14:00:40 #startmeeting Workstation WG 14:00:40 Meeting started Wed Jul 6 14:00:40 2016 UTC. The chair is stickster. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:40 Useful Commands: #action #agreed #halp #info #idea #link #topic. 14:00:40 The meeting name has been set to 'workstation_wg' 14:00:45 #meetingname workstation 14:00:45 The meeting name has been set to 'workstation' 14:00:48 #topic Roll call 14:00:50 .hello pfrields 14:00:51 stickster: pfrields 'Paul W. Frields' 14:00:56 .hello mclasen 14:00:58 mclasen: mclasen 'Matthias Clasen' 14:01:27 .hello otaylor 14:01:28 otaylor: otaylor 'Owen Taylor' 14:02:09 .hello kalev 14:02:10 kalev: kalev 'Kalev Lember' 14:02:25 .hello ryanlerch 14:02:26 ryanlerch: ryanlerch 'ryan lerch' 14:03:32 #chair mclasen otaylor kalev ryanlerch 14:03:32 Current chairs: kalev mclasen otaylor ryanlerch stickster 14:04:51 #topic Any agenda changes? 14:05:04 hello 14:05:11 #chair cschalle 14:05:11 Current chairs: cschalle kalev mclasen otaylor ryanlerch stickster 14:06:19 #info looks like nothing to list, not surprising given short notice 14:06:34 #topic F25 -- rpm-ostree status and next steps 14:06:49 So hi all, our agenda today is basically carryover from last meeting 14:07:40 Also I want to apologize for short notice to the list for this meeting. I spent most of yesterday catching up after basically being away for most of 11 calendar days. 14:07:50 But rather than lose momentum, I figured we could catch up with where we are on other F25 topics 14:08:41 no updates from me on this one 14:10:11 mclasen: Can you remind (for purposes of minutes) what's next up for rpm-ostree? Is the proof of concept still underway? 14:10:24 .hello rdieter 14:10:26 hi 14:10:33 rdieter_work: rdieter 'Rex Dieter' 14:10:34 o/ Rex 14:10:40 #chair rdieter_work 14:10:40 Current chairs: cschalle kalev mclasen otaylor rdieter_work ryanlerch stickster 14:10:49 next steps on this are: a) figuring out the installer story and b) buildsystem integration 14:10:58 .hello andreasn 14:10:59 andreasn: andreasn 'Andreas Nilsson' 14:11:34 I'll take the action to get an update on the installer story for next time 14:12:20 mclasen: That reminds me, dcantrell was asking about gathering work items for installer, so yeah, let's talk about that further offline 14:13:07 #info Next steps for rpm-ostree: figuring out how this will work with installer; and integrating into the buildsystem (discussion underway) 14:13:23 Anything else we need to say about rpm-ostree here? 14:14:54 OK, I take that as a no... 14:15:06 #topic F25 -- Overarching themes/plans 14:16:29 So something we have been lacking IMHO is any discussion of larger developer story for the Workstation. I would like to know whether I'm alone in that, and if not, what story/theme we could pursue during the F25/26 timeframe 14:16:45 #chair 14:16:45 Current chairs: cschalle kalev mclasen otaylor rdieter_work ryanlerch stickster 14:17:22 stickster: I'm working on a talk about the Fedora developer story for flock ... not sure I have a lot to share yet 14:18:56 stickster: mostly tjhings we can do are a) providing coherent guidance, even if it means making decisions b) doubling down and not accepting pain points (e.g. defaulting vagrant to vagrant-libvirt but not fixing major issues with that) c) offering slick integration points - there's still quite a bit more we can do with gnome terminal 14:18:59 * mclasen will sadly miss that - I hope flock talks are recorded ? 