[09:56] <kraut> moin
[02:35] <zul> @schedule montreal
[02:35] <ubotu> Schedule for America/Montreal: 14 Aug 11:00: Ubuntu Server Team meeting | 14 Aug 15:00: Technical Board | 15 Aug 16:00: Edubuntu | 16 Aug 16:00: Ubuntu Development Team | 20 Aug 18:00: Forum Council meeting | 21 Aug 11:00: Kernel Team
[02:36] <christian> join linux
[03:03] <dendrobates> @schedule New York
[03:03] <ubotu> Schedule for America/New_York: 14 Aug 11:00: Ubuntu Server Team meeting | 14 Aug 15:00: Technical Board | 15 Aug 16:00: Edubuntu | 16 Aug 16:00: Ubuntu Development Team | 20 Aug 18:00: Forum Council meeting | 21 Aug 11:00: Kernel Team
[03:06] <stgraber> @schedule Zurich
[03:06] <ubotu> Schedule for Europe/Zurich: 14 Aug 17:00: Ubuntu Server Team meeting | 14 Aug 21:00: Technical Board | 15 Aug 22:00: Edubuntu | 16 Aug 22:00: Ubuntu Development Team | 21 Aug 00:00: Forum Council meeting | 21 Aug 17:00: Kernel Team
[03:17] <malleze> ?
[04:19] <kraut> @schedule berlin
[04:19] <ubotu> Schedule for Europe/Berlin: 14 Aug 17:00: Ubuntu Server Team meeting | 14 Aug 21:00: Technical Board | 15 Aug 22:00: Edubuntu | 16 Aug 22:00: Ubuntu Development Team | 21 Aug 00:00: Forum Council meeting | 21 Aug 17:00: Kernel Team
[04:19] <kraut> are there any predefinied topics about the server-meeting?
[04:19] <kraut> any agenda?
[04:22] <kraut> ah, found it
[04:27] <nealmcb> https://wiki.ubuntu.com/ServerTeam/Meeting
[04:28] <nealmcb> https://wiki.ubuntu.com/ServerTeam/Roadmap
[04:37] <Nafallo> nealmcb: thanks
[04:38] <Nafallo> nice! my suggestion is already on there :-)
[04:58] <dendrobates> #startmeeting
[04:58] <MootBot> Meeting started at 14:51. The chair is dendrobates.
[04:58] <MootBot> Commands Available: [TOPIC] , [IDEA] , [ACTION] , [AGREED] , [LINK] , [VOTE] 
[04:58] <dendrobates> Just making sure mootbot is working.
[04:58] <keescook> MootBot's clock is off...
[04:59] <jono> yo
[05:00] <jono> where is the agenda?
[05:00] <mathiaz> jono: https://wiki.ubuntu.com/ServerTeam/Meeting
[05:00] <dendrobates> and away we go.
[05:00] <nealmcb> current time is 15:00:39
[05:00] <nealmcb> UTC
[05:00] <jono> ok I would like to add an item to discuss the community team
[05:01] <dendrobates> jono would you like to start, so we make sure we fit ouy in?
[05:01] <jono> dendrobates: happy to if no one minds
[05:01] <dendrobates> jono: the floor is yours
[05:02] <jono> great stuff
[05:02] <jono> I just really wanted to get an update on how the community building is going and discuss the next steps
[05:02] <jono> from what I can tell, things are going great
[05:02] <jono> what is the progress with the team so far?
[05:02] <dendrobates> We are pretty happy, with the response.
[05:03] <jono> excellent
[05:03] <jono> and these meetings are regular?
[05:03] <dendrobates> We need to work on ways to give community members very concrete tasks to workon.
[05:03] <jono> right
[05:03] <jono> can I also suggest to have a docs day?
[05:04] <tarvid> good point, i made two promises and only partially kept one
[05:04] <dendrobates> sometimes it is just as easy to do something yourself as to define it well enough for someone else to work on it,
[05:04] <jono> the aim of the docs day is to fill out https://wiki.ubuntu.com/ServerTeam/KnowledgeBase with details of how to get involved in the team
[05:04] <jono> I would sugges that some focus goes into filling out the knowledge base - this will help drive in new contributors
[05:04] <tarvid> er make that 3
[05:05] <mathiaz> another thing that may be interesting is a mentoring program
[05:05] <nealmcb> [LINK]   https://wiki.ubuntu.com/ServerTeam/KnowledgeBase
[05:05] <MootBot> LINK received:   https://wiki.ubuntu.com/ServerTeam/KnowledgeBase
[05:05] <mathiaz> I've seen that on the ubuntu-doc mailing list.
[05:05] <jono> I would recommend that before you do the mentoring, get the docs in place
[05:05] <mathiaz> they've sent an anoucement about docuMentors
[05:05] <mathiaz> and people are signing up.
[05:05] <jono> think of the knowledge base as the content that new contributors will use to get up to speed - if they have no content to help them, they will get bored or lost and move on
[05:06] <jono> so I would recommend the team invests in creating this documentation first before mentoring and further outreach happens
[05:06] <dendrobates> [ACTION]   Fill out https://wiki.ubuntu.com/ServerTeam/KnowledgeBase with details of how to get involved in the team
[05:06] <MootBot> ACTION received:   Fill out https://wiki.ubuntu.com/ServerTeam/KnowledgeBase with details of how to get involved in the team
[05:06] <jono> :)
[05:06] <jono> I recommend you folks pick a day for your docs day and really spend much of that day doing docs
[05:07] <jono> and encourage the wider community to write docs on that day
[05:07] <jono> you will want to make a big deal of the docs day to encourage everyone to get involved and write docs
[05:07] <mathiaz> what about every other wednesday ?
[05:07] <nealmcb> (see more about mootbot commands you can use to structure the meeting minutes at https://wiki.ubuntu.com/ScribesTeam/MootBot)
[05:07] <mathiaz> alternating with the ubuntu-server meeting ?
[05:08] <jono> mathiaz: regular docs days tend to die a death - I would recommend you do just one for now
[05:08] <mathiaz> ok. next friday ?
[05:08] <jono> really focus this day on getting the docs sorted
[05:08] <jono> a week on friday?
[05:08] <mathiaz> so that we can announce it on the mailing list ?
[05:09] <dendrobates> I think the sever community is unique in that many members have Jobs at large corps, that limit the days/hours they can contribute.  I think we might need to stagger the times of things like doc days and meetings to get the most involvement.
[05:09] <jono> fri 24th?
[05:09] <jono> dendrobates: agreed
[05:09] <nealmcb> works for me - sounds fun
[05:09] <jono> and much of the docs day work might happen in the evening
[05:09] <jono> fri 24th sounds great
[05:10] <jono> but you want to make everyone in the ubuntu community know about the docs day - get it on the fridge, everyone blog it, do a few podcast interviews, shout out the window - whatever :)
[05:10] <dendrobates> [ACTION]  Organize Doc Day for July 24th.
[05:10] <MootBot> ACTION received:  Organize Doc Day for July 24th.
[05:10] <dendrobates> oops.
[05:10] <nealmcb> :-)
[05:10] <mathiaz> dendrobates: hummm.. that's a bit far awya
[05:10] <jono> MootBot is way cool :)
[05:11] <dendrobates> [ACTION]  Organize Doc Day for Aug 24th.
[05:11] <MootBot> ACTION received:  Organize Doc Day for Aug 24th.
[05:11] <dendrobates> I have no idea how to remove an action.
[05:11] <sommer> just so I'm clear this is doc day for creating docs on getting involved in the server team?  Not server howto type docs?
[05:11] <jono> sommer: yep, its all about how to contribute to the team
[05:12] <sommer> cool
[05:12] <jono> sommer: the plan is to make it as easy as possible for a new contributor to get involved
[05:12] <mathiaz> yes. It's about updating the knwoledge base.
[05:12] <jono> stuff like:
[05:12] <jono>  * how to join the team
[05:12] <jono>  * where resources
[05:12] <jono> are
[05:12] <jono>  * technical processes
[05:12] <keescook> not that we should discourage people from writing howto docs, though!
[05:12] <jono> etc
[05:12] <penguim> hi all
[05:12] <jono> keescook: :)
[05:12] <jono> hey penguim
[05:12] <jono> ok, lets talk about the TODO
[05:12] <jono> is there a TODO that indicates what needs to be done?
[05:13] <mathiaz> yes. The Roadmap: https://wiki.ubuntu.com/ServerTeam/Roadmap
[05:14] <jono> cool stuff
[05:15] <mathiaz> should we go through the roadmap ?
[05:15] <dendrobates> not yet.
[05:15] <mathiaz> dendrobates: or you wanna discuss the other item first ?
[05:16] <dendrobates> [TOPIC]   Review ACTION points from previous meeting.
[05:16] <MootBot> New Topic:   Review ACTION points from previous meeting.
[05:16] <tarvid> i have a short howto on getting the drupal5 package to work, where should I put it
[05:16] <dendrobates> Lets go through this quickly.
[05:16] <jono> right I am gonna look away and do other stuff, ping me if I am needed :)
[05:16] <dendrobates> thanks jono.
[05:16] <jono> np :)
[05:17] <nealmcb> see also https://help.ubuntu.com/community/Drupal
[05:17] <mathiaz> previous meeting log: https://wiki.ubuntu.com/MeetingLogs/Server/20070731
[05:17] <dendrobates> I know we are all busy getting ready for FF.  So has anyone completed any action items?
[05:17] <mathiaz> I haven't sent the annoucement for UWN.
[05:18] <dendrobates> DONE. dendrobates and jdstrand should discuss auth-client-config packaging issues.
[05:18] <mathiaz> I still need to contact tarvid for AppArmor testing.
[05:18] <mathiaz> tarvid: can you still provide some help for that ?
[05:18] <dendrobates> That was about python-central, and jstrand fixed it.
