I didn't understand your comment about "alecu's gives his +1" [15:27] nessita, is this related to splitting the sso? [15:27] nessita, I mean, not running the installer on the sso process? [15:27] nessita, I'm +1 to not running the installer on the sso process. [15:27] alecu: I answered to ralsina_cooking's report, last night, proposing moving the installer process outside sso [15:27] alecu: what about the idea I proposed? does it make sense? [15:29] nessita, I read it, but I have some doubts: [15:29] alecu: I'm here to answer [15:29] alecu: I'm here to answer "i don't know" :-P [15:29] nessita, what about the "registration": will we be duplicating all the registration steps on the installer? [15:30] nessita, the captcha and validation code, and all that? [15:30] nessita, or will the installer use the _ui.py modules from sso? [15:30]