fedora_base_design_working_group
LOGS
15:00:59 <pknirsch> #startmeeting Fedora Base Design Working Group (2013-11-08)
15:00:59 <zodbot> Meeting started Fri Nov  8 15:00:59 2013 UTC.  The chair is pknirsch. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:59 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:01:23 <pknirsch> #meetingname  Fedora Base Design Working Group
15:01:23 <zodbot> The meeting name has been set to 'fedora_base_design_working_group'
15:01:49 <pknirsch> Hello and welcome everyone!
15:01:56 <jreznik> hey pknirsch!
15:01:59 <haraldh> hey
15:02:11 <pknirsch> lets see if we have everyone here. Is there a way to check that with zodbot?
15:02:24 * pknirsch wonders if the #chair command would help
15:02:55 <pknirsch> #commands
15:02:55 <zodbot> Available commands: #accept #accepted #action #agree #agreed #chair #commands #endmeeting #halp #help #idea #info #link #lurk #meetingname #meetingtopic #nick #rejected #restrictlogs #save #startmeeting #topic #unchair #undo #unlurk
15:03:12 <pknirsch> ah well, lets do a roll call and i'll #info it.
15:03:13 <jreznik> well, at least for the first meeting I'd say some sort of overview of who's in base groups is not a bad idea, so it could help identifyin who's here or no :)
15:03:23 <pknirsch> absolutely
15:03:34 <pknirsch> so i already see jreznik and haraldh
15:03:37 <jreznik> if you chair me, I can help taking notes :)
15:04:10 <pknirsch> #chair jreznik
15:04:10 <zodbot> Current chairs: jreznik pknirsch
15:04:13 <haraldh> #chair pknirsch jreznik haraldh
15:04:28 <pknirsch> #chair haraldh
15:04:28 <zodbot> Current chairs: haraldh jreznik pknirsch
15:06:15 <pknirsch> alright, who else we got? jwb, dgilmore, notting, dwalsh, jon disnard, subhendu gosh?
15:06:49 <haraldh> everybody seems to be asleep :)
15:06:50 <jwb> yep, here.  sorry
15:06:57 <pknirsch> np, hi jwb :)
15:07:02 <pknirsch> #chair jwb
15:07:02 <zodbot> Current chairs: haraldh jreznik jwb pknirsch
15:07:07 <jreznik> I don't see jon online
15:07:25 <jreznik> haraldh: I already had one red bull today :)
15:07:38 <pknirsch> 4th diet coke today :)
15:07:44 <haraldh> tss
15:08:07 <haraldh> you guys are obviously awake :)
15:08:19 <pknirsch> :)
15:08:49 <pknirsch> hey Jon :)
15:08:57 <masta> hey there
15:09:02 <pknirsch> #chair masta
15:09:02 <zodbot> Current chairs: haraldh jreznik jwb masta pknirsch
15:09:02 <jreznik> hey masta!
15:09:24 * dgilmore is here
15:09:49 <pknirsch> #chair dgilmore
15:09:49 <zodbot> Current chairs: dgilmore haraldh jreznik jwb masta pknirsch
15:10:15 <pknirsch> alright, i think we can get going then, just to avoid the EU folks to have to spend the whole evening on IRC
15:10:24 <jreznik> pknirsch: +1 :)
15:10:31 <pknirsch> ;)
15:10:55 <masta> ok
15:11:36 <pknirsch> So first up, topics for today can be found here: https://lists.fedoraproject.org/pipermail/devel/2013-November/191316.html
15:11:58 <pknirsch> The first task from FESCO for each WG was the draft for governance
15:12:13 <pknirsch> #topic Governance discussion & draft
15:13:02 <pknirsch> As i mentioned i  the agenda, my proposal would be to look at some of the other WGs governance proposals and either pick one of them or use several of them and combine them to make one for us
15:14:02 <jwb> so far i think they're all going with "the existing members pick a person if a seat becomes available"
15:14:09 * jreznik would prefer to be closer to other WGs
15:14:32 <jreznik> jwb: seems like a good compromise for me
15:14:54 <pknirsch> jwb: any objections to that governance model for Base?
15:15:03 <jwb> not from me
15:15:16 <pknirsch> ok, anyone else thinks we should do something else?
15:15:20 <haraldh> and when does a seat become available?
15:15:23 <pknirsch> and if so, what? :)
15:15:37 <jwb> haraldh, whenever someone steps down
15:15:42 <jreznik> haraldh: depends, people can step down
15:15:52 <haraldh> no overall time limits?
15:16:09 <haraldh> lifetime seats?
15:16:09 <jreznik> and in case somebody would not be very active over time, I think the rest of group can talk to him
15:16:17 <jwb> no.  i asked if that was something we wanted in Workstation and the consensus seemed to be no term limits
15:16:30 <jwb> server and fesco approved without term limits
15:16:52 <pknirsch> hm, and imho if someone is inactive and unreachable the remaining 8 could then vote to release him of his duties in his absence and replace him?
15:17:02 <jwb> https://fedoraproject.org/wiki/Server/Governance_Charter
15:17:12 <jwb> that's the fesco approved server governance
15:17:19 <jwb> members just confirm every 6mo
15:17:23 <masta> thanks jwb
15:17:31 <pknirsch> ah good
15:17:39 <dgilmore> pknirsch: i think that timelimits are silly
15:17:42 <pknirsch> so if someone doesn't confirm the seat automatically becomes available
15:18:03 <pknirsch> dgilmore: aye, not arguing in favor of timelimits
15:18:03 <dgilmore> we could have re-evaluate check points
15:18:17 * pknirsch likes the 6 month rule for recheck
15:18:38 <dgilmore> if people get to the point they are not effective or they need to step down, i believe they will
15:18:45 <pknirsch> true
15:19:10 <jwb> the thing to note here: WGs are under FESCo jurisdiction.  if there becomes problems, they can address them
15:19:36 <haraldh> https://fedoraproject.org/wiki/Server/Governance_Charter looks good :)
15:19:37 <pknirsch> Proposal: Take the server governance charter and replace the names
15:19:40 <sgallagh> Right, ultimately FESCo can disband or reform the group if such a thing became necessary
15:19:58 <jwb> pknirsch, sounds fine to me
15:20:05 <haraldh> #agreed
15:20:25 <pknirsch> +1 from me obviously ;)
15:20:44 <dgilmore> +1 from me
15:20:49 <haraldh> so, if nobody objects, do it :)
15:21:16 <masta> seems fine
15:21:24 <masta> read it over, so +1
15:22:03 <jreznik> +1, all seems similar
15:22:11 <pknirsch> #agreed Proposal: Take the server governance charter and replace the names and present to FESCO for approval
15:22:38 <pknirsch> Any volunteers to copy over the wiki and put in our names?
15:23:05 <jwb> should probably create a landing page too
15:23:06 <haraldh> <-
15:23:19 <jwb> like fedoraproject.org/wiki/Base
15:23:36 <pknirsch> #action haraldh  Proposal: Take the server governance charter and replace the names and present to FESCO for approval
15:23:51 <pknirsch> jwb: true, good point
15:24:01 <dgilmore> jwb: yeah we should put everything under /Base
15:24:12 <masta> yes /Base
15:24:23 <pknirsch> quick and easy to type and remember
15:25:00 <pknirsch> so charter would be fedoraproject.org/wiki/Base/Governance_Charter
15:25:07 <jwb> right
15:25:13 <pknirsch> +1 :)
15:25:36 * jreznik can volunteer to prepare it (copy/paste)
15:25:59 <pknirsch> jreznik, haraldh: feel free to work on that together then :)
15:26:26 * haraldh is currently editing /Base
15:26:28 <pknirsch> Ok, anything else on governance we need to cover?
15:26:39 <jreznik> there's not much work for too people so I let haraldh but I think the whole group will help there
15:26:51 <pknirsch> aye
15:27:14 <pknirsch> i just want to avoid "someone" will do it scenarios where it someone == noone ;)
15:27:46 <dgilmore> pknirsch: :)
15:27:52 <pknirsch> alright, lets move on to the next topic.
15:28:03 <pknirsch> #topic Regular meeting times
15:28:54 <dgilmore> for the most this time works well for me
15:28:58 <pknirsch> any objections to keeping the meeting at this time/day each week? or any counter proposals based on the availability from whenisgood.net
15:29:07 <dgilmore> december ill be tricky, but ill get over it
15:29:11 <pknirsch> mhm
15:29:12 <dgilmore> will be
15:29:47 <jreznik> it's ok for me too
15:29:54 <pknirsch> dgilmore: i'll try to send out agendas 1 day prior to the meeting to you so you can already comment on them in case you won't be able to make it
15:30:02 <masta> fine for now
15:30:18 <dgilmore> pknirsch: thanks, ill be in Australia Nov 24-Dec 27
15:30:31 <dgilmore> and the start time is 1am there
15:30:32 * pknirsch makes a note of the date
15:30:37 <jwb> fine with me
15:30:40 <pknirsch> ye, thats harsh :)
15:30:40 <dgilmore> but i will do my best to attend
15:31:14 <jreznik> stacks guys has two meeting time - to make it easier for people in different timezones
15:31:32 <jreznik> once it's early/late for one group, next time for another
15:31:49 <pknirsch> how do they coordinate that then? do they have a chair thats on both?
15:31:52 <jreznik> but I'm not sure how much it makes sense for one month exception
15:32:06 <dgilmore> lets not make an exception
15:32:37 <pknirsch> ok, lets keep it at this time/day for now then.
15:32:49 <masta> +1
15:32:58 <jreznik> ok
15:33:00 <pknirsch> #info Keep meeting time at 15:00 UTC for the time being
15:33:19 <pknirsch> next topic:
15:33:28 <pknirsch> #topic Base mailing list discussion
15:33:38 <dgilmore> pknirsch: one question, will we change with daylight savings?
15:34:01 <pknirsch> dgilmore: Ah right, thats something jwb mentioned as well. We probably should, yes
15:34:13 <pknirsch> dgilmore: question is, which ones? ;)
15:34:16 <dgilmore> pknirsch: and which daylight savings?
15:34:17 <pknirsch> EU or US ;)
15:34:18 <jreznik> dgilmore: we now have half year, once it will come, we can think about it
15:34:19 <pknirsch> hahaha
15:34:26 <pknirsch> yea
15:34:36 <dgilmore> lets deal with it later then
15:34:38 <dgilmore> carry on
15:34:45 <pknirsch> ok, mailing list
15:34:47 <jreznik> in half year, schedule for many people can change, so...
15:34:52 <pknirsch> i don't have any preference really
15:34:58 <pknirsch> devel is great for exposure
15:35:04 <pknirsch> but stuff can drown there due to volume
15:35:27 <jwb> i'd kind of like to stick with devel
15:35:42 <dgilmore> pknirsch: im torn, in many cases i think devel is best, but the signal to noise ratio is high there
15:35:44 <pknirsch> ye, i have a slight preference for that as well
15:35:45 <jwb> it's the one place where most of the other group members should be subscribed already, and where FESCo stuff comes up
15:35:51 * pknirsch nods
15:36:07 <pknirsch> can we mark our subjects with the [Fedora Base WG] tag maybe?
15:36:15 <pknirsch> that at least helps me find the messages quickly
15:36:24 <jwb> sure
15:36:39 <dgilmore> pknirsch: [Base]
15:36:42 <pknirsch> and i agree, Base should really happen on devel
15:36:49 <pknirsch> dgilmore: ah, even better :)
15:36:52 <dgilmore> easier to type
15:36:55 <pknirsch> yea
15:37:06 <pknirsch> doesn't fill up the subject unnecessarily with clutter then
15:37:08 <pknirsch> +1
15:37:11 <haraldh> #agreed
15:37:28 <dgilmore> +1 on devel use [Base] in headers
15:37:32 <dgilmore> subject
15:37:46 <pknirsch> mhm
15:37:53 <pknirsch> masta, jreznik?
15:38:41 <masta> I've no objection to this
15:38:54 <jreznik> no objections, I prefer devel
15:38:58 <pknirsch> good,
15:39:33 <pknirsch> #agreed Stay with fedora-devel mailinglist, use [Base] in subject to indicate emails around Base Design discussions
15:39:33 <haraldh> jreznik, your take on https://fedoraproject.org/wiki/Base/Governance_Charter and https://fedoraproject.org/wiki/Base now :)
15:40:13 <pknirsch> thanks haraldh !
15:41:16 <jreznik> haraldh: what is still missing there? heh, dgilmore as plumber :) I'll continue with that, thanks
15:41:58 <haraldh> jreznik, oops.. corrected :)
15:42:51 <pknirsch> ok, now lets get to the probably biggest topic for today:
15:42:54 * dgilmore cleaned up his entry
15:42:57 <pknirsch> #topic Role of Base Design group discussion
15:43:14 * pknirsch needs a quick bathroom break, but please feel free to continue without me
15:43:42 <dgilmore> what is it e are responsible for? what and when will we deliver things?
15:44:04 <haraldh> hmm
15:44:11 <jwb> as a start, defining what package set Base means and what rules it should follow
15:44:26 <jwb> i think of this mostly as "critpath + some stuff"
15:44:34 <dgilmore> i kindof think of base are critpath
15:44:48 <dgilmore> and the tooling to compose and deliver releases
15:44:54 <haraldh> I think of it as minimal install at least
15:44:56 <jreznik> do we have a list of critpath?
15:45:01 <jwb> somewhere, yes
15:45:28 <jwb> but i think the "some stuff" should also be included.  e.g. things all (or most) of the other WGs want to see
15:45:28 <jreznik> as previously, a lot of stuff was dragged into the critpath that's not critpath at all
15:45:28 <jwb> s/see/use
15:45:31 <haraldh> minimal install for all installation options we support
15:45:34 <dgilmore> https://fedoraproject.org/wiki/Critical_path_package?rd=Critical_path_packages#Where_can_I_find_the_critical_path.3F
15:45:57 <dgilmore> https://admin.fedoraproject.org/pkgdb/lists/critpath?tg_format=plain&collctn_list=f20
15:46:06 <dgilmore> thats the f20 critpath list
15:46:13 <haraldh> PackageKit?
15:46:13 <jreznik> another thing is - I'd like to avoid being strictly package wise but aim more on functionality as it was proposed on the last fudcon by mitr
15:46:21 <haraldh> hmm, no.. not all critpath
15:46:31 <haraldh> cups?
15:46:32 <haraldh> no
15:46:42 <haraldh> emacs?
15:46:43 <haraldh> no
15:46:47 <haraldh> so, no for me
15:46:51 <jwb> haraldh, it's driven by dependencies pulled in around a defined set of pacakges
15:47:03 <haraldh> emacs is in a dep?
15:47:04 <jreznik> and also workstation guys are not strictly tied to packages but functionality provided (aka support wayland, pulseaudio)
15:47:12 <jwb> haraldh, likely somewhere, yes
15:47:19 <dgilmore> haraldh: https://fedoraproject.org/wiki/Update_Critpath_SOP lists how its generated
15:47:24 <jwb> jreznik, and containers
15:47:26 <jreznik> it could also help with situations such as - removal of sendmail
15:47:29 <jwb> wich server also want
15:47:36 <jwb> or at least want to look at
15:47:55 <jreznik> instead of insisting on sendmail as package we can define - mta should be present (in any form)
15:48:23 <dgilmore> jreznik: we should define the minimal fedora experience
15:48:52 <jreznik> dgilmore: yep, I know packages are easiest way but be a bit above package level would help a lot
15:49:03 <haraldh> I would say, the minimal required packages to get a system running on all installation options we support (like iSCSI...)
15:49:05 <dgilmore> the things yo can be sure will be there, regardless of if you're using cloud, server, worksation  or some other product
15:49:14 <jreznik> (and on the other hand I understand we can't define everything, so some compromise is needed)
15:49:35 <haraldh> system running == booting to a shell with an init system
15:49:55 <dgilmore> haraldh: i think we want more than that
15:50:02 <jreznik> yep
15:50:03 <haraldh> so, define "more"
15:50:15 <jreznik> aka with defined interfaces
15:50:22 <dgilmore> yum/dnf, rpm, editor
15:50:36 <haraldh> agreed
15:50:42 <dgilmore> possily a mta interface
15:50:50 <haraldh> hmm
15:50:53 <haraldh> really?
15:51:08 <dgilmore> probably not
15:51:10 <haraldh> wasn't mail killed by the NSA?
15:51:11 <haraldh> :)
15:51:13 <pknirsch> isn't mta already something a product should define?
15:51:14 <jwb> i think the more is "functionality all Products will use"
15:51:18 <jwb> and mta isn't that.
15:51:21 <pknirsch> ye
15:51:25 <dgilmore> workstation probably doesnt want it and server will
15:51:38 <pknirsch> but e.g. something env/stacks have asked me to bring up here is Containers
15:51:49 <pknirsch> or containerized applications
15:51:54 <jwb> right
15:51:55 <haraldh> dgilmore, but does a container need/want "yum/dnf, rpm" ?
15:51:57 <dgilmore> pknirsch: like SCL etc
15:52:07 <pknirsch> well, SCL would use it
15:52:13 <pknirsch> but base should provide the technology
15:52:14 <pknirsch> aka
15:52:16 <masta> since some of our consumers would not want an MTA, I would strive to also not have it in Base
15:52:17 <pknirsch> tech vs. content
15:52:17 <dgilmore> haraldh: yes
15:52:51 <haraldh> masta, agreed
15:53:23 <pknirsch> but we will need rpm and yum i suppose as the majority of our products will build on top of that
15:53:34 <haraldh> so, what do we have: support for all installations + init + rpm + yum/dnf + editor
15:53:45 <haraldh> filesystem check
15:53:52 <jwb> i don't think we're going to draw crisp and clear lines here today, but getting a general consensus on the overriding realm of what Base looks at would be good
15:53:53 <haraldh> filesystem tools
15:54:05 <dgilmore> haraldh: probably anacona and its deps
15:54:10 <dgilmore> anaconda
15:54:30 <haraldh> probably not
15:54:40 <haraldh> you can easily install systemd without anaconda :)
15:54:49 <haraldh> s/systemd/systems
15:54:53 <haraldh> damn freud :)
15:54:59 <jwb> ...
15:55:07 <jwb> systems or containers?
15:55:11 <haraldh> both
15:55:18 <dgilmore> haraldh: sure, but anaconda is the fedora installation tool, unless we plan to throw that out
15:55:29 <jwb> i don't think base is going to deviate from anaconda as the default installer
15:55:41 <dgilmore> i dont thing so either
15:55:51 <pknirsch> but is base itself installable?
15:56:00 <pknirsch> (just throwing out ideas)
15:56:03 <haraldh> so, do Products pick the entire Base or single packages from Base?
15:56:10 <jreznik> as minimal install it should be
15:56:22 <dgilmore> pknirsch: i think we need to provide the installer framework
15:56:27 <pknirsch> and if it is installable, it needs an installer
15:56:27 <jwb> i was thinking Base is the minimal set all products build on, not just individual packages
15:56:34 <pknirsch> which should be common for most products
15:56:38 <haraldh> Hmm, ok, in the term of "all installation options we support", adding anaconda makes sense
15:56:47 <pknirsch> but any derived product could still define their own installer
15:56:49 <pknirsch> right?
15:56:56 <jwb> yes, if they're crazy enough
15:57:00 <pknirsch> ye
15:57:01 <pknirsch> :)
15:57:01 <haraldh> right
15:57:03 <jreznik> or derived anaconda experience
15:57:22 <dgilmore> pknirsch: right
15:57:25 <jreznik> ala workstation does not need enterprise storage
15:57:33 <pknirsch> mhm
15:57:38 <haraldh> so, yes, Base should probably be self contained and installable
15:57:38 <pknirsch> so they could drop support for that
15:57:42 <jwb> so Base to me sounds like: installer, compose tools, minimal install (for some definition there), and functionality all products want to use
15:57:45 <jreznik> (it was discussed on devel)
15:57:56 <dgilmore> Server and Workstation would use different kickstarts to provide different install options
15:58:12 <pknirsch> mhm
15:58:25 <dgilmore> but thats a compose time thing
15:58:37 <pknirsch> and could pick a different set of components for their product, ye
15:58:40 <dgilmore> anaconda would be the same
15:58:45 <pknirsch> even with things missing from Base
15:59:08 <pknirsch> i like jwb's definition
15:59:16 <dgilmore> pknirsch: same
15:59:25 <haraldh> "functionality all products want to use"
15:59:28 <haraldh> which means?
15:59:31 <pknirsch> though i would replace "all products" with "the majority"
15:59:36 <jwb> sure, majority works
15:59:38 <pknirsch> of products
15:59:41 <haraldh> if one product does not want to use it, it is not base?
15:59:44 <dgilmore> if you did a minimal install of any product everything installed should be part of base
15:59:45 <haraldh> ok
15:59:50 <jwb> haraldh, like when they come to us and say "we want containers"
15:59:58 <pknirsch> as we don't want to restrict any product to go a different route
16:00:16 <jwb> it's worth pointing out that "minimal install" and "functionality the majority of products want to use" are not identical
16:00:19 <pknirsch> though at some point the question why they'd use Base then at all might come up ;)
16:00:23 <jwb> in that we can do the latter without including it in the former
16:00:27 <jreznik> so if majority - other products should be able to override some base design stuff
16:00:41 <dgilmore> jwb: right base can be more than a minimal install
16:00:42 <jwb> jreznik, if they do the work
16:00:47 <haraldh> so we have "Base" and "Common" :)
16:00:49 <masta> I'd like Base to stay small enough for embedded situation
16:01:01 <jwb> masta, you'd like "minimal install" to do that
16:01:03 <haraldh> masta, me, too
16:01:08 <dgilmore> masta: thats minimal install not base
16:01:20 <jwb> Base can go beyond install.  we're talking about function, not packages
16:01:24 <pknirsch> aye
16:01:27 <pknirsch> just wanted to say:
16:01:41 <dgilmore> a minimal install for instance wont have the compose tools, which we seem to be saying will be part of base
16:01:41 <pknirsch> Base can include more than a minimal set of packages, but offer a minimal install
16:01:47 <pknirsch> exactly
16:02:14 <pknirsch> that could be reflected in specific comps groups i suspect
16:02:21 <dgilmore> pknirsch: yep
16:02:35 <masta> ok, fair enough
16:02:44 <pknirsch> with <minimal> being yum, rpm, vim, kernel
16:02:49 <pknirsch> e.g.
16:02:59 <pknirsch> just as an example
16:03:03 <dgilmore> yep
16:03:14 <pknirsch> would that work for you haraldh ?
16:03:55 <pknirsch> i personally suspect we'll see quite a few things moving back and forth over the coming months anyway between WGs
16:03:59 <pknirsch> who is doing what, etc
16:04:11 <dgilmore> pknirsch: right
16:04:32 <dgilmore> i think as a great starting point what jwb proposed is spot on
16:04:40 <pknirsch> but i agree personally with masta and haraldh, i would love to have the option to have a very minimal install
16:04:43 <pknirsch> aye
16:05:14 <pknirsch> and one very important point for me is that Base is at least in my personal opinion a helper for all other products
16:05:21 <jwb> yes
16:05:23 <dgilmore> pknirsch: right
16:05:40 <dgilmore> i dont think we will mandate that they must do foo
16:05:41 <pknirsch> so anything we discuss here has to be brought up with the other WGs to see if that will work for them
16:05:46 <pknirsch> aye
16:05:56 <pknirsch> we'll here to provide something for them
16:05:58 <haraldh> pknirsch, I would say that the minimal install is our critpath, and the add-ons, which are commonly used by the products are second class
16:06:01 <dgilmore> but instead this is the frameworks and platform we provide for you to build on
16:06:02 <pknirsch> not to force them to use FOO
16:06:11 <pknirsch> right
16:06:44 <jreznik> dgilmore: exactly
16:06:56 <pknirsch> haraldh: how would you separate those then? logically and technically? do they belong to Base? Are they treated the same way? etc.
16:07:46 <dgilmore> with a releng hat on ive been trying to envision what the release tree will look like
16:08:06 <dgilmore> im not sure we want a Base tree
16:08:22 <dgilmore> and if we did i guess it would provide a pxe tree and boot.iso only
16:08:25 <jreznik> (btw. on how long meeting we agreed? 1.5h?)
16:08:40 <haraldh> hmm, why not start with the anaconda plus minimal install for all options and then decide bit by bit on extending the Base tree
16:08:50 <dgilmore> jreznik: i would say no more than that
16:08:57 <pknirsch> jreznik: until we're done ;) but if you have to leave at some point no worries
16:09:11 <dgilmore> haraldh: 15:57 < jwb> so Base to me sounds like: installer, compose tools, minimal install (for some definition there), and functionality all products want to use
16:09:13 <pknirsch> dgilmore, jreznik: yea
16:09:52 <haraldh> dgilmore, "functionality"  all products want to use ... ok
16:09:54 <pknirsch> dgilmore: Why would you not see the need for a Base tree?
16:10:09 <jreznik> I think infra around is really more important than just packages (releng infra, qa infra - possibly automated as much as possible)
16:10:25 <dgilmore> pknirsch: well i just wonder what purpose it would serve
16:10:29 <pknirsch> jreznik: +1
16:10:59 <pknirsch> dgilmore: well, if we can install Base, then wouldn't we want a closed repo revolving around it then?
16:11:11 <dgilmore> pknirsch: we could make a base tree or just make Everything have an installer in it
16:11:12 <pknirsch> just wondering out aloud here
16:11:27 <pknirsch> how would i know then that i'm only using Base packages?
16:11:29 <dgilmore> pknirsch: maybe we do
16:11:47 <dgilmore> pknirsch: i really do not know
16:12:03 <dgilmore> and i need to work out what the tree will look like
16:12:21 <pknirsch> ye, as i said, it's just me thinking out aloud, not sure if we need it either
16:12:21 <dgilmore> and take it to the mirrors. so that we can get input from them
16:12:28 <pknirsch> yup
16:12:33 * jreznik wonders how other meta-distributions are defined - Mer especially
16:12:37 <kmacleod> iirc, someone on-list mention having the "minimal network install iso" would be in-scope
16:13:21 <pknirsch> mhm
16:13:27 <haraldh> I don't think we need a Base repo.
16:13:35 <haraldh> We only need a "comps"
16:13:40 <pknirsch> good point
16:13:49 <masta> just comps
16:13:51 <haraldh> and we already have that
16:13:57 <haraldh> "minimal" and "base"
16:14:21 <haraldh> and I would trim those to fit our definition
16:14:34 <pknirsch> repoquery --requires --resolve --all yum rpm vi kernel | wc
16:14:43 <pknirsch> 38      38    1178
16:14:44 <pknirsch> heheh
16:14:48 <haraldh> and I would trim those to fit our definition
16:15:05 <haraldh> esp. if dnf is rewritten in C
16:15:10 <masta> haraldh: that sounds good to me
16:15:19 <pknirsch> jup
16:15:21 <pknirsch> +1
16:15:36 <pknirsch> so lets settle for an initial proposal then.
16:15:48 <dgilmore> one question, i guess we need to have the base python and gcc also
16:16:07 <pknirsch> why gcc?
16:16:10 <haraldh> dgilmore, you mean, that base can compile itsself?
16:16:14 <dgilmore> pknirsch: libgcc
16:16:28 <dgilmore> haraldh: no that would extend things to far
16:16:54 <pknirsch> hm, my repoquery didn't list libgcc
16:17:18 <pknirsch> (f19 where i tested it on a ppc64)
16:17:47 <jwb> glibc on x86_64 requires libgcc
16:18:03 <jwb> nothing explicitly requires glibc because that would be pointless
16:18:10 <jwb> so your repoquery didn't catch everything ;)
16:18:27 <pknirsch> ye, but you get it in via automatic dependencies from rpm
16:18:37 <jwb> then ppc64 is weird
16:18:39 <pknirsch> as glibc is in :)
16:18:49 <pknirsch> set, let me fpast it
16:19:05 <jwb> do we really need to review this right now?
16:19:06 <dgilmore> both fall into minimal install and anaconda deps anyay
16:19:17 <dgilmore> so i guess i just made noise
16:19:35 <haraldh> jwb, no
16:19:47 <dgilmore> i don't think we need to work out specifics now
16:19:54 <pknirsch> http://fpaste.org/52688/38392758/
16:20:03 <pknirsch> ye
16:20:11 <pknirsch> it's something we can later
16:20:19 <dgilmore> i think going with "15:57 < jwb> so Base to me sounds like: installer, compose tools, minimal install (for some definition there), and functionality all products want to use" is what we want now
16:20:20 <pknirsch> so definition of purpose proposal:
16:20:22 <pknirsch> aye
16:20:29 <pknirsch> everyone fine with that?
16:20:32 <jreznik> dgilmore: +1
16:20:36 <jwb> wait
16:20:41 <jwb> we said s/all/majority
16:20:42 <haraldh> +1
16:20:45 <pknirsch> right
16:20:50 <jwb> other than that, fine with me
16:21:30 <pknirsch> #info Base definition: installer, compose tools, minimal install (for some definition there), and functionality the majority products want to use
16:21:58 <jreznik> jwb: ah, missed that bit, yes, majority
16:21:59 <dgilmore> proposal #agreed Base definition: installer, compose tools, minimal install (for some definition there), and functionality the majority products want to use
16:22:23 <haraldh> ok guys, I have to leave unfortunately, but I guess the most important definition was made :)
16:22:35 <pknirsch> do we need to cover containers in more detail today? i think no.
16:22:36 <dgilmore> haraldh: cheers
16:22:44 <dgilmore> pknirsch: i dont think so
16:22:45 <jwb> pknirsch, not today :)
16:22:49 <jreznik> do we want to use a bit nicer words as server guys do to define mission statement or just Base is ...
16:22:50 <pknirsch> ok
16:23:14 <pknirsch> jreznik: i can try to word wrangle it a bit and put the mission statement in /Base
16:23:24 <dgilmore> jreznik: i think in the wiki it would be expanded
16:23:35 <jreznik> yep, that's my point
16:23:39 <dgilmore> but in the meeting short and brief is okay
16:23:39 <pknirsch> jup
16:23:53 <jreznik> reading Server's one and don't get it at all :)
16:23:59 <pknirsch> hehe
16:24:05 <jreznik> so maybe less words but contain real description is actually better
16:24:12 * pknirsch nods
16:24:20 <pknirsch> i'll try to write something up by Monday EOB
16:24:47 <dgilmore> pknirsch: cheers
16:25:04 <jwb> ok, i need to drop off too
16:25:07 <pknirsch> ok
16:25:13 <jwb> i'll review anything that comes after in the minutes.  thanks all
16:25:20 <pknirsch> that was all that was on the agenda for today anway
16:25:35 <dgilmore> pknirsch: i think we should wrap up
16:25:39 * jreznik is working on wiki + putting meeting time to irc channel wiki + fedocal
16:25:50 <pknirsch> #topic Open Floor
16:26:02 <pknirsch> anything else for today then?
16:26:04 <jreznik> I'll send you a note to review wiki once it's all done
16:26:12 <pknirsch> thanks :)
16:26:20 <jreznik> with using Server as basis for our
16:26:34 <dgilmore> jreznik: thanks
16:27:17 <jreznik> one thing we already touched - how are we going to coordinate with other WGs?
16:27:36 <jreznik> for me it's a black area and we definitely want some sort of coordination
16:27:38 <dgilmore> everyone don't forget to join #fedora-base
16:27:54 <dgilmore> jreznik: i think we need to work that out
16:28:18 <pknirsch> jreznik: good point. maybe sending out our weekly meeting notes to the coordinators of each WG?
16:28:18 <jreznik> yep, definitely willing to help here
16:28:21 <dgilmore> jreznik: we didnt touch on release schedules and supported lifecycles etc
16:28:38 <jreznik> dgilmore: yep, that's my point + a few more other topics
16:29:27 <jreznik> maybe reposition Board a bit to be more council like constituted from WGs representatives, other teams...
16:30:03 <dgilmore> jreznik: or have a monthly WG council meeting with at least two people from each WG
16:30:03 <jreznik> as we really need also other teams to be pulled in (and not only qa everyone talks about but marketing, docs...)
16:30:16 <jreznik> dgilmore: that's something I'd like to propose
16:30:31 * jreznik was thinking about it this week
16:30:36 <dgilmore> we need to make sure communications happen and not assume so
16:31:07 <jreznik> only WGs or should we invite other teams? at least the top ones (docs, websites, marketing, design)?
16:31:25 <dgilmore> jreznik: probably all the above
16:31:26 <jreznik> as it all has to fit together in the end
16:31:37 <pknirsch> agree
16:31:41 <dgilmore> maybe ambassadors also
16:31:42 <jreznik> I'll take an action item to start working on this
16:31:53 <pknirsch> communication is very important, thanks jreznik !
16:32:07 * Southern_Gentlem is watching
16:32:25 <jreznik> well, even me sometimes fails to communicate properly...
16:32:31 <pknirsch> Southern_Gentlem: everyone is welcome here! :)
16:32:41 <pknirsch> jreznik: welcome to the club ;)
16:32:52 <jreznik> not today but I'll try to talk to other WGs folks, teams and will try to come with a plan next week
16:33:01 <pknirsch> cool
16:33:17 <dgilmore> jreznik: i know im guilty of it
16:33:24 <dgilmore> i am trying to be better
16:33:25 <pknirsch> #info jreznik working on a communication plan with other WGs
16:33:42 <pknirsch> yea. sometimes it's really hard when you're drowning in work.
16:33:46 <jreznik> dgilmore: same here, so trying to find an excuse to be better and help there
16:33:54 <jreznik> pknirsch: you can use action
16:34:04 <pknirsch> #action  jreznik working on a communication plan with other WGs
16:34:06 <pknirsch> there
16:34:07 <pknirsch> better
16:34:10 <pknirsch> ;)
16:34:11 <dgilmore> pknirsch: si
16:34:24 <dgilmore> pknirsch: when we agree on things you can use #agreed
16:34:38 <pknirsch> i think we all agree jreznik should do it :P
16:34:39 <dgilmore> sho sup differently in the meeting summary
16:34:53 <jreznik> in that style - proposal #agreed, get ack, copy it without proposals
16:35:17 <dgilmore> pknirsch: for things like when we agreed to what we are going to start out with as base
16:35:20 <jreznik> do we have our own mizmo here to prepare hers nice summaries? :)
16:35:46 <pknirsch> ha jreznik, you wish!
16:38:25 <dgilmore> okay
16:38:29 <dgilmore> anything more?
16:38:39 <dgilmore> if not we should close up shop
16:39:31 <dgilmore> pknirsch: ?
16:39:37 <pknirsch> not from me. just one agenda item for next wek i'd like to talk about it schedule resp. releases
16:39:38 <pknirsch> for base
16:39:44 <pknirsch> but we can discuss that next week
16:39:48 <pknirsch> when everyone is here again
16:40:07 <dgilmore> perfecto
16:40:08 <pknirsch> alright, then lets wrap up. Thanks everyone for joining and contributing, we got a lot done on our first meeting!
16:40:37 <masta> thanks
16:41:32 <pknirsch> #endmeeting