17:00:19 #startmeeting murano 17:00:20 Meeting started Tue Sep 8 17:00:19 2015 UTC and is due to finish in 60 minutes. The chair is sergmelikyan. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:21 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:00:24 The meeting name has been set to 'murano' 17:00:30 o/ 17:00:48 o/ 17:00:58 Hi 17:01:12 hi 17:03:06 Agenda for todays meeting is available here: https://wiki.openstack.org/wiki/Meetings/MuranoAgenda/Archive 17:03:25 We don't have action items, so we can jump right on to the next topic 17:03:51 sorry, correction for the agenda link https://wiki.openstack.org/wiki/Meetings/MuranoAgenda 17:04:12 #topic Future of stackforge/yaql 17:04:30 I am not sure what exactly we should discuss here 17:05:06 @Nikolay_St do you have any inside regarding this topic? 17:05:16 mogration to openstack space? 17:05:32 It's going to happen automatically 17:06:08 I think the point was to ask if we all agree with that 17:06:57 what is there to disagree? 17:07:24 sergmelikyan: have we add yaql to the list? 17:08:08 there are pros and cons to this. I'm myself not sure what I'd like more. However if you all agree with this we can skip to the next topic 17:08:22 #link: https://wiki.openstack.org/wiki/Stackforge_Namespace_Retirement 17:08:44 it's here, so everything good from this side. 17:08:56 yep 17:09:22 @stan_lagun I don't see any difference between being on openstack/ with being on stackforge/ other than name 17:09:52 @stan_lagun just in case, can you provide more details regarding cons? 17:09:52 stan_lagun: ^^ 17:11:49 sergmelikyan: the difference is people's perception. People rarely use projects from openstack outside openstack. I'd prefer that yaql be either developed outside like SQLAlchemy or become official project like oslo.*. I don't think that current state of how yaql being developed is not what people expect from openstack project 17:13:34 stan_lagun: I guess you meant that yaql is developed not in the way what people expect from openstack project? 17:14:37 yes 17:15:30 Can you explain a little bit more what we should improve? 17:17:00 Project is developed 95% by 1 person without community being involved or even care, without yaql team, without QA etc. We have no elected cores, neither community meetings like this. No one tries to promote yaql. We have no plans to move it to Big Tent etc 17:17:19 though it is a major part of Murano, Mistral and Gon knows who else 17:17:24 *God* 17:17:27 stan_lagun: you will be surprised it's already in big tent :) 17:19:36 We (murano) care about YAQL, Mistral is also cares about YAQL 17:20:48 cores, community meetings and all nice stuff comes to well documented project and I hope we will finally start documenting YAQL, katyafervent and kzaitsev1ws have started working in this direction but... 17:21:25 stan_lagun: by the way your contribution to documentation as core member of the project is also invaluable 17:21:48 The problem that there is no one who is working on this full time. No one tries to promote and advertise it etc. My development in yaql is 50% take place on my personal time 17:22:32 Okay, this sounds not productive. If you believe yaql should be in openstack lets move on 17:22:40 stan_lagun: thank you 17:23:02 before we start to promote yaql we should have good docs for it 17:23:09 #topic New murano API 17:23:21 oh 17:23:29 I added this topic 17:23:55 katyafervent2: to have good docs someone need to write them. And it takes much time. And no one has it 17:23:55 I know at least one issue which can be fixed only with new API 17:24:07 #link https://bugs.launchpad.net/murano/+bug/1454723 17:24:08 Launchpad bug 1454723 in murano "[Dashboard] Incorrect environment status after failed action" [High,Confirmed] - Assigned to Nikolay Starodubtsev (starodubcevna) 17:24:40 Nikolay_St: this is the last thing in why we need new API to be honest though it is also important 17:24:40 we can't do anything with it now because we don't 'Application' object 17:25:03 stan_lagun: I just say that it's what I know about it 17:25:30 stan_lagun: I guess you have much more to say on this topic 17:26:14 so, I suggest start to collect the ideas about murano APIv2 17:26:41 yeah, do we have a list with requirements for a new api? 17:27:25 i think we can start with blueprint if we still dont have one 17:27:50 katyafervent2: and also create etherpad and write a letter to dev-list 17:27:54 I can do it 17:27:56 There are many things that are missing. We have no way to efficiently work with MuranoPL from outside - validate contracts from UI etc. We need a way to execute MuranoPL code for automatic object model generation to be possible. We need an API to be able to query object model. We need to do something with sessions and we need much more clearer implementation that spagetti code we have right now 17:28:12 Nikolay_St: before deciding that we need new API, we should start documenting requirements and start from there 17:28:15 and the first thing is that we need to design it 17:28:38 * sergmelikyan looser, didn't scroll before sent message 17:29:27 stan_lagun: you want to design before we collects some feedback? 17:29:34 Nikolay_St: I would appreciate that 17:29:55 sergmelikyan: you can create action item ;) 17:29:57 Nikolay_St, you can also hold a meeting to discuss that 17:30:07 I was asked to summarize our expectations from our UI vNext aka Merlin 17:30:17 stan_lagun: oh 17:30:17 and it is strongly connected with new API 17:30:22 I forgot about that 17:30:36 you're right. 17:30:46 but may be we can start in parallel? 17:32:45 i think so, and summerize all the thoughts after 17:32:52 #action Nikolay_St to send e-mail about collecting requirements for new API 17:35:42 so, that's all from my side on this topic 17:36:14 #topic Open Discussion 17:36:44 I have one topic 17:37:33 The question is if we are going to release MuranoPL/1.1 format for Liberty and 0.1 for core-library or we are going to leave it 1.0 and 0.0? 17:39:36 stan_lagun: I don't have strong opinion on this topic, let's postpone until the next meeting? 17:41:25 no problem. But we need to decide before release. The difference is that if I in my app use any function from yaql 1.0 that is not available in 0.2 (and there are hundreds of those) or something from core library that was introduced in Liberty there is no way for me to specify it. Apps written for yaql 1.0 will just fail when run on Murano Kilo 17:41:33 I need to spend sometime diving deeper to this topic 17:41:57 @stan_lagun thx 17:42:23 lets create action item 17:42:36 #action stan_lagun re-raise question about package version on the next meeting 17:49:01 sergmelikyan: should we end meeting? 17:49:48 mgershenzon: hi 17:50:00 you're a little bit late today 17:50:08 do you have any topic to discuss? 17:55:24 #endmeeting