[05:18] <tarvid> i will be at it shortly, just put up the sandbox at home
[05:18] <tarvid> did the drupal5 package first
[05:19] <dendrobates> [TOPIC]   Discuss example PHP home page for LAMP install.
[05:19] <MootBot> New Topic:   Discuss example PHP home page for LAMP install.
[05:19] <mathiaz> ok. That's all for me for my last actions.
[05:19] <jdstrand> dendrobates: have you thought more about our discussion regarding your packages and 'purge'?
[05:20] <tarvid> i am especially interested in seeing if AppArmor can be used to make shell commands safe on a shared server
[05:20] <dendrobates> jdstrand: I have not had time.
[05:20] <sommer>  mathiaz: I'm also interested in helping test AppArmor.
[05:20] <jdstrand> dendrobates: ok-- we can talk more on #ubuntu-server later
[05:20] <mathiaz> sommer: cool. I'll get in touch with you also.
[05:21] <sommer> mathiaz: thanks
[05:21] <dendrobates> It has been stated as a goal that we need a php page example for the lamp install.
[05:21] <mathiaz> The idea about the php install page is to have the equivalent of the 'It works' page
[05:21] <tarvid> I always wind up adding a phpinfo.php page
[05:22] <mathiaz> we also want to make sure that the mysql database is running correctly.
[05:22] <keescook> how about a displayed checklist of "OK" items?
[05:22] <keescook> query the DB, "db OK", etc
[05:23] <keescook> (the PHP test page can use the "test" db)
[05:23] <sommer> can't you install PHP without MySQL though.
[05:23] <mathiaz> yes you can.
[05:23] <nealmcb> and many want to use postgresql
[05:23] <mathiaz> but we'd like to test the LAMP install.
[05:23] <sommer> ah...makes sense.
[05:24] <mathiaz> people can install a LAMP stack, and we'd like to have something to show that things are working correctly.
[05:24] <tarvid> phpinfo has been my best guide
[05:24] <jdstrand> I like keescook idea-- shows the different components of LAMP are working
[05:24] <mathiaz> Should we test that select is working from the database ?
[05:24] <mathiaz> tarvid: well... It doesn't test if the mysql database is working
[05:25] <tarvid> that is correct but at least it shows you which versions are installed
[05:25] <jdstrand> tarvid: not to mention, if the user doesn't update it, it gives out a *ton* of info
[05:25] <keescook> checking for postgresql ... skipped        checking for mysql ... OK     database found!   etc
[05:25] <mathiaz> so we could have test page that shows that apache is running..
[05:25] <mathiaz> mysql is running
[05:25] <mathiaz> may be postgresql
[05:26] <dendrobates> network glitch.  Sorry.
[05:26] <mathiaz> and which version of php is running.
[05:26] <mathiaz> whith a list of php modules loaded ?
[05:26] <tarvid> for example, a drupal install installs apache2-mpm-itk
[05:26] <nealmcb> Do we really want a test page that knows a db password?
[05:26] <tarvid> and postfix
[05:26] <sommer> Is there a "test" db setup with the install?
[05:27] <tarvid> on a shared server, mysql passwords can be read
[05:27] <jdstrand> mathiaz: personally, I am with you all the way to the modules.  I just keep thinking of people not updating index.php (or whatever) for a while, and there being to omuch info given
[05:27] <tarvid> from configuration files
[05:27] <mathiaz> mysql_install_db installs some tests database.
[05:28] <sommer> I'd think that'd work then, the username/password is for test and you could include docs on how to remove after install.
[05:28] <nealmcb> but having a remotly-accessible way to log in to even a test database seems a bit risky
[05:28] <tarvid> we are headed towards regression testing and that is not a bad idea
[05:28] <dendrobates> it should not be remotely accessable.
[05:29] <dendrobates> you can grant local rights only.
[05:29] <dantalizing> what will you do if someone installs wordpress....lamp is a dependency, and wp tries to install a index.php too?
[05:29] <mathiaz> we'll probably face the same problem as the default apache website.
[05:29] <dendrobates> this would only be for the tasksel lamp task.
[05:30] <dendrobates> it should clobber nothing.
[05:30] <mathiaz> so what is needed is to write a simple php web page
[05:30] <mathiaz> that prints some information.
[05:30] <dantalizing> but wouldnt wp fail with "trying to overwrite file...in package lamp"?
[05:31] <dendrobates> It would have to be clobberable as well, just like ant index.html that ships.
[05:31] <tarvid> i think we should distinguish between a class of users who wants to get "something" running and a more experienced class that knows what they want
[05:31] <mathiaz> and then we need to add it to the tasksel task.
[05:32] <dendrobates> yes.
[05:32] <mathiaz> advanced users would reconfigure apache anyway
[05:32] <mathiaz> and remove the default web host.
[05:32] <mathiaz> defautl website.
[05:33] <dendrobates> are there any php programmers out there who would like to give it a try?
[05:33] <jdstrand> I can do it
[05:34] <dendrobates> [ACTION]  jdstrand will take a first stab at an example php page for the LAMP tasksel.
[05:34] <MootBot> ACTION received:  jdstrand will take a first stab at an example php page for the LAMP tasksel.
[05:35] <dendrobates> [TOPIC]   Brainstorm new server-related  ubotu factoids, e.g. webmin, mail, mua, mta, pop, imap, nic, openssl, inetd.
[05:35] <MootBot> New Topic:   Brainstorm new server-related  ubotu factoids, e.g. webmin, mail, mua, mta, pop, imap, nic, openssl, inetd.
[05:35] <nealmcb> [IDEA]  people on #ubuntu-server should watch for FAQs, see if ubotu has related factoids and if not, propose new factoid text.
[05:35] <MootBot> IDEA received:  people on #ubuntu-server should watch for FAQs, see if ubotu has related factoids and if not, propose new factoid text.
[05:35] <nealmcb> the ones I listed in the agenda don't exist for the most part
[05:35] <mathiaz> the first item that we should add is about webmin.
[05:35] <mathiaz> it's been asked a couple of times on the irc channel.
[05:36] <dantalizing> has anyone used ebox? i thought it was being added and is supposedly easier??
[05:36] <nealmcb> who knows the history, and can provide examples of config file mutilation?
[05:36] <dantalizing> i'll shut up now
[05:36] <infinity> soren's been working on cleaning up eBox, hasn't he?
[05:36] <tarvid> tasksel is in perl
[05:36] <dendrobates> I know nothing about ubotu factioids.  Care to give a quick primer?
[05:37] <nealmcb> https://wiki.ubuntu.com/UbuntuBots
[05:37] <nealmcb> I
[05:37] <mathiaz> infinity: why webmin was dropped from the archive ?
[05:37] <dendrobates> OK so what actions do we have out of this topic?
[05:37] <nealmcb> That describes ways to add new factoids, but I think only certain folks like seveas can do so now
[05:38] <infinity> mathiaz: Because it doesn't play nicely with dpkg in any way, shape, or form.  It takes over your config files, mangles them, and leaves the packaging system confused and upset.
[05:38] <infinity> mathiaz: And webmin users become completely unsupportable by us.
[05:38] <Mithrandir> mathiaz: "RoM; outdated; unmaintained"
[05:39] <nealmcb> infinity: can you put some text together and maybe give an example, contrast with ebox et al?
[05:40] <dendrobates> Before we move on do we have any action items for this topic?
[05:40] <mathiaz> I'll write some text on webmin
[05:40] <nealmcb> thanks!
[05:40] <mathiaz> send it to the ubuntu-server mailing list
[05:40] <mathiaz> and figure out how to update the bot.
[05:41] <dendrobates> [ACTION]  mathiaz will write a factiod about webmin.
[05:41] <MootBot> ACTION received:  mathiaz will write a factiod about webmin.
[05:41] <dendrobates> [TOPIC]   Update server related pages on help.ubuntu.com - Documentors section.
[05:41] <MootBot> New Topic:   Update server related pages on help.ubuntu.com - Documentors section.
[05:41] <nealmcb> s/iod/oid/
[05:41] <infinity> Uhm, okay, I just read the channel log.
[05:42] <infinity> What package are you proposing ships this "PHP Test Page"?
[05:42] <infinity> And, dear god, why?
[05:42] <dendrobates> This is a server team task from mdz.
[05:42] <mathiaz> There has been some request on the irc channel about help pages.
[05:43] <infinity> Seriously?
[05:43] <dendrobates> the php test page I mean.
[05:43] <mdz> infinity: as an example, to show that it works
[05:43] <mathiaz> and people were complaining about the lack of accuracy about the wiki pages
[05:43] <mathiaz> on help.ubuntu.com
[05:43] <infinity> Well, I could write it in about 5 minutes.
[05:44] <mathiaz> so what would be a list of pages on help.ubuntu.com that we should try to update ?
[05:44] <mdz> infinity: currently, you install the LAMP task, and the system should be able to run PHP pages, but there aren't any on the system to try.  I'm suggesting we put one there.
[05:44] <dendrobates> There was nothing in the task that stated it had to be the default webpage, though.
[05:44] <mdz> infinity: make sense?
[05:44] <infinity> mdz: Yeah, makes a modicum of sense, just not entirely sure where to put it, so it's both visible, but not in the way.
[05:45] <mdz> infinity: same place as the default HTML one, I'd think
[05:45] <infinity> (And how to ship it... A .deb in the LAMP task to ship one file seems excessive)
[05:45] <mdz> infinity: or do you mean package-wise?
[05:46] <infinity> mdz: /var/www/lamp-test/index.php would seem an alright place to shove it, or something similar, but yeah, which package to shove it in is a fun one.  PHP might make the most sense.
[05:46] <mdz> (I'm on a conference call, but hopefully I've clarified what I was looking for)
[05:46] <infinity> mdz: I could whip up a generic test page that shows if MySQL or PostsreSQL is up and running on localhost, status of some other random stuff, whatever.
