20:02:31 <sarob> #startmeeting milk
20:02:32 <openstack> Meeting started Mon Apr 28 20:02:31 2014 UTC and is due to finish in 60 minutes.  The chair is sarob. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:02:33 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
20:02:36 <openstack> The meeting name has been set to 'milk'
20:02:42 <sarob> roll call
20:03:17 <sarob> -
20:04:16 <sarob> roll call
20:06:39 <sarob> -
20:25:14 <sarob> milk team roll call
20:25:24 <sarob> sing out
20:25:40 <sarob> MyJoshua, ping!
20:25:42 <MrJoshua> here
20:25:49 <sarob> MrJoshua, ping!
20:25:54 <sarob> hey
20:25:57 <MrJoshua> Just recovered from computer crash
20:26:05 <sarob> sad panda
20:26:12 <MrJoshua> ha
20:26:34 <MrJoshua> Do you have Kevons (sp) email address?
20:26:41 <sarob> i believe so
20:27:10 <MrJoshua> Send it to me if you would please.
20:27:20 <sarob> i think we need to get a few of us to set aside a half day to make some progress
20:27:27 <MrJoshua> So we’re suposed to have a bunch of interns now.
20:27:30 <MrJoshua> I agree
20:27:32 <sarob> we can collab
20:27:35 <sarob> over irc
20:27:44 <sarob> interns, through erik?
20:27:51 <MrJoshua> yes, and yes
20:27:55 <sarob> yup
20:28:07 <MrJoshua> I’m much more open to do that now.
20:28:10 <sarob> so lets drop back to square one
20:28:19 <sarob> build out about 10 sprints
20:28:32 <sarob> 2 weeks apiece
20:28:46 <sarob> min goals
20:29:28 <sarob> sprint 1: define API model
20:29:48 <sarob> sprint 2: apiaryIO frame
20:29:52 <sarob> sprint 3...
20:30:48 <sarob> with sprint 10 being web services, clients, and API with scripted build
20:31:10 <sarob> we can start breaking up work
20:31:32 <sarob> mrjoshua
20:31:46 <MrJoshua> Ok, well the fist step I think is that inital meeting and see who we have from erik.
20:32:04 <MrJoshua> I prefer having a working thing at the end of each effort.
20:32:20 <sarob> hey im all for that
20:32:45 <sarob> we small bits of success to build on
20:32:45 <MrJoshua> Then we can point to something and meausre it’s usefulness over time.
20:32:53 <sarob> right
20:32:53 <MrJoshua> yes
20:33:09 <sarob> with the end of summer as the timeline
20:33:15 <sarob> 2 week sprints
20:33:28 <sarob> your thoughts on objectives
20:33:32 <sarob> for each sprint
20:34:05 <sarob> we have 3-5 people with 5-10 hours per week is my guess
20:34:21 <MrJoshua> I have new insights on archtecture now that the C4 project is in place.
20:34:28 <sarob> sweet
20:34:33 <sarob> lay it on me
20:36:12 <MrJoshua> I think I understand better how we can build a new `type` of OpenStack element that applies a structure to an OpenStack deployment perhaps a bit like OpsWorks at AWS but focused on production.  This is a practical application then.
20:36:42 <sarob> #link https://www.youtube.com/watch?v=sK50So-yYRU
20:37:11 <sarob> hmm okay
20:38:25 <MrJoshua> The api componants are, storage, db (metadata), transport channels (ill come back to this), and crypto / security
20:39:02 <MrJoshua> Storage is arcive, active, local and other domains.
20:39:05 <sarob> does this line of thinking still apply to using paas like cloudfoundry dea to build,deploy
20:39:18 <MrJoshua> yes
20:39:52 <sarob> okey dokey
20:40:26 <sarob> so do have a design recommendation based on opsworks?
20:40:37 <MrJoshua> db (metadata) can be abstract like Karma, but if we make some concrete examples than I think it’ll be better.  To this end there are some standared columns that we’ll need to define, I have a list.
20:40:52 <sarob> sarob like list
20:40:55 <MrJoshua> No this is based on my work with C4, and it’s deployment.
20:41:21 <sarob> deployment configuration
20:41:26 <sarob> yeeess
20:41:29 <sarob> ?
20:41:37 <MrJoshua> yes
20:41:48 <sarob> sarob understand
20:42:02 * sarob sarob thinking of thinking
20:42:05 <MrJoshua> db / metadata has some syncronization modles that we need to look at.
20:42:18 <sarob> hmm, okay
20:42:25 <sarob> example?
20:42:40 <sarob> i have a hard stop at :55
20:42:42 <sarob> btw
20:43:18 <MrJoshua> one example http://en.wikipedia.org/wiki/Vector_clock but I think we have better.
20:45:02 <MrJoshua> transport should have spacificly defined modes/channels.  Along the lines of automatic pre-proxying of asset data, streaming, or only for archinve etc.
20:45:03 <sarob> so breaking down synchronous into smaller asynchronous events?
20:45:11 <MrJoshua> yes
20:45:34 <sarob> okay
20:45:48 <sarob> im eager to see your list
20:46:07 <sarob> id guess that FIMS effort has taken some of these into account
20:46:58 <MrJoshua> Yes, within the general model that I’m putting forward and we can all revew, we can see what part FIMS ocupies already.
20:47:15 <sarob> nice
20:48:16 <MrJoshua> Finally security in addition to general accout/access managment we will want to have asset incryption with spacific proporites, such as sending assets to a spacific person or group, etc.
20:48:37 <MrJoshua> Also, sighing features etc.
20:48:45 <MrJoshua> er signing
20:49:47 <MrJoshua> So when do we want to setup a meeting and I can prespent this stuff? And we can start editing it for an initial sprint schedule.
20:51:00 <sarob> sounds good to me
20:53:03 <MrJoshua> So do you want to talk to erik about setting up some time for this half day think you’re talking about and I can present this for everyone then?  Or setup something seperate?
20:53:27 <sarob> half day engineering prep sounds great
20:53:48 <sarob> lets talk via email
20:53:52 <sarob> i have to run
20:53:56 <MrJoshua> k
20:53:59 <sarob> #endmeeting