14:17:14 <haraldh> #startmeeting Fedora Base Design Working Group (2015-06-01)
14:17:14 <zodbot> Meeting started Mon Jun  1 14:17:14 2015 UTC.  The chair is haraldh. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:17:14 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
14:17:20 <jreznik> hi haraldh!
14:17:20 <haraldh> #meetingname  Fedora Base Design Working Group
14:17:20 <zodbot> The meeting name has been set to 'fedora_base_design_working_group'
14:17:26 <haraldh> #chair haraldh msekleta jreznik dgilmore vpavlin masta dwalsh
14:17:26 <zodbot> Current chairs: dgilmore dwalsh haraldh jreznik masta msekleta vpavlin
14:17:34 <haraldh> hi :)
14:17:37 <msekleta> hi everyone
14:17:43 <jreznik> hey msekleta
14:17:46 <haraldh> long time no see (read)
14:17:52 <msekleta> indeed
14:18:06 <masta> hello
14:18:24 <jreznik> and I have to leave again in 15 minutes, unplanned wfh and I have to move to the office for an hour :(
14:18:27 <masta> myself and dgilmore are presently in a meeting in Westford, so might be distracted
14:18:27 <msekleta> jreznik, hi
14:19:11 <haraldh> so, if you have any items to discuss, state them now
14:19:47 <jreznik> Fedora Modularization - https://fedorahosted.org/council/ticket/26
14:20:03 <jreznik> seems like we're getting to the point Base will be really needed
14:21:31 <haraldh> #topic Fedora Modularization - https://fedorahosted.org/council/ticket/26
14:22:07 <jreznik> it's an objective proposal by langdon
14:22:20 <jreznik> aka what's next in Fedora.Next
14:22:37 <jreznik> and it fits the reason for Base to exist
14:22:47 <langdon> is there a council meeting now?
14:22:51 <jreznik> so I recommend all Base WG members at least CC on the ticket
14:22:54 <langdon> ohh .. sorry.. this is base..
14:22:56 <jreznik> langdon: it's Base WG
14:23:01 <langdon> i am here if you need me :)
14:23:13 <jreznik> but as I wrote above, this ticket is the reason why we have Base WG (mostly)
14:23:20 <jreznik> (if you agree ;-)
14:24:29 <haraldh> ok, put myself on CC
14:24:50 <langdon> jreznik, well.. i would say the "data gathering" should be by edition wgs, implementation by envs&stacks and base.. i recommended to the e&s group that they prepare a list of qs they were curious/concerned about to ask the editions to get answered..
14:25:02 <langdon> more info on this here: http://1angdon.com/2015/05/28/fedora-moularization-prototypes/ as well
14:25:24 <haraldh> although I am not sure, what the Base WG can add, because we have the other products mainly as "users"
14:25:30 <langdon> i also, based on the comments on the ticket, need to do a bit of a rewrite/expansion.. but .. life/work...
14:26:00 <jreznik> langdon: data gathering should be already done with env & stack and base together... you can't just throw anything on other groups, that's why I pointed out Base WG to this ticket
14:26:34 <langdon> jreznik, well.. "end user requirements" seem to be the purview of the editions, no?
14:27:16 <jreznik> haraldh: that's the problem - we are not yet there... Base is just something that was called Fedora, not that much done as we initially planned to have well defined base to be in hands of Base WG and then build on it
14:27:49 <jreznik> langdon: you're right if you look on it from this POV
14:28:30 <jreznik> there are also more groups than editions/wgs... so it's really broad scope :(
14:28:59 <jreznik> in the ideal world everyone helps and everyone is happy :D I don't think we live in ideal world but real :)
14:29:34 <jreznik> but as I said - my main idea now was to spread it to Base WG group and let others know it's coming so they are aware and can join in early phase
14:29:45 <jreznik> I don't think we have much more to say today
14:30:02 <haraldh> jreznik, thanks for bringing it to attention, though :)
14:30:18 <langdon> jreznik, i guess i imagined that base and e&s would get their requirements from the "people who depend on them" .. base for the content, e&s for the "how"... so.. you need e&s and the editions to give you what base needs to support
14:31:16 <jreznik> langdon: that was initial idea :) but honestly, we failed in gathering that requiremens I'd say (my personal opinion)
14:31:37 <jreznik> so I take it also as base restart
14:32:15 <jreznik> I'm sorry but I have to leave now, just bad day today - I'll be available next week and I'll re-read log once I'm back
14:32:22 <langdon> jreznik, well.. we are hoping to get the data back by flock.. i think everyone has been so worried about just getting the edition stuff sorted out that we haven't really had a lot of time/energy to focus on the next steps
14:32:49 <jreznik> I agree
14:33:18 <haraldh> I was more thinking about E&S stating their package requirements and base would be more or less the intersection
14:35:06 <haraldh> any other topic? do we have candidates for membership?
14:35:34 <lnykryn> If you have some free seats, I would like to apply.
14:35:51 <langdon> haraldh, i think e&s is thinking about the "how" ... so.. (purely for example) they choose scl to support the rings model, then base would need to add scl as part of base, right? or if they decided "docker containers" .. base would need to add docker.. right?
14:38:42 <haraldh> lnykryn, sure :)
14:39:40 <msekleta> langdon, isn't that responsibility of e&s, WG has "stacks" in its name. So if they want provide ruby via SCL shouldn't they take care of both, scl-utils+packaging *and* ruby
14:40:26 <haraldh> if no other product has it in, I don't see, why it should be part of base, if it is "optional"
14:41:50 <langdon> msekleta, haraldh i would think "base" = scl-utils, if wkstn wants "ruby in an scl" they ship it.. relying on base's scl-utils.. e&s just tells wkstn and base the answer to "how we ship multiple versions of things" (in this scenario, they "said" scl)
14:42:37 <langdon> if all three editions want ruby in an scl, they can petition base to include it..
14:43:05 <haraldh> yes
14:43:30 <haraldh> agreed
14:45:22 <haraldh> lnykryn, too bad we are low on members today, so I will add it to the agenda for the next time
14:45:33 <lnykryn> haraldh: ok
14:46:23 <msekleta> langdon, base already covers rpm so I figure that we might add scl-utils too as it is related
14:47:53 <langdon> msekleta, i guess i am making the argument that if e&s says "this is how we do x, which relies on y" then e&s needs to work with base to get y included in base for use by the editions (or other WGs in general)
14:49:53 <lnykryn> langdon: I don't think that what you have stated should be covered by base wg, for example different version of ruby should by covered by e&s and how we packaged things should be covered by FPC
14:50:18 * msekleta agrees with lnykryn
14:50:55 <msekleta> in this particular case (scl-utils) there is huge overlap with responsibilities of FPC
14:52:04 <langdon> lnykryn, msekleta well.. that indicates some communication gap :) because I am not sure e&s agrees.. FPC covers "packaging" yes.. but that is not the whole of "things like this" ... FPC, for example, might draw up the guidelines of scl which were determined to be a "good thing" by e&s.. or similar for containers..
14:52:35 <langdon> as for ruby..i don't think i was saying base should do ruby.. i also don't think e&s should do ruby..
14:52:36 <langdon> i think e&s is more advisory
14:52:55 <langdon> however, it sounds to me like a joint meeting between base and e&s might be a good idea :)
14:53:06 <langdon> and/or discussion with the council
14:56:18 <haraldh> langdon, sure
15:06:38 <haraldh> ok, seems like we have no other topics
15:08:15 <msekleta> haraldh, maybe we could figure out until next meeting if we carry on with work Phil's guys started around dependency minimalization
15:08:37 <haraldh> I asked Nils about it
15:08:41 <haraldh> just now
15:09:05 <haraldh> and he said, that because of build failures and lack of time, it kind of stalled right now
15:09:25 <haraldh> but, I already put that to the agenda of the next meeting
15:09:34 <msekleta> haraldh, thanks
15:12:10 <haraldh> ok, then...
15:12:14 <haraldh> #endmeeting