[05:46] <jdstrand> infinity, mdz: there was discussion that this should test for mysql too
[05:47] <mathiaz> infinity: if we want to test for mysql, it may not be the best palce.
[05:47] <tarvid> we are talking about the webmin deb and not the raw package aren;t we?
[05:47] <infinity> mathiaz: Nothing wrong with it testing for a database and not finding one.
[05:47] <jdstrand> infinity: I had said that I would do it, but you are welcome to if you want it
[05:47] <dendrobates> ok lets get back on topic.   Update server related pages on help.ubuntu.com - Documentors section.
[05:48] <mathiaz> I think we should list the wiki pages that are problematic according to the complains in the IRC channel.
[05:49] <mathiaz> and then we can point people wanting to work on server related documentation to it.
[05:49] <dendrobates> OK, is there someone who would like to do this?
[05:50] <sommer> I can help...not sure which ones are being complained about though.
[05:50] <dendrobates> mathiaz: have you heard specific complaints?
[05:50] <mathiaz> hum... dovecot+postfix
[05:51] <mathiaz> yesterday I think.
[05:51] <ScottK> Yes, yesterday
[05:51] <tarvid> i have a brief drupal5 article for the wiki, i would like to get it into the proper format and then find out what people think
[05:51] <mathiaz> tarvid: did you check out https://help.ubuntu.com/community/Drupal ?
[05:51] <tarvid> i also have an esmtp howto
[05:52] <dendrobates> So this would be an ongoing task, but should not be too much work.
[05:52] <mathiaz> dendrobates: correct.
[05:52] <tarvid> interesting, i search for drupal and this page was not returned
[05:52] <dendrobates> sommer: are you willing to do this on an ongoing basis?
[05:53] <sommer> dendrobates: sure.
[05:53] <nealmcb> tarvid: did you search via google? via the wiki?
[05:53] <sommer> I'm currently working on migrating from sendmail+dovecot to postfix so very topical for me.
[05:53] <tarvid> via the wiki
[05:53] <mathiaz> tarvid: wiki.ubuntu.con or help.ubuntu.com ?
[05:54] <tarvid> i found it this time, must have been a typo
[05:54] <tarvid> probably help.ubuntu.com
[05:54] <tarvid> i can work on an update
[05:54] <dendrobates> so is this the correct action:  sommer will listen for documentation complaints on #ubuntu-server and update update them
[05:54] <mathiaz> tarvid: that would be excellent.
[05:55] <dendrobates> -1 update
[05:55] <sommer> I can also work on the dovecot+postfix doc.
[05:55] <mathiaz> sommer: is there a document on help.ubuntu.com ?
[05:56] <mathiaz> there are a couple of Mail related ressource on help.ubuntu.com
[05:56] <dendrobates> mathiaz: do you want him to update the server page so documentors can fix, or do you want him to fix the documentation?
[05:56] <dendrobates> I'm thinking the former.
[05:56] <mathiaz> dendrobates, sommer: update the server page.
[05:57] <nealmcb> which server page?
[05:57] <mathiaz> dendrobates, sommer: not fixing them. So that we have a list of things to work on.
[05:57] <mathiaz> https://wiki.ubuntu.com/ServerTeam/Roadmap
[05:57] <mathiaz> in the documentor section.
[05:57] <nealmcb> ahh - thanks
[05:58] <dendrobates> [ACTION]  sommer will listen for documentation complaints on #ubuntu-server and update the documentor section of  https://wiki.ubuntu.com/ServerTeam/Roadmap.
[05:58] <MootBot> ACTION received:  sommer will listen for documentation complaints on #ubuntu-server and update the documentor section of  https://wiki.ubuntu.com/ServerTeam/Roadmap.
[05:58] <mathiaz> I'll add a task to the Roadmap so that sommer can list them there.
[05:58] <sommer> mathiaz: so I should submit drafts linked to that section?
[05:58] <dendrobates> We are running out of time.  Do we need to schedule these more often?
[05:59] <mathiaz> hum.. we're almost done.
[05:59] <sommer> My thought is it may not be a bad idea.
[06:00] <sommer> at least until the Server Team contributing docs are ironed out.
[06:00] <dendrobates> [TOPIC]   Discuss new tasksel for openssh server
[06:00] <MootBot> New Topic:   Discuss new tasksel for openssh server
[06:01] <dendrobates> What does everyone think of adding a tasksel for openssh-server?
[06:01] <tarvid> openssh-client is installed by default, it would be better to install openssh-server and not install client
[06:02] <lamont> server by default wouldn't be allowed to listen  on other than loopback, except for LAMP....
[06:02] <lamont> it's that pesky security policy...
[06:02] <tarvid> client encourages users to work on the server without the benefit of cut and paste
[06:02] <dendrobates> If a user chooses a tasksel task it is not by default.
[06:03] <lamont> dendrobates: right
[06:03] <mathiaz> LAMP has an exception for the security policy - so why not openssh-server ?
[06:03] <nealmcb> the proposal is to have an option, "off" by default, during install to install openssh-server
[06:03] <lamont> in LAMP, it makes perfect sense to install openssh-server
[06:04] <mathiaz> I guess it would make sense to install it with every tasksel.
[06:04] <dendrobates> But it is certainly a commonly installed package, and we have been asked to look for new possible tasksel tasks.  This seems like it would be a simple choice.
[06:04] <nealmcb> does someone have a link to the policy and this lamp exception?
[06:04] <tarvid> tasksel could use a mail section
[06:04] <tarvid> mail coices should include esmtp
[06:05] <mathiaz> so what about adding openssh-server to the LAMP tasksel ?
[06:05] <mathiaz> and also to the bind9 task ?
[06:05] <tarvid> bind9 has rndc
[06:05] <mathiaz> but not to the default server install
[06:06] <infinity> There is no "LAMP exception", when you install LAMP, you're specifically requesting a server with open ports.
[06:06] <dendrobates> I'm not sure we should do that, why not just have it as a separate task?
[06:06] <jdstrand> that is my feeling with openssh-server task
[06:06] <infinity> It's not the default choice, after all.
[06:06] <nealmcb> infinity: that is what I thought
[06:06] <jdstrand> right, create a separate openssh-server task, it is not default.  It is a checkbox to install.
[06:06] <jdstrand> non-default, so ports are open
[06:07] <jdstrand> in the furture there will be a lot of tasks that are added that would benefit from openssh-server, having it separate makes sense to me.
[06:07] <dendrobates> I like it as a separate task, then the users knows they are opening the port
[06:07] <tarvid> you will far fewer broken server installs if users get off the server as soon as possible
[06:08] <dendrobates> anyone have any experience with tasksel?
[06:08] <jdstrand> tarvid: they need to check the lamp box anyway, they can check the ssh server box just as easily-- or am I missing something?
[06:08] <infinity> jdstrand: LAMP is selected from the boot screen, usually.
[06:08] <tarvid> nowhere do we encourage users to get off the server
[06:08] <infinity> jdstrand: We special-case it, as the "star task".
[06:09] <tarvid> users benefit from documentation and a lot of mine ask for a desktop on the server
[06:09] <infinity> tarvid: While I agree, in theory, the user can use whatever console they want, it's hardly my concern.  And if they're going to install a local X and other things, they're just as likely to try remote X, or try to break their box remotely. :)
[06:10] <nealmcb> tarvid: by "get off the server" you mean stop using the local display, and that it is good to make it easy to ssh in?
[06:10] <tarvid> yes
[06:10] <jdstrand> infinity: ok-- so are you suggesting a separate task, that LAMP should pull in as well?
[06:10] <jdstrand> infinity: perhaps with a note stating it happened
[06:10] <tarvid> mandrake had a package called msec which is not a bad concept
[06:10] <infinity> jdstrand: No, I'm pretty much with anyone who thinks openssh-server should be part of the LAMP task.
[06:11] <infinity> jdstrand: If we also want it in its own task, that's fine.
[06:11] <tarvid> might put netstat into the test php page so users know which ports are open
[06:12] <tarvid> why would a new user think to try tasksel in the first place, it is far from obvious
[06:12] <infinity> tarvid: The test page will be available on a public interface, having it expose too much info about the machine isn't bright.
[06:12] <nealmcb> I like the idea of a separate task for openssh.  adding it to lamp would be confusing
[06:12] <infinity> tarvid: The installer runs tasksel.
[06:12] <dantalizing> i dont get it...so a developer who wants lamp gets openssh-server too?
[06:12] <tarvid> ah security by obscurity
[06:12] <jdstrand> infinity: while I can't really think offhand why you would want LAMP without ssh, I think the extra port open needs to be documented somewhere.  Least surprise and all
[06:12] <dendrobates> I think it should be available as a task to users who don't select LAMP.  Perhaps the installer could automatically check the box when you select LAMP.
[06:13] <infinity> tarvid: "Don't show everyone your netstat output" isn't "security by obscurity".
[06:13] <Seveas> jdstrand, for servers that have an ILO card or are xen instances, an SSH server may be unneccessary
[06:13] <infinity> tarvid: It's privacy.
[06:13] <tarvid> it is to any hacker who has an nmap or equivalent
[06:13] <Seveas> (jdstrand: and I administer several of those ;))
[06:13] <jdstrand> Seveas: good point
[06:13] <dantalizing> if i'm a web developer of any kind, i want lamp and not necessarily ssh-server
[06:13] <jdstrand> (Seveas: I don't ;)
[06:14] <infinity> tarvid: netstat shows things nmap can't.
[06:14] <infinity> tarvid: Namely, who's connecting to you, and other fun stuff.
[06:14] <dendrobates> Ok so a separate openssh-server task seems to be the consensus.
[06:14] <jdstrand> why not have LAMP and LAMP+SSH?
[06:14] <Mithrandir> a task for a single package?
[06:14] <Mithrandir> sounds overkill to me..
[06:14] <tarvid> i wasn't suggest a raw dump of netstat simply a list of open ports
[06:14] <nealmcb> Seveas: good point about xen
[06:15] <infinity> A task for a single package does strike me as a bit silly.
[06:15] <tarvid> we are missing the objective to help new users get something running, hopefully what they want
[06:15] <dendrobates> it is a little, but I think some users would appreciate the ability to select sshd at startup.
[06:16] <tarvid> and some would prefer postgresql and some would prefer exim ...
[06:16] <tarvid> or postfix+courier
[06:17] <nealmcb> but this single package is pretty much the only non-default package you need to be able to admin a server remotely, so I vote for a visible task  for it
[06:18] <dendrobates> does anyone want to work on this?
[06:18] <dantalizing> this is only in reference to an ubuntu-server install, correct? not ubuntu?
[06:18] <dendrobates> dantalizing: yes.
[06:18] <nealmcb> https://help.ubuntu.com/community/Tasksel
[06:18] <tarvid> some day we are going to have to address maintaining numerous workstations
[06:19] <nealmcb> dantalizing: I would want to see it as an option for non-server installs also
[06:19] <sommer> tarvid: there was a good presentation on clonezilla at Ubuntu Live.
[06:20] <infinity> Tasks are at the archive level, if it exists for one flavour, it exists for all of them.
[06:20] <tarvid> i will take a look at that
[06:20] <ScottK> Well courier is in Universe and the package for it isn't in great shape.  I've given merging courier to prospective MOTUs as an "are you really ready to be a MOTU" exercise.
[06:21] <dendrobates> yes, but we are talking about installer integration.
[06:21] <dendrobates> we need to close the meeting.  I suggest we continue on #ubuntu-server.
[06:21] <jdstrand> I am not sure a decision can be made of whether we should enable ssh by default with LAMP unless we know how many users want it and don't want it.  The only other option is to offer choices.
[06:22] <jdstrand> By installing it by default, we will make another set of users say 'I need to uninstall this'
[06:22] <nealmcb> next meeting time?
[06:22] <nealmcb> do we want one next week?
[06:22] <mathiaz> next week is the kernel team meeting at this time.
[06:22] <dendrobates> mathiaz and I will look into another meeting, but at this time it is in two weeks.
[06:22] <mathiaz> so we'd have to move the meeting.
[06:23] <dendrobates> #endmeeting
[06:23] <MootBot> Meeting finished at 16:16.
[06:23] <nealmcb> thanks, all!
[06:23] <tarvid> can some of this be moved to the mailing list so we canb proceed asynchronously
[06:23] <mathiaz> I'll write up some notes.
[06:24] <dendrobates> mathiaz: lets see what mootbot does for us.
[06:25] <mathiaz> dendrobates: yeah. I think it will generate some preview things. But it may be much easier to edit it.
[07:17] <zul> @schedule montreal
[07:17] <ubotu> Schedule for America/Montreal: 14 Aug 15:00: Technical Board | 15 Aug 16:00: Edubuntu | 16 Aug 16:00: Ubuntu Development Team | 20 Aug 18:00: Forum Council meeting | 21 Aug 11:00: Kernel Team | 22 Aug 08:00: Edubuntu
[07:18] <Evergete> @schedule rome
[07:18] <ubotu> Schedule for Europe/Rome: 14 Aug 21:00: Technical Board | 15 Aug 22:00: Edubuntu | 16 Aug 22:00: Ubuntu Development Team | 21 Aug 00:00: Forum Council meeting | 21 Aug 17:00: Kernel Team | 22 Aug 14:00: Edubuntu
[08:57] <calc> hello
[08:57] <mdz_> hi
[08:57] <mjg59> Evening
[08:58] <calc> btw i have a test patch that may fix the ooo bug completely
[08:58] <calc> i already found where the bug is and was able to work around it locally by disabling the bad patch, if this corrected patch works I can upload OOo with working fontconfig hinting support as well (thats what i had to disable to get it to work locally)
[09:02] <mdz_> mjg59: seen Keybuk?
[09:02] <mdz_> have pinged sabdfl
[09:02] <mjg59> Not since earlier
[09:03] <mjg59> Can wait a few minutes, I guess
[09:03] <mdz_> phoning
[09:03] <mdz_> Keybuk: just phoned you
[09:03] <Keybuk> oh?
[09:03] <mdz_> yes, I don't think I've ever successfully reached you on your phone though :-P
[09:04] <mdz_> SMSed sabdfl
[09:04] <Keybuk> ah yes, I have a missed call
[09:04] <calc> heh
[09:05] <calc> when my phone is in my pocket i can't hear it even when its set to loudest setting :\
[09:06] <mdz_> keescook: we're waiting for sabdfl to process developer applications, but we can start discussing your item now
[09:07] <keescook> mdz_: okay, excellent.  the proposal was designed during the sprint, and we figured it'd need TB approval.
[09:07] <mdz_> keescook: the proposal seems to more or less document current practice
[09:08] <mdz_> the one point I'm uncertain about is the second bullet
[09:08] <mdz_> #startmeeting
[09:08] <MootBot> Meeting started at 19:01. The chair is mdz_.
[09:08] <MootBot> Commands Available: [TOPIC] , [IDEA] , [ACTION] , [AGREED] , [LINK] , [VOTE] 
[09:08] <mdz_> [TOPIC]  Review StableReleaseUpdates/MicroReleaseExceptions -- KeesCook
[09:08] <MootBot> New Topic:  Review StableReleaseUpdates/MicroReleaseExceptions -- KeesCook
[09:08] <keescook> right, the basic "unknown" was "who approves packages that get micro-version updates instead of minimal patches for SRU?"
[09:08] <mdz_> (pasting for MootBot's benefit)
 keescook: the proposal seems to more or less document current practice
[09:08] <mdz_>  the one point I'm uncertain about is the second bullet
[09:09] <keescook> the 2nd bullet is for when a micro update includes a security fix, and extracting it is hard (i.e. firefox)
[09:09] <mdz_> keescook: this point seems to be addressing a different issue than the others
[09:09] <mjg59> mdz_: Can you clarify which level of indentation you're referring to?
[09:09] <mdz_> mjg59: outermost
[09:09] <keescook> it would allow for the security team to publish a micro update via -security instead of via an SRU
[09:10] <mdz_> On a case-by-case basis, if the security team rates the chance of vulnerability exploitation greater than the chance of regression for a given update, it can skip SRU and go directly through the security queue.

[09:10] <mdz_> keescook: the security team already publishes updates via -security per default, no?
[09:10] <mdz_> there's always been a sort of implicit exception for security vis SRU
[09:10] <mdz_> [LINK]  https://wiki.ubuntu.com/StableReleaseUpdates/MicroReleaseExceptions
[09:10] <MootBot> LINK received:  https://wiki.ubuntu.com/StableReleaseUpdates/MicroReleaseExceptions
[09:10] <keescook> mdz_: correct, this is basically providing a process outline for what is already done
[09:11] <keescook> the "new" part is the "how do we add packages to the micro-update exception list"?
[09:11] <keescook> this grew from a desire to do mysql micro-releases in SRU
[09:11] <keescook> and php
[09:11] <keescook> (for example)
[09:11] <mdz_> keescook: ok, so we're addressing two separate exceptions: 1. security updates in general being exempt from SRU by nature, and 2. exceptions to SRU for specific packages where it would otherwise apply ?
[09:12] <keescook> 1) is already well-documented (bullet 2 is a clarification about how security and SRU processes might collide)
[09:12] <mdz_> the first being documentation of current practice, and the second a proposed change
[09:13] <keescook> 2) "we want to do micro-updates of packages, current SRU process requires a 'minimal patch': we want to do upstream versions"
[09:13] <keescook> this proposal is about SRU (the 2nd bullet just confused things by bringing up security)  The 3rd bullet was pitti and my first guess at who should own the exception list, we also considered the TB
[09:14] <mdz_> keescook: should we remove the second point then?
[09:14] <keescook> I don't might dropping it; it was clear to me, but someone at the sprint wanted it clarified, so I saw no harm in adding it.
[09:15] <mdz_> keescook: I don't see harm in clarifying that in the SRU page, if it isn't already, but it has nothing to do with the proposal regarding micro-updates as far as I know
[09:15] <mjg59> In the absence of the second point, who is responsible for handling the decision process?
[09:15] <keescook> okay, noted.
[09:15] <mdz_> SRU says: "This process is to be followed for all updates except those to fix security updates, which are only released by the Ubuntu security team. Security procedures are documented at SecurityUpdateProcedures."
[09:16] <keescook> mjg59: that's part of the decision I'm seeking.  pitti and I were proposing the security team own it, but it felt like something the TB might want to own.
[09:16] <mdz_> mjg59: the decision process is addressed in the third point
[09:16] <mdz_> "The security team will have the responsibility to maintain and review the list of packages that are exceptions to the SRU rule, as well as approve package additions."
[09:16] <mdz_> I'm not so sure that the security team is the right place to make this decision
[09:16] <mdz_> especially since security updates are specifically excluded from the policy
[09:16] <keescook> I'm happy changing it; we just wanted to have a starting point to work from. :)
[09:17] <keescook> well, note, of course, that the micro-update rule _does_ apply to security updates.  SUP requires minimal patches too.
[09:17] <keescook> this would be the documented exception to SUP as well (since we already do this for firefox, etc)
[09:17] <mdz_> keescook: ok, please drop the security exception then, since SRU already has an exception in it
[09:18] <mdz_> keescook: you can reference the same exception in SUP if you like
[09:18] <keescook> I will drop point 2, and clarify the exception as an SRU/SUP exception.
[09:18] <mdz_> ok
[09:19] <mdz_> so I think a reasonable process for adding packages to the list is for the proposer (whomever that might be) to assess whether the package meets the criteria, and mail that to the tech board for confirmation
[09:19] <mdz_> mjg59,Keybuk: ?
[09:20] <mjg59> Is the assumption that we'll approve by mail?
[09:20] <keescook> what level of +1ing will be required via email?
[09:20] <mdz_> mjg59: yes
[09:20] <mdz_> keescook: I think confirmation by any TB member is sufficient in general
[09:21] <Keybuk> that seems reasonable
[09:21] <mjg59> Ok. If we're going to this level of documentation, I'd probably prefer it if the approval process is mentioned
[09:21] <keescook> one thing that occurs to me now is do we need to change the SRU verification process at all?  with a micro-bump, there may be larger areas of testing needed.
[09:21] <mdz_> agreed, the approval process should be included
[09:22] <mdz_> keescook: the current policy basically leaves it to the discretion of the SRU team, no?
[09:23] <keescook> I guess that's true, yes.
[09:23] <mdz_> in which case, I think that's sufficient.  they can judge case-by-case what sort of testing is needed
[09:24] <mdz_> IIRC the tech board more or less blessed the existing exceptions
[09:24] <mdz_> so they can be grandfathered in
[09:24] <mdz_> any other comments or concerns about kees' proposal?
[09:25] <keescook> okay, I will mark them as such; I've adjusted the exception wiki to reflect the TB approval process.  can you ACK?
[09:25] <mdz_> keescook: if you make the edits now, we can go ahead and approve it officially
[09:25] <mjg59> Sounds good to me
[09:25] <mdz_> keescook: I'd note that the request for an exception should include an explanation of how the package meets the criteria
[09:25] <mdz_> that is, the tech board isn't expected to do the research, just verify it
[09:25] <keescook> mdz_: ah, right
[09:26] <keescook> "The request is expected to include justification against the above criteria."
[09:26] <mdz_> fine with me
[09:27] <mdz_> ok
[09:28] <keescook> okay, edits made; I've marked it "ratified", based on +1s from mdz_ and mjg59
[09:28] <mdz_> [VOTE]  keescook's proposal at https://wiki.ubuntu.com/StableReleaseUpdates/MicroReleaseExceptions, revision 4
[09:28] <mjg59> Right, sounds fine by me
[09:28] <MootBot> Please vote on:  keescook's proposal at https://wiki.ubuntu.com/StableReleaseUpdates/MicroReleaseExceptions, revision 4.
[09:28] <mdz_> +1
[09:28] <MootBot> +1 received from mdz_. 1 for, 0 against. Count is now 1
[09:28] <mjg59> +1
[09:28] <MootBot> +1 received from mjg59. 2 for, 0 against. Count is now 2
[09:28] <Keybuk> +1
[09:28] <MootBot> +1 received from Keybuk. 3 for, 0 against. Count is now 3
[09:28] <mdz_> #endvote
[09:28] <MootBot> Final result is 3 for, 0 against. Total: 3
[09:29] <keescook> hadn't seen the voting features of MootBot before.  nice.  :)
[09:29] <mdz_> [AGREED]  ratified  https://wiki.ubuntu.com/StableReleaseUpdates/MicroReleaseExceptions
[09:29] <MootBot> AGREED received:  ratified  https://wiki.ubuntu.com/StableReleaseUpdates/MicroReleaseExceptions
[09:29] <mdz_> [ACTION]  keescook to update StableReleaseUpdates accordingly
[09:29] <MootBot> ACTION received:  keescook to update StableReleaseUpdates accordingly
[09:29] <keescook> great, thanks everyone.  :)
[09:29] <mdz_> sabdfl says he is having connectivity problems
[09:29] <mdz_> we will go ahead without him
[09:29] <mjg59> Ok.
[09:30] <mdz_> [TOPIC]  application from Chris Cheney for ubuntu-core-dev: https://lists.ubuntu.com/archives/motu-council/2007-July/000236.html
[09:30] <MootBot> New Topic:  application from Chris Cheney for ubuntu-core-dev: https://lists.ubuntu.com/archives/motu-council/2007-July/000236.html
[09:30] <mdz_> calc: still here?
[09:30] <calc> mdz_: yes
[09:31] <mdz_> ok, good
[09:31] <mdz_> (reading your wiki page)
[09:31] <Keybuk> calc: you applied for MOTU only recently, correct?
[09:31] <calc> Keybuk: around June 13 iirc
[09:32] <calc> hmm June 23 actually
[09:32] <mdz_> calc: you've been a Debian developer for quite a while, but only recently got involved with Ubuntu development.  What are you finding different about it?
[09:33] <calc> mdz_: well merges aren't needed on the debian side of course, but also translations being done via launchpad, and the split of upload ability between main/universe, probably other things i am forgetting at the moment
[09:34] <calc> and the much quicker release cycle :)
[09:34] <mdz_> calc: the release cycle is a big one.  core developers are expected to "have a strong working knowledge of Ubuntu project procedures, especially those related to the release process and support commitments, and an understanding of the reasons why they exist"
[09:35] <calc> ok
[09:35] <mdz_> calc: feature freeze is this Thursday.  what will change in terms of development activity at that time?
[09:36] <calc> only bug fix uploads are allowed past that point unless an exception is requested and granted
[09:37] <calc> which i will be filing for OpenOffice since its final release is close to the release date of Gutsy overall
[09:37] <mdz_> calc: indeed.  this implicitly means it is the deadline for feature projects
[09:37] <mdz_> calc: eek.  how close?
[09:38] <calc> the place i am looking just mentions sept 2007
[09:38] <calc> i think another page may have more detailed date, but i don't recall the url for it
[09:38] <mdz_> calc: what does our release manager think about it?
[09:39] <calc> openoffice and ubnutu releases have generally been close together
[09:39] <calc> er ubuntu
[09:40] <calc> eg oo 2.2 mar 2007, 2.0.4 sep 2006, etc
[09:40] <calc> mdz_: i believe that he knows about the release schedule but i'll ask him about it again to make certain he is completely aware of the situation
[09:41] <mdz_> the more advance notice you can give, the better.  last-minute exceptions make for grumpy release managers
[09:41] <sabdfl> hi all sorry for the delay, network issues
[09:41] <mdz_> sabdfl: reviewing  application from Chris Cheney for ubuntu-core-dev: https://lists.ubuntu.com/archives/motu-council/2007-July/000236.html
[09:41] <mdz_> sabdfl: (calc)
[09:41] <calc> mdz_: yes, i am going to make sure not to let things run late like *cough* tribe-4 *cough* :\
[09:41] <sabdfl> no reservations on my side
[09:42] <calc> mdz_: i ran into several unforeseen snags at the last minute so i'm going to make sure to get it done in plenty of time from now on, so I won't be annoying the RM anymore
[09:42] <mdz_> calc: another defining characteristic for core developers is participation in planning feature projects.  do you have any plans in this area for gutsy+1?
[09:42] <calc> mdz_: not yet
[09:42] <mdz_> calc: do you know how the feature planning process works overall?
[09:43] <calc> mdz_: i am interested in the hardware database situation but last time i looked it appeared to be on hold
[09:43] <calc> mdz_: i have reviewed it before but I don't recall all the details of how it works
[09:44] <calc> mdz_: high level overview that i do remember is create a plan in the wiki and then present it at UDS for other people to critique
[09:44] <mdz_> calc: I'm not sure how well documented it is, to be honest
[09:44] <mdz_> calc: most ideas get refined at UDS, yes
[09:45] <mdz_> sabdfl,mjg59,Keybuk: do you have any questions about calc's application?
[09:45] <mdz_> keescook: would you like to say anything on Chris' behalf, as his sponsor?
[09:46] <keescook> yup, he's a quick learner for all the merge work we did together, and the OOo updates all went without regression, so I'm quite happy there too
[09:46] <sabdfl> none, +1
[09:46] <keescook> his pre-existing debian packaging knowledge is better than mine.  ;)
[09:47] <mjg59> Given his level of involvement in Debian and what I've seen so far, I'm pretty happy
[09:47] <mdz_> calc: are you making some progress in getting involved upstream?
[09:47] <calc> mdz_: yes, on the ooo-build side of things I have had my first set of patches committed so I can now file for direct svn access to the gnome svn repo
[09:47] <mdz_> calc: that's excellent, thanks
[09:47] <Keybuk> the short time in MOTU is a concern for me since he won't be as familar with our procedures and how they differ from what he's used to, but otherwise +1
[09:48] <mdz_> [VOTE]   application from Chris Cheney for ubuntu-core-dev
[09:48] <MootBot> Please vote on:   application from Chris Cheney for ubuntu-core-dev.
[09:48] <calc> mdz_: on the OOo (sun) side last I checked they hadn't officially accepted my copyright waiver yet so I don't know if I can make changes there yet
[09:48] <mjg59> +1
[09:48] <MootBot> +1 received from mjg59. 1 for, 0 against. Count is now 1
[09:48] <calc> mdz_: but I am subscribed to the various lists so I keep track of things on the lists also
[09:48] <Keybuk> +1
[09:48] <MootBot> +1 received from Keybuk. 2 for, 0 against. Count is now 2
[09:48] <mdz_> +1, based on long Debian track record and recent involvement.  you seem to be getting up to speed very quickly
[09:48] <mdz_> +1
[09:48] <MootBot> +1 received from mdz_. 3 for, 0 against. Count is now 3
[09:49] <calc> Keybuk: feel free to lart me anytime I make a mistake, but I'll do my best to not screw up anything :)
[09:49] <sabdfl> +1
[09:49] <MootBot> +1 received from sabdfl. 4 for, 0 against. Count is now 4
[09:49] <mdz_> #endvote
[09:49] <MootBot> Final result is 4 for, 0 against. Total: 4
[09:49] <mdz_> calc: congratulations and thanks
[09:49] <calc> thank you all :)
[09:49] <sabdfl> erk, MootBot, listen up in advance ;-)
[09:49] <mdz_> [ACTION]  somebody to update Launchpad
[09:49] <MootBot> ACTION received:  somebody to update Launchpad
[09:49] <calc> er
[09:49] <mdz_> [TOPIC]  application for ubuntu-core-dev from Lionel Le Folgoc: [WWW]  https://lists.ubuntu.com/archives/motu-council/2007-June/000162.html
[09:49] <MootBot> New Topic:  application for ubuntu-core-dev from Lionel Le Folgoc: [WWW]  https://lists.ubuntu.com/archives/motu-council/2007-June/000162.html
[09:49] <mdz_> mr_pouit: are you here?
[09:50] <mr_pouit> mdz_: yes
[09:50] <mdz_> exellent
[09:50] <mdz_> gpocentek: are you here?  you seem to be Lionel's main sponsor
[09:51] <sabdfl> calc: are you ccheney in LP?
[09:51] <Keybuk> calc: it's ok, I don't mind unexpected hugs
[09:51] <mdz_> mr_pouit: I didn't know we had French forums.  are they entirely separate from the english language forums?
[09:52] <calc> sabdfl: yes
[09:52] <sabdfl> done
[09:52] <mr_pouit> mdz_: yes, they are hosted on a server with several other european locoteams iirc
[09:53] <mdz_> mr_pouit: is that out of necessity, or just because it was more convenient?
[09:54] <mr_pouit> I think it was more convenient, but the "main" admin (yann2) would give a better answer
[09:55] <mr_pouit> (because there is also a separate wiki and planet...)
[09:55] <sabdfl> mr_pouit: could you summarise your plan on the cdrecord front?
[09:56] <keescook> I have to run off, but I'd like to make a quick comment on mr_pouit's application: USN-497-1 was almost entirely his work, and even included a proposed USN description.  excellent work, imho. (https://launchpad.net/bugs/132046)
[09:56] <ubotu> Launchpad bug 132046 in xfce4-terminal "SECURITY: URL handling allows remote shell command execution" [Medium,Fix released] 
[09:57] <mr_pouit> sabdfl: my plans were to promote libburnia, which is an alternative currently under development, but unfortunately it is nt mature enough yet, so for gutsy at least, I think cdrkit is still the solution
[09:57] <mdz_> mr_pouit: you mention that you've been active in reviewing packages on REVU.  what kind of things do you look for when reviewing packages?
[09:57] <mdz_> keescook: thanks
[09:58] <sabdfl> are there areas other than cd burning and cubuntu that you are interested in?
[09:58] <sabdfl> erk... x
[09:58] <Mithrandir> (are non-TB people allowed to ask applicants questions?)
[09:58] <mdz_> Mithrandir: sure
[09:58] <mdz_> Mithrandir: just not vote ;-)
[09:58] <Mithrandir> mdz_: sure. :-)
[09:59] <mdz_> though MootBot doesn't enforce that at the moment
[09:59] <Mithrandir> mr_pouit: what's the relation of libburnia and libisofs/libburn?
[10:00] <mr_pouit> Mithrandir: I think there was a lib called libisofs which existed, but a fork was created, and called libburnia, but ships libburn and libisofs
[10:01] <mr_pouit> mdz_: I try to do as if I was making the package
[10:01] <Mithrandir> mr_pouit: so http://icculus.org/burn/ is superseded by http://libburnia.pykix.org/ ?
[10:02] <mr_pouit> mdz_: so copyright issues, respect of most points of the debian policy (except nmus for instance), I buid it in pbuilder, and run it then...
[10:02] <mdz_> mr_pouit: are there any particular policy errors which are common in your experience?
[10:03] <mr_pouit> Mithrandir: I think that the one at icculus.org isn't very active anymore
[10:03] <mr_pouit> so yeah
[10:03] <Mithrandir> mr_pouit: ok, thanks.
[10:04] <highvolt1ge> mr_pouit: wow, it would be really cool to have libburnia in ubuntu!
[10:04] <mr_pouit> highvolt1ge: it's already in universe
[10:04] <highvolt1ge> s/ubuntu/main :)
[10:04] <mjg59> It's good to see that there's work being done in this field
[10:04] <mjg59> The continued dependence on driving cdrkit is less than idea
[10:05] <pygi> hello
[10:05] <mr_pouit> pygi: :)
[10:06] <mr_pouit> (pygi is taking care of libburnia)
[10:06] <mr_pouit> mdz_: the most common are copyright issues...
[10:06] <mdz_> (delivery at the door, brb)
[10:06] <mr_pouit> (missing full licenses texts in the orig)
[10:10] <mjg59> mr_pouit: What are your plans for integrating improved CD support in main?
[10:10] <mdz_> (back)
[10:10] <mdz_> mr_pouit: what about technical issues?  what kinds of technical errors have you noticed in the packages you've reviewed?
[10:10] <mjg59> As in, integration into the distribution (rather than just providing stuff)
[10:11] <sabdfl> mr_pouit: are you interested in the GUI / user experience of cd burning, or just the kernel / console aspect?
[10:12] <mr_pouit> mjg59: I use and try to take care of brasero for some time, and I think it could be the good app to improve burning experience (for gtk used at least)
[10:13] <mjg59> mr_pouit: Sounds promising
[10:13] <mr_pouit> sabdfl: more for the gui aspect with brasero at the moment
[10:14] <sabdfl> ok
[10:14] <mr_pouit> mdz_: issues with libraries (sonames for instance, files in worng packages {,-dev})
[10:14] <sabdfl> no further questions, yer-honner
[10:14] <pygi> please allow me to kick in just for a second?
[10:15] <mjg59> Sure
[10:15] <pygi> I'd just like to say that from the low-level stuff regarding cd-recording mr_pouit (and everyone else who wants it) has and will have all the support from me for as long as I'm here
[10:16] <pygi> (with no idea how long will that be anymore, but ...)

[10:17] <mjg59> Ok. Any more questions?
[10:17] <mdz_> mr_pouit: how well does the Ubuntu release process serve Xubuntu?  are there things that you think would work better if changes were made?
[10:18] <mr_pouit> mdz_: I think that's almost ok... I only feel that xubuntu needs a bit more love.
[10:18] <mr_pouit> for instance, I don't think enough people test alpha release
[10:19] <mr_pouit> (and for xubuntu, a big problem is gnome and its dependencies...)
[10:19] <mdz_> mr_pouit: how is gnome a big problem?
[10:20] <mr_pouit> xubuntu was aimed for not so fast machines, and including gnome apps would have increased memory needs
[10:21] <mr_pouit> (I am not totally convinced that's still the case)
[10:21] <mdz_> mr_pouit: so the problem is what exactly?  the lack of gnome apps?
[10:21] <mr_pouit> it's sometimes hard to find gtk only apps which have similar features to gnome ones
[10:22] <mdz_> assessing memory requirements is tricky.  Ubuntu is difficult to install on some systems with 256M of memory, but many systems share 25% of that with video
[10:23] <mdz_> mr_pouit: ok, so the problem isn't gnome, but a lack of alternatives to gnome :-)
[10:23] <mr_pouit> yes :)
[10:23] <mdz_> any other questions for mr_pouit?
[10:24] <mdz_> [VOTE]   Lionel Le Folgoc for ubuntu-core-dev
[10:24] <MootBot> Please vote on:   Lionel Le Folgoc for ubuntu-core-dev.
[10:25] <sabdfl> +1
[10:25] <MootBot> +1 received from sabdfl. 1 for, 0 against. Count is now 1
[10:25] <mjg59> +1
[10:25] <MootBot> +1 received from mjg59. 2 for, 0 against. Count is now 2
[10:25] <mdz_> +1 good support from sponsors and record of strong participation
[10:25] <MootBot> +1 received from mdz_. 3 for, 0 against. Count is now 3
[10:25] <mdz_> Keybuk: ?
[10:26] <mdz_> ok, need to move on.  don't know what happened to Keybuk
[10:26] <mdz_> #endvote
[10:26] <MootBot> Final result is 3 for, 0 against. Total: 3
[10:26] <mdz_> mr_pouit: congratulations
[10:26] <mdz_> [ACTION]  sabdfl to update launchpad
[10:26] <MootBot> ACTION received:  sabdfl to update launchpad
[10:26] <mr_pouit> thank you all :)
[10:27] <mdz_> [TOPIC]  Compiz by default -- MatthewGarrett
[10:27] <MootBot> New Topic:  Compiz by default -- MatthewGarrett
[10:27] <mjg59> Want me to introduce this?
[10:27] <mdz_> mjg59: I've read over your email, but yes, please
[10:27] <mjg59> Ok
[10:27] <mjg59> The summary is that compiz still has pretty much the same level of functional regression as it did at this point in the feisty release cycle
[10:28] <mjg59> The main improvement has been in its level of window management
[10:28] <sabdfl> done
[10:28] <mjg59> We're still faced with two major problems - anything using Xv and any windowed 3D applications
[10:28] <mdz_> its configuration compatibility with metacity has also improved greatly
[10:29] <sabdfl> Xv the old image display program?
[10:29] <mvo> basic Xv support is in place for video-i810 and video-ati
[10:29] <mdz_> sabdfl: no, the Xvideo extension
[10:29] <mjg59> sabdfl: Xv as in accelerated X video playback
[10:29] <mvo> nothing composited though
[10:29] <sabdfl> ok
[10:29] <sabdfl> thanks
[10:29] <mdz_> the thing which does video scaling and colorspace conversion in hardware
[10:29] <sabdfl> that's because of overlays?
[10:29] <mvo> I have a pending patch for -video-intel
[10:29] <mjg59> sabdfl: Yes
[10:29] <mdz_> Xv works fine for me, it just isn't composited
[10:30] <mjg59> Some quantity of basic functionality can be restored, which gets us to the point where video playback works providing you're not doing anything that transforms the window
[10:30] <mvo> windowed 3d will not be ready for gutsy
[10:30] <mdz_> 3D is a bit messier
[10:30] <mdz_> mvo: Xv compositing was demoed at UDS, no?
[10:31] <mvo> mdz_: I was not in charge back then I can not say for sure. we have patches for mplayer that work
[10:31] <mvo> but nothing for gstreamer
[10:31] <mvo> and its not a straightforward port
[10:32] <mdz_> the issues with Xv and windowed 3D are cosmetic and pretty easy to work around by disabling desktop effects
[10:32] <mvo> upstream thinks that the compiz video interface is not the final one so they are reluctant to add support for a moving target (from what I have heared)
[10:32] <mjg59> Right, David Reveman has implemented a means of providing accelreated playback with composited output
[10:32] <mdz_> I'm more concerned about the 3D performance issue, which is less obvious and a functional problem
[10:32] <mjg59> But it requires moderately heavy patches to the Xv application
[10:32] <mjg59> So it's not a general solution
[10:32] <mvo> indeed
[10:33] <mjg59> The 3D performance issue is due to all GL operations now having to go via the X server
[10:33] <mjg59> There's no intrinsic reason for this to cause a significant loss of performance
[10:33] <mjg59> But right now, it does
[10:33] <mjg59> And it's going to take a while for that to be rewritten
[10:33] <mdz_> with the nvidia driver, I saw 3D performance issues even full-screen
[10:34] <mjg59> I'd argue that the issues are somewhat more than cosmetic - the damage left behind by 3D applications right now is likely to obscure sections of any 3D-using application that's more ocmplicated than a single window
[10:34] <mjg59> mdz_: That shouldn't be the case - nvidia have an indirected X implementation that works properly
[10:35] <mdz_> I can have a look at it tomorrow if I know what to look for
[10:35] <mdz_> but in short, it was dog slow
[10:36] <mjg59> As mdz says, these are issues that can be worked around by disabling desktop effects
[10:36] <mjg59> But they are functional regressions, and I don't know that we've fully weighed the cost/benefit analysis
[10:37] <mjg59> (This is why RH still haven't shipped compiz by default in Fedora)
[10:37] <sabdfl> if a user turns off desktop effects, do both problems disappear?
[10:37] <mjg59> Yes
[10:37] <mjg59> The question is whether we want bling but broken or working but dull
[10:37] <sabdfl> we don't restart X, do we?
[10:37] <mjg59> No
[10:38] <mdz_> I don't see a functional problem with video
[10:38] <sabdfl> surely extensions / aiglx etc are there all the time then?
[10:38] <mdz_> which chipsets are affected by that?
[10:38] <sabdfl> can we detect when an app is using wither Xvideo or GL?
[10:38] <mjg59> mdz_: Anything other than nvidia to varying extents
[10:38] <mjg59> sabdfl: The window manager can't, no
[10:38] <mdz_> mjg59: I'm on ATI here and it works fine, just isn't composited
[10:38] <sabdfl> and pop up a warning - which takes them to the desktop effects / appearance dialog?
[10:38] <mjg59> sabdfl: Nope
[10:38] <mdz_> windowed and fullscreen
[10:39] <mjg59> sabdfl: It's not the presence of the extensions - it's whether you're rendering to an offscreen framebuffer or not
[10:39] <mvo> I pateched -video-{intel,i810} today to be at the same level as -ati
[10:39] <mvo> patched even
[10:39] <sabdfl> mjg59: can something else monitor X and tell when those subsystems are activated by an app?
[10:39] <sabdfl> ok
[10:40] <mjg59> sabdfl: In theory it's possible
[10:40] <mdz_> not really, that information is in the guts of the X server
[10:40] <mjg59> But it would involve huge hacks to the server
[10:40] <sabdfl> ok
[10:40] <sabdfl> jez-askin
[10:40] <mdz_> i.e., not 2 days before feature freeze
[10:40] <mdz_> I have different feelings about the 3 different issues mjg59 has raised
[10:41] <mdz_> I think that video, if mvo is correct and it is as functional on intel as what I see on ati and nvidia, is a non-issue
[10:41] <mdz_> it looks funny while window is being dragged, but then is perfect as soon as it is placed
[10:42] <mdz_> and full-screen works without a hitch
[10:42] <mdz_> I shouldn't say non-issue
[10:42] <mdz_> but I think the bling factor outweighs the cosmetic issue
[10:42] <mdz_> since it's cosmetic vs. cosmetic
[10:43] <mdz_> the second issue, of windowed 3D applications not being composited, seems like another example of the same
[10:43] <mjg59> mdz_: It's not purely a matter of them not being composited in that case
[10:43] <mdz_> but worse
[10:43] <mjg59> It's the fact that you get no damage signals
[10:43] <mjg59> So applications aren't redrawn until you move them
[10:43] <mdz_> when i drag the window and release it, the stuff at the old location is clobbered with a frame of glxgears
[10:44] <mjg59> Yes
[10:44] <mdz_> and that's pretty ugly
[10:44] <mjg59> It gives the impression of something that doesn't work properly
[10:44] <mdz_> so I'm on the fence about that one at the moment
[10:44] <mjg59> And I don't think our default desktop should give that impression
[10:44] <mdz_> mostly because 3D applications are a) fairly rare, and b) in a professional setting, not deployed on intel graphics
[10:45] <mdz_> nobody uses intel graphics for serious 3D work
[10:45] <mjg59> mdz_: It's not limited to Intel
[10:45] <mdz_> and that's the only place where desktop effects are enabled by default
[10:45] <mjg59> Ati will also have the issue
[10:45] <mdz_> well, and some ancient ATI like my lpatop
[10:45] <mdz_> laptop
[10:45] <mvo> it will be enalbed on -ati too
[10:45] <mjg59> mdz_: Well, ancient is unfair - anything up to a bit over a year ago
[10:46] <mdz_> oh, things must have gotten better in -ati since I was last informed
[10:46] <mjg59> At least, we do nothing to prevent it being run on r3/400
[10:46] <mdz_> so it works on r300?
[10:46] <mjg59> I hope so, otherwise a lot of people are going to be very unhappy
[10:46] <mvo> I do not have a r300 for testing, but it should work
[10:47] <mdz_> I'm even more uncertain, then
[10:48] <mjg59> The r300/r400 Radeons were common in business-class laptops for the past few years
[10:48] <mdz_> I don't know to what extent people rely on that configuration
[10:49] <mdz_> sabdfl,Keybuk: thoughts?
[10:49] <mdz_> (I sent an SMS to Keybuk, but no response.  I hope he's OK...)
[10:50] <sabdfl> mjg59: do you expect these issues to be much better in the next 4-6 months?
[10:50] <mjg59> sabdfl: In terms of video, yes
[10:50] <mjg59> Intel already have textured video support, but it only works with exa
[10:50] <mjg59> Which we can probably switch to post-gutsy
[10:50] <mjg59> 3D? Very hard to say
[10:51] <mjg59> The basic code exists, but it depends on TTM. I don't think anyone is going to put any money on when that'll land
[10:51] <mdz_> mjg59: exa post-gutsy -> even in the context of 8.04 LTS?
[10:51] <mjg59> At the moment only RH are doing any real development on it
[10:51] <mjg59> mdz_: Well, yeah, that's the obvious concern
[10:51] <sabdfl> TTM?
[10:51] <mjg59> Intel, at least, are putting most of their dev work into EXA now
[10:52] <mjg59> sabdfl: A new memory manager for 3D
[10:52] <sabdfl> EXA>
[10:52] <sabdfl> talloc based?
[10:52] <mjg59> sabdfl: A new(ish) acceleration architecture for 2D operations
[10:52] <mjg59> sabdfl: No - the main difference is that memory allocations are no longer static
[10:52] <mjg59> Textures can be moved between main memory and card memory much more easily
[10:52] <sabdfl> if we expect it to be ok for gutsy+1 LTS, then i would say let's do it, the feedback will give us more confidence for the LTS dcision
[10:53] <mjg59> I have no faith in TTM being ready within 12 months
[10:53] <sabdfl> if we expect it to be not-great for the LTS, then we should realise we are looking at a year's delay
[10:53] <mjg59> These problems are hard
[10:53] <sabdfl> ok
[10:53] <mjg59> There's a lot of real development work that needs to be done
[10:55] <mdz_> as a (subjective) data point, I haven't seen any reviews of Gutsy comment on the fact that compiz is enabled by default in these situations
[10:55] <mdz_> I do see positive press about the effects themselves though
[10:55] <mjg59> My personal opinion is that this technology is still not ready for default use
[10:56] <mjg59> Bear in mind that reviewers at this point know that they're dealing with a beta release
[10:56] <sabdfl> mjg59: are you OK with the experience on Intel, and the roadmap for Intel graphics between now and 8.04?
[10:56] <mjg59> The general community mindset often seems to be that public criticism of non-working features before final release is unfair
[10:57] <mjg59> sabdfl: I doubt the 3D issues will be solved on Intel by 8.04
[10:57] <mjg59> The code currently exists, but requires features that probably aren't going to hit the mainstream kernel by then
[10:57] <mjg59> And currently performance is in the 12fps range
[10:58] <mdz_> and while it's cranking away at 12fps, at least on nvidia, the rest of the system is unusable
[10:58] <sabdfl> but, turning off the effects solves that problem completely
[10:58] <mdz_> the X server seems bottlenecked
[10:58] <mdz_> sabdfl: correct
[10:58] <mjg59> sabdfl: Yes, but I think asking users to turn off effects in order to get 1995-level functionality is a bit unreasonable
[10:59] <mdz_> hey, we barely had glxgears in 1995
[10:59] <sabdfl> hmm... the only 3D graphics I see on my laptop are the screensavers
[10:59] <mdz_> at all
[10:59] <mjg59> mdz_: Yeah, but they at least worked properly :)
[10:59] <mdz_> probably around 12fps
[11:00] <mjg59> (Without leaving damage)
[11:00] <sabdfl> do the screensavers disable composite before they kick in?
[11:00] <mjg59> sabdfl: They're full-screen, so less of an issue
[11:00] <mdz_> there's no issue with moving the window, as they're full-screen
[11:01] <sabdfl> from my perspective, it feels like trading cosmetic issues
[11:01] <sabdfl> and being less technical, i would lean to the new-and-exciting rather than the older, more functional capability
[11:01] <mdz_> sabdfl: not in the case of the performance problem
[11:01] <mjg59> Kind of. In one case we maintain the status quo - in another we add stuff we didn't previously have, at the cost of breaking things that used to work
[11:02] <mdz_> there are functional regressions with 3D
[11:02] <sabdfl> the performance problem goes away, though, if you disable the effects?
[11:02] <mdz_> correct
[11:02] <mjg59> sabdfl: Yes, but how do we communicate that to users?
[11:02] <mdz_> new desktop wallpaper
[11:02] <mdz_> "if anything looks wonky, try this"
[11:02] <sabdfl> we have excellent forums :-)
[11:03] <mdz_> it's easy to write it off if the user has enabled the effects, then has a problem like this which can be solved by reverting
[11:04] <mdz_> trickier when they haven't done anything and has a problem out of the box
[11:04] <mdz_> s/has/have/
[11:04] <mjg59> mdz_: Translation nightmare
[11:04] <calc> i hear that compiz causes much higher power consumption on laptops too (not sure if this is actually true) so should it be default on laptops?
[11:04] <mdz_> mjg59: pictograms
[11:04] <ompaul> sabdfl, you are talking about people who may not know anything other than the CD that came with some windows magazine .. so you need to make it verbose - perhaps a stalling of the install that has someone say System Administration Video Troubleshooter
[11:04] <ompaul> sabdfl, s/say/Type - so they know where to find it on the menu
[11:04] <sabdfl> calc: odd, i just noticed that the laptop fan turned on as soon as i enabled effects
[11:05] <sabdfl> mdz: what, we don't have gnash desktop backgrounds working yet?
[11:05] <mdz_> mvo: would it be possible to display a notification on first login *only* in cases where compiz is running by default?
[11:05] <sabdfl> moving pictograms!
[11:05] <mvo> mdz_: that should be possible
[11:06] <calc> perhaps tie compiz into the laptop-mode stuff so it is turned off when not on AC power?
[11:06] <mjg59> calc: No.
[11:06] <mdz_> how about displaying a note to the user explaining that the wobbly effects are a technology preview, and explaining how to turn them off?
[11:06] <mjg59> "I unplug AC and all my windows rearrange themselves"
[11:06] <sabdfl> that would give a very strange user experience, calc
[11:06] <mdz_> "we believe they work well for a majority of cases, but there are known problems with X and Y"
[11:06] <calc> mjg59: hmm, yea that would suck
[11:07] <mvo> mdz_: sure, we could do set, we may do the opposite too, if we believe that desktop effects would work on the system, display a notification with a easy option to turn them on
[11:07] <mvo> s/set/that/
[11:07] <mdz_> mvo: ooohhh
[11:07] <mdz_> could it be a one-click option from the notification?
[11:08] <mvo> mdz_: yes, can add actions
[11:08] <mvo> mdz_: buttons to notifications
[11:08] <mdz_> that's an interesting idea
[11:08] <sabdfl> better for that notification to bring up the appearance dialog
[11:08] <calc> sabdfl: yes, especially since its not just a 3d on/off but completely different WM :\
[11:08] <sabdfl> so people know that's where to find it
[11:08] <sabdfl> folks, i have an early start tomorrow
[11:08] <mdz_> this is a critical issue for 7.10
[11:08] <sabdfl> mjg59: thanks very much for raising this, and the analysis
[11:09] <sabdfl> i'll abstain in a vote for now
[11:09] <mdz_> we don't need to lock ourselves in, but we need to pay attention to it, and soon
[11:09] <sabdfl> mdz, could you send me a log of remaining conversation?
[11:09] <sabdfl> i'm inclined to say "let's take a small risk on it"
[11:09] <mdz_> if you're leaving, then it's only mjg59 and me; mjg59 is against it and I'd abstain at this point
[11:09] <sabdfl> but not enough to vote
[11:10] <sabdfl> mjg59: would you be swayed by large numbers of "please do it we know it has issues but it's great"?
[11:10] <sabdfl> comments?
[11:10] <mdz_> I'd like to pursue ways to communicate better with the user about this feature
[11:10] <sabdfl> by people who don't hack on that code but who are strong ubuntu users?
[11:10] <mjg59> sabdfl: I think it depends which markets we're worried about here
[11:11] <mdz_> early adopters
[11:11] <sabdfl> this will depend on who you ask :-)
[11:11] <mdz_> and complete novices
[11:11] <sabdfl> myself, i'd say developers, early adopters, and media
[11:11] <Keybuk> (and I've just caught up on log, sorry for the dropping off)
[11:11] <sabdfl> howdy scott
[11:12] <sabdfl> i think novices are also eye  candy aware
[11:12] <sabdfl> though perhaps less tolerant of a total failure like X doesn't start, they don't mind some screen issues
[11:12] <mdz_> they are also less likely to trip on 3D use cases
[11:12] <sabdfl> all of these folks love to show their linux machine doing something that windows cannot
[11:13] <mjg59> My concern is that this sends out the message that we're more worried about bling than we are about working software
[11:13] <sabdfl> mdz_: +1
[11:13] <sabdfl> mjg59: working? will either crash?
[11:13] <mdz_> mjg59: perhaps some of us are *cough*sbdf*cough* ;-)
[11:13] <mjg59> sabdfl: 3D stuff, visual artifacts
[11:13] <mjg59> I haven't been using compiz
[11:14] <mjg59> (For this sort of reason)
[11:14] <sabdfl> sbdf?
[11:14] <Keybuk> mjg59: lies, I've seen drop-shadows on your windows at Uncle Steve's before <g>
[11:14] <mjg59> Keybuk: Yeah, that lasted about a day
[11:14] <Keybuk> really?  what made you turn it off?
[11:14] <mdz_> sabdfl: ??
[11:15] <sabdfl> oh dear
[11:15] <sabdfl> cats
[11:15] <sabdfl> pigeons
[11:15] <mjg59> Ok. I think I'll leave my argument at this:
[11:15] <sabdfl> first, i should say my compiz and beryl settings were so far from default i had weirdly wobbly windows
[11:15] <mjg59> Ubuntu has a reputation for Just Working
[11:15] <mjg59> Compiz does not Just Work
[11:16] <mdz_> sabdfl represents at once the uber-adopter who wants the latest bits at all costs, and the perfectionist intolerant of all faults :-)
[11:17] <mdz_> mjg59: it Mostly Works
[11:17] <sabdfl> ok. ok. mvo showed me a way to reset to defaults, will leave it switched on for now and reserve judgement
[11:17] <sabdfl> we do have a reputation for both "cutting edge cool" and "just works"
[11:17] <sabdfl> and that's what we're trading here
[11:17] <mvo> sabdfl: compizconfig-settings-manager has a global reset button too (since today :)
[11:17] <sabdfl> bugger. ctrl-shift-left is not working now. could that be compiz? to select word left
[11:18] <sabdfl> thanks mvo :-)
[11:18] <mdz_> sabdfl: that duality spells trouble
[11:19] <sabdfl> mdz_: also, opportunity :-)
[11:19] <sabdfl> mjg59: thanks again for raising this now, it's in the nick of time and totally appropriate for TB
[11:19] <mdz_> I think there are valid arguments both ways on this issue
[11:19] <mdz_> I don't feel comfortable taking a decision just yet
[11:19] <mdz_> I'd like to throw it open to a larger crowd
[11:20] <mdz_> at least ubuntu-devel-discussish
[11:21] <sabdfl> can we start with mjg59's TB email to u-d-d?
[11:21] <mjg59> Sure
[11:21] <mdz_> not verbatim, but with annotations from this discussion
[11:22] <mdz_> mvo: do you think you could launch that discussion?
[11:22] <mvo> mdz_: sure, I can write a mail about this tomorrow
[11:23] <mdz_> mvo: ok, I've forwarded you the original mail from mjg59 to tech-board
[11:23] <mvo> mdz_: ok, I will use that and the discussion from today
[11:23] <mdz_> [ACTION]  mvo to start more public discussion with summary of the issues
[11:23] <MootBot> ACTION received:  mvo to start more public discussion with summary of the issues
[11:24] <mdz_> I think it would be wise to defer the automatix discussion to a later date
[11:24] <mjg59> Ok
[11:24] <mdz_> the short version from me is that we should continue to not ship automatix, and do an updated analysis of the use cases it tries to meet post-7.10
[11:25] <mdz_> and see if we can improve them with the existing tools
[11:25] <mdz_> but anyway
[11:25] <mdz_> #endmeeting
[11:25] <MootBot> Meeting finished at 21:18.
[11:25] <sabdfl> i would like us to have a happier relationship with automatixz
[11:25] <mdz_> thanks, everyone, for hanging around this long
[11:26] <sabdfl> but i think that needs to be based on helping them to get the results they are looking for in a way consistent with debian and ubuntu policy
[11:26] <sabdfl> and i haven't been encouraged by their response there
[11:26] <sabdfl> ok, another time
[11:26] <sabdfl> night all!
[11:33] <nealmcb> the next community council meeting was just scheduled, but the time is contradictory - the visible time is 1100 UTC, but the link is for 1300 UTC
[11:40] <nealmcb> I wrote to mikeb- about it...