--- Log opened Sun Jun 22 21:41:47 2008
--- Day changed Mon Jun 23 2008
07:15 <linville> disregard my request for meeting info :-)
08:36 <@grover> oh well email's free
08:44 -!- grover changed the topic of #lpc-tpc to: wiki url http://plumbers.linuxplumbersconf.org/doku.php?id=abstracts:review | meeting 10am
08:50 <@keithp> grover: -EPERM
09:47 <@grover> one moment
09:48 <@grover> -EAGAIN
09:53 <@keithp> grover: 0
09:53 <@grover> yay!
09:54 <@grover> :)
10:00 * willy 's clock says it's time to start
10:00 <@grover> 60sec
10:01 <@grover> ok!
10:01 <kees> (for all the people that are on the TPC and can't access the URL in the topic, please let me know what your wiki login is, and I can add you to the group)
10:01 <@grover> hi all thanks for attending
10:01 <wagle> mine works
10:02 <kees> wagle: yeah, I think I mapped most people's logins correctly, but I thought I'd mention it anyway
10:02 <@grover> ok so basically this meeting should be pretty quick
10:02 <djwong> yay :)
10:02 -!- Topic for #lpc-tpc: wiki url http://plumbers.linuxplumbersconf.org/doku.php?id=abstracts:review | meeting 10am
10:02 -!- Topic set by grover [~grover@69.55.236.191] [Mon Jun 23 08:44:37 2008]
10:03 -!- grover changed the topic of #lpc-tpc to: wiki url http://plumbers.linuxplumbersconf.org/doku.php?id=abstracts:review |
10:03 -!- grover changed the topic of #lpc-tpc to: wiki url http://plumbers.linuxplumbersconf.org/doku.php?id=abstracts:review | Agenda: Introductions
10:03 <@grover> I'm sure a lot of you know each other but maybe everyone could introduce themselves and also add when you'd be free for future mtgs
10:04 <linville> intro: John Linville, kernel maintainer for wireless lan, red hat employee, east coast USA time zone
10:04 <@keithp> intro: Keith Packard, Xorg maintainer, Intel, PST8PDT
10:04 <rdd> randy dunlap, oracle, oregon/pacific time.
10:05 <willy> intro: Matthew Wilcox, kernel hacker for Intel, east coast Canada ;-)
10:05 <davidz> intro: David Zeuthen, works on fd.o and GNOME stuff, works for Red Hat, EST time zone
10:05 <rostedt> intro: Steven Rostedt, Linux RT maintainer and author of ftrace (with Ingo Molnar), Red Hat employee
10:05 <gregkh> intro: greg kroah-hartman, novell, kernel driver monkey, PST
10:05 <jejb> intro: James Bottomley, SCSI subsystem maintainer, just an observer
10:05 <lennart> intro: Lennart Poettering, doing all kinds of audio related stuff including PulseAudio, red hat employee, berlin/germany, CEST
10:05 <djwong> djwong: Darrick J. Wong, jack of too many trades, most recently working on hardware sensors (power meters and general environmental stuff) and libsas/aic94xx, ibm, oregon/pacific time
10:05 <kees> intro: Kees (pronounced "Case") Cook. kernel.org admin, Ubuntu security dude (Canonical employee), US Pacific TZ, plumbers wiki admin, general code monkey
10:05 <philips> Brandon Philips, Kernel dev at Novell/SuSE, organizer for LPC, Pacific time
10:05 <wagle> intro: Perry Wagle, monta vista, pacific
10:05 <kristen> intro: Kristen Accardi, Kernel dev at Intel, conference organizer, Pacific time.
10:06 <@grover> intro: Andy Grover, Oracle, worked on pm and net stuff, tpc chair, PST
10:06 * rostedt forgot to say EDT
10:07 <@grover> great, thanks everyone! I guess we can discuss next meeting time at the end
10:07 <@grover> btw I am logging this and will post it after the mtg
10:08 <@grover> just to possibly state the obvious we have people with two different roles here
10:08 <@grover> 1) Microconf Runners are in charge of a 2.5 hr block, and get to decide talks, organization, pretty much everything about that microconf
10:09 <@grover> 2) TPC members are here to help MC Runners review talk abstracts, and we additionally are responsible for defining the content of the conference's Track 3
10:10 <@grover> I'm hoping everyone is clear which case they fall into :)
10:10 <djwong> are MC runners allowed to help out with track 3, or are they supposed to manage their own MC and stay out of #3?
10:11 <@grover> absolutely, but #1 focus is your microconf
10:11 <rdd> uh, what is track 3?
10:11 <willy> track 3 is the papers which aren't in a microconf
10:12 <rdd> aka General?
10:12 <willy> yes
10:12 <rdd> k, thnx
10:12 <@grover> see http://plumbers.linuxplumbersconf.org/doku.php?id=program_planning Other Sessions, too (not just talks necessarily)
10:13 -!- grover changed the topic of #lpc-tpc to: wiki url http://plumbers.linuxplumbersconf.org/doku.php?id=abstracts:review | Agenda: Wiki use for reviewing submissions
10:13 <@grover> can everyone access that url?
10:14 <wagle> yup
10:14 <djwong> i can
10:14 <rdd> yes
10:14 * davidz too
10:14 <rostedt> yep
10:14 <linville> yes
10:14 * rostedt also prefers puppies over bunnies
10:14 <@grover> can anyone not? :)
10:14 <@grover> ok great
10:14 <lennart> grover, i cannot
10:14 <kees> grover: quick question, are all the MC and TPC folks here for this meeting? (is this all of us?)
10:14 <kees> lennart: what is your wiki login?
10:14 <lennart> grover, username is "lennart"
10:15 <@grover> thx kees
10:15 <djwong> jamey sharp doesn't seem to be here
10:15 <kees> lennart: okay, try now
10:15 <willy> he and sarah are on vacation
10:15 <kristen> kees: some are missing. ah - there's the one I was thinking of :)
10:15 <lennart> kees, thanks, works fine now
10:15 <jejb> Mauro is missing too
10:15 <@grover> we have ~18 people so I guess a few absences each mtg is inevitable
10:16 <gregkh> mauro has a "real" job and probably can't be on irc right now :)
10:16 <@grover> hehe
10:16 <wagle> no Bart, but he said he was going to be late or something
10:17 <@grover> btw in addition to this channel there is also #linux-plumbers, which always has lpc people idling in it, if needs be
10:17 <wagle> everyone else on the tpc list on the wiki is here
10:17 <@grover> everything we do should also end up on the wiki
10:18 <wagle> well, except those mentioned about
10:18 <wagle> above
10:18 <@grover> has everyone had a chance to read over the directions on that page? any questions?
10:18 <mjg59> grover: I'm missing from the acl, I think
10:19 * grover watches kees dart into action
10:19 <kees> mjg59: okay, try again
10:20 <kees> wagle: which wiki page were you getting the list from?
10:20 <mjg59> kees: Got it
10:20 <wagle> kees: http://plumbers.linuxplumbersconf.org/doku.php?id=program_planning
10:21 <@grover> I've put the submissions we've received so far up, as you can see
10:21 <kees> wagle: thanks, I've linked to it from the review page
10:21 <@grover> there are a couple of questions we need to agree on
10:22 <wagle> hmm.. i missed hemminger, dont see him
10:22 <kees> grover: the TPC list in program_planning looks incomplete -- no MC runners listed. is there a more canonical list?
10:22 <wagle> yeah, was wondering about that
10:22 <@grover> the mc runners are listed on the microconfs page
10:23 <kees> ("more" canonical? .... "a little dead")
10:23 <willy> closer to canonical? ;-)
10:23 <kees> grover: how does "track 3" map to the microconf topic list? 3rd in the list?
10:24 <wagle> only one runner per microconf?
10:24 <@grover> kees: don't understand the q
10:24 <willy> kees: Microconfs are tracks 1 and 2.
10:24 <@grover> wagle: correct
10:24 <kees> willy: oh! right
10:24 <kees> track 3 == non-MC
10:25 <kristen> maybe they should be relabeled to be more clear.
10:25 <@grover> correct. please feel free to make things clearer, it's a wiki!
10:25 <kees> (urgh, I can read -- grover added the "how to use this matrix" since I last looked heh)
10:25 <willy> true, but it's also an organisational question ;-)
10:25 <davidz> would it make sense to have track 3, it being of more general interest and all, not collide with track 4?
10:25 <@grover> track...4?
10:26 <wagle> grover: does that mean only one reviewer for microconf papers?
10:26 <davidz> grover: oh, sorry... I think I got it wrong.. I thought track 3 == microconf 3
10:26 <@grover> wagle: no we should review all papers, but final decision is up to MC runner
10:27 <wagle> grover: /me gets it now
10:27 <@grover> any more questions before the questions I had?
10:27 <kees> seems like it's a good idea for reviewers to detail their opinion about a given abstract then, so that MC has something to read from the other TPC folks beyond just "+" and "-"
10:27 <kristen> I have a comment on this line in the program_planning page:
10:27 <kristen> “No topic block ends until it produces some working code”?
10:28 <kristen> I was wondering if you were going to cover more the goals of the MC?
10:28 <kristen> I sort of disagree that you should require working code :)
10:28 <@grover> sure can
10:29 <@grover> don't treat the wiki text as canonical, early on it was just where ideas were gathered
10:29 <kristen> ok - thanks for the clarification.
10:29 <@grover> ok so microconfs.
10:30 <@grover> the idea was that this is a working conference, where real, tactical issues get solved faster by having people in the same place
10:31 <@grover> there is progress made by gathering stakeholders in each area together, as well as better cross-area collaboration by having the chance to meet up with people interested in other areas
10:32 <@grover> it's ok to not save the world in each microconf, but it's a goal to make some progress -- either code, or come to consensus on a design decision, or something.
10:32 <kristen> for example - bringing together 2 separate projects that might need to work more closely with each other to have a discussion about a system wide problem to solve.
10:33 <@grover> pushing out the responsibility for inviting people and each mc program to someone involved in the area is meant to make sure the mc is best suited to the given area, and all the right people are there (because honestly we could guess, but we'd be wrong!)
10:35 <@grover> kind of a different way of structuring things, hopefully it will work out well
10:36 <@grover> so one thing that it appears is not being listed in peoples' submissions is if they are targeting a given MC or not
10:36 <@grover> it's not always obvious, at least to me
10:37 <kees> perhaps move that bullet item to the top of the "abstract should include these things:" list on the CFP page?
10:37 <wagle> what if they (accidentily or on purpose) "target" more than one?
10:37 <@grover> so we may need to review them to decide, and then tag them approriately
10:37 <kees> MC fight! ;)
10:37 <rostedt> the miniconfs are 2 1/2 hours, so does this mean that we just divide the submissions by that time and that is how long each submitter has?
10:38 <djwong> i think we were targeting 2-3 25min talks and then discussions after the talks?
10:38 <djwong> or at least that's what _i_ was intending to go for :)
10:38 <@grover> rostedt: it's up to you how much time to give each presenter, and to devote to open discussion, or a panel talk, or a code sprint, or a presentation by you
10:38 <willy> rostedt: There's a suggested schedule, but it's up to you
10:38 <rostedt> thanks
10:38 <kristen> while we don't want to tell people how to run their MC - I think if you want to produce some kind of output from your MC, having more than just talks would be good.
10:38 <rdd> but there could be some travel budget constraints if they need travel help
10:38 * rostedt just hopes the people he asks actually shows up
10:39 <kristen> rostedt: and other MC runners - you will have to nag people.
10:39 <mjg59> Is there a number that we'd expect per miniconf?
10:39 <@grover> kees: we should add that to the CFP but some will invariably omit it anyways
10:39 <@grover> mjg59: number of speakers?
10:39 <kees> grover: sure, but an extra prompt can be handy
10:39 <mjg59> grover: Attendees
10:40 <mjg59> Or is it more a matter of "as many as needed"?
10:40 <kristen> mjg59: the conference is limited to 350 people.
10:40 <wagle> you going to have whiteboards, or big paper pads, or somesuch?
10:40 <kristen> mjg59: the rooms I believe can seat 100-150
10:40 <@grover> mjg59: the total conference size is 350, but are there really ~35 key stakeholders in each mc area? I'd be happy if each mc gets 10 key people and then a bunch of random other interested people
10:41 <kristen> we have much smaller rooms available to have smaller more focused groups meet.
10:41 <@grover> there will be stuff so work can get done, as well as a bunch of other rooms to escape to
10:41 <@grover> either whiteboards or pads, not sure
10:41 <kristen> you could make your MC divide into groups if you want to have smaller groups discussing things too.
10:43 <@grover> so what I propose is that for submissions not targeted to a single MC, I'll just pick one, and then that'll be a starting point for discussion, either on the wiki or on the mailing list
10:43 <@grover> and we'll update the cfp webpage so hopefully we'll know what mc people had in mind
10:43 <@grover> I'll contact the submissions we have so far and ask them, as well
10:44 <wagle> i cant remember being happy with the results of being divided into 3-4 person groups
10:44 <@grover> does that sound ok to everyone?
10:45 <@grover> cool
10:45 <wagle> sounds ok
10:45 <linville> seems reasonable
10:46 <wagle> so alal the papers will be accessible (on-line) to everyone (on the committee)?
10:46 <@grover> next question: should comments default to public (i.e. the submitter may see as feedback) or private (i.e. no one but tpc sees)? We can have a scheme to mark comments pub or priv, so the real question is what should the *default* be
10:46 <wagle> all
10:47 <@grover> wagle: I was going to put all papers on the wiki page, unless there's a reason not to?
10:47 <wagle> default private, but people should be pressured to give out detailed criticism
10:48 <kristen> my experience on other Paper Review committees was that it was easier for me to make comments if I knew they were private by default.
10:48 <@grover> what has been standard on other review committees?
10:48 <kristen> for OSCON at least they are private.
10:48 <rostedt> OLS posts just the abstract of papers
10:48 <wagle> grover, no reason not to.. was imagining different microconfs vying for the same paper maybe.. (?)
10:48 <rostedt> the final papers are distributed after the conf
10:48 <jejb> most review systems that give public feedback have special places for private and public comments ... I suggest you just default to private since you don't have such a complex system
10:48 <kristen> I think when I did summer of code reviews they were private too.
10:49 <willy> OLS keeps comments private ... unless ajh feels like giving them out.
10:49 <@grover> ok so we'll go with what's documented on the wiki, all comments are private unless prefixed with a "pub:" tag
10:49 <wagle> silly flames really dont need to be passed on
10:49 <rostedt> I think submitters may also feel more comfortable with having the comments private
10:49 <@keithp> grover: we should encourage people to make public comments that can be passed back to the author
10:50 <jejb> most major conferences give public feedback ... but I think for the first plumbers, just follow the OLS everything's private rules
10:50 <willy> reviewers shouldn't be writing silly flames.
10:50 <kristen> providing constructive feedback would be great.
10:50 <willy> when I was reviewing papers for OLS, I was appalled by some of the comments other reviewers were leaving.
10:50 <jejb> unless you have an automated system, it becomes quite a chore doing it
10:50 <jejb> I'd suggest stealing someone else's review system for next time ...
10:51 * rostedt never had a problem with comments from OLS reviewers
10:51 * rostedt is also use to being called an ass ;-)
10:51 <@grover> jejb: yes next time we will get fancy
10:51 <kristen> jejb: yes, we will definitely examine our system and see if we can make improvements for next year.
10:51 <kristen> so any feedback on the review process itself is much appreciated.
10:52 <@grover> any more review/wiki/feedback questions or comments?
10:53 <@grover> ok two more things to cover 1) assignments for next time 2) when's next time gonna be
10:53 <wagle> someone said we already have submissions?
10:53 <@grover> a) Andy will contact submitters so far and get their targeted mc, if any
10:54 <jejb> Actually, could I suggest that MC runners might like to invite people in their area to make submissions with a nice personalised email
10:54 <@grover> wagle: they're on that wiki page
10:54 <jejb> just a few each
10:54 <willy> Dear James. Please come to my Microconf. Love, willy.
10:54 <@grover> b) everyone should review current submissions (there are 5) and solicit more submissions
10:54 <@grover> lol
10:55 <rdd> s/come to/be a part of/
10:55 * mjg59 gets confused about alternations of numerically and alphabetically ordered lists again
10:55 <wagle> this one?: http://plumbers.linuxplumbersconf.org/doku.php?id=abstracts:review
10:55 <@grover> wagle: yup
10:55 <wagle> the example, followed by "Bunnies are Cute" threw me off.. 8)
10:56 <@grover> all but the first one are real
10:56 <wagle> cool
10:56 <@grover> are there any more tasks we need to do before next meeting?
10:57 <jejb> willy, Something like Dear James, I'm organising the Future Storage session at the Linux Plumbers conference, and as one of the experts in the field, I'd really appreciate it if you could find the time to put together a submission (CFP url is here). Thank you so much for doing this ...
10:57 <@grover> how could you refuse???
10:58 <@grover> ok, when's the next meeting gonna be?
10:58 <jejb> It's important that each MC runner do this otherwise you're relying on pot luck for the submissions pool ...
10:58 <davidz> jejb: yeah
10:59 <@grover> yeah mc runners, we're really relying on you guys for your mc's level of health
10:59 <kees> grover: 1 week? 2 weeks?
10:59 <@grover> 1 week I was thinking
10:59 <philips> 1 week, same time/place?
10:59 <wagle> good meeting time would be somewhere around the time the abstracts are due, or something else requiring some thought
10:59 <rostedt> I take it it would be bad to just spam a bunch of people in the field and say "Hi All, I'm running the miniconf in tracing/debugging and since you are all involved, please send in some submissions! -- Steve"
10:59 <@grover> any time where we might get better attendance?
11:00 <rdd> better??
11:00 <philips> grover: email those who couldn't attend and find out what times work for them- then alternate meeting times?
11:00 <wagle> how many people are missing from this meeting?
11:00 <@grover> rdd: we're missing ~4 people, incl Bart
11:00 <jejb> rostedt, experience shows that personal email produces the best results
11:01 <rostedt> jejb: I'm sure. as I said, it would be bad to do that ;-)
11:01 <@grover> ok I'll contact them and see what time is good. watch your email, there may be some scheduling back-and-forth, so prepare
11:01 <rdd> grover: yes, i thought that this was Good attendance
11:01 <rdd> or even Better ;)
11:02 <@grover> ok that's all I had to talk about. Thanks for your time everyone! I'm very excited about this process!
11:02 <davidz> cool
11:02 <kees> \o/
11:04 -!- grover changed the topic of #lpc-tpc to: meeting over, thanks!
--- Day changed Tue Jun 24 2008
13:18 <@grover> ok gonna go whip up that irc log
--- Log closed Tue Jun 24 13:19:07 2008