14:19:49 mclasen: unclear, but I'm looking into it -- we should be able to get laptops to at least do on the fly videos similar to last year 14:19:51 otaylor: I would be interested in gnome-terminal ideas; had some chat we rishi about it 14:20:53 otaylor: These sound like worthwhile things, although (1) is rather vague to me; but not sure they reach the point of an overall theme 14:22:44 stickster: 1) is of course vague without actually making the decisions :-) THere's a million paths to devleopment these days, and developer.fedoraproject.org mirrors that - trying to figure out what we consider most important, make sure they are painless, and make sure that developer.fedoraproject.org points to them 14:23:51 mclasen: Yeah, we can discuss them more as I get more advanced in my thinking - also need to get some design sanity in at some point 14:24:04 aday had some terminal mockups in toronto 14:25:45 stickster: Well, we can see as things gell up if there's an overall theme, but it's all incremental work - we can create a nice highlight of things we added but we aren't going to have some major new tool 14:26:17 otaylor: ' 14:26:18 oops 14:27:28 otaylor: One thing we can consider, although it depends on progress on the Hubs project, is hooking developers up with relevant groups through their respective Hub. I don't mean a hub of packagers, but rather e.g. a hub of people writing web apps in Ruby 14:28:13 * stickster imagines in the future the hubs will self-organize around areas of interest, basically like a Fedora SIG but without any bureaucracy 14:28:23 that's probably further off than F25 though 14:29:01 although we may end up doing that through the fedora developer portal as opposed to anything in the platform 14:30:06 cschalle: any thoughts on this? 14:30:07 Could be cool - that goes along a lot with making decisions - if there's a "fedora way of doing ruby apps" if the operating system is the least of the variation then it's not much of an affinity group 14:30:17 * mclasen wonders if ruby developers would feel any need to join a hub ... other than github, that is 14:30:50 mclasen: github gives you project collaboration, not trying to replace that or get in the way. Our hubs are about having a group of people to ask questions, exchange information, learn, etc. 14:30:58 stickster, well my main goal would be that as soon as we have the 3rd party proposal approved I want to reach out to IDE makers to see if we can get flatpaks of most major IDEs 14:32:51 *nod, beefing up content is going to be important to long-term flatpak success 14:33:46 well and getting more up to dates IDE is important for our general developer focus :) 14:33:55 stickster: yeah, there is not really anywhere for, say a ruby dev to ask questions atm 14:34:02 #fedora maybe? :/ 14:34:07 ryanlerch: that's a wasteland 14:34:20 "how do I make my nvidia driver work" (ad infinitum) 14:34:49 for that to work well, we need some other 'preferred developer workflow' questions answered, though 14:34:50 stickster, well we are hoping to have that question fixed properly for F25 (NVidia) 14:34:53 for a developer... they're going to be looking for people working on the stuff they work on 14:35:09 cschalle: My point wasn't about Nvidia, it was about the difference in who's present 14:35:17 but good to know 14:35:24 stickster, I know, just brought it up since you made that joke :) 14:35:35 1/2 joke ;-\ 14:35:48 e.g if vagrant is the preferred way to do development, we need to make sure that flatpak doesn't get in the way of it 14:36:46 mclasen: That seems very relevant. Case in point, in our team we're looking to roll out Vagrant files for our under-development apps (e.g. Bodhi) 14:36:49 mclasen: You are talking about packaging ides? 14:37:06 cschalle was 14:37:22 but yeah, my comment was in response to that 14:37:28 mclasen: yeah - just making sure I understand the context 14:39:28 mclasen: We can't really change the workflow of how Eclipse or Android Studio or PyCharm work - luckily most of these IDes are going to be set up for completely homedir development with their own isolation 14:39:50 right 14:40:46 Sandboxing will be tricky, but maybe we just punt that. 14:41:42 otaylor: tricky in what sense? 14:42:17 because they need arbitrary access to homedir material? 14:42:21 stickster: In the sense that an IDE is a big app that is written with the epxectation of accessing your homedir (or at least some directories within it) freely 14:42:29 ha, *jinx 14:42:37 * stickster happy he got something a bit correct 14:43:10 stickster: We could have "Allow Eclipse access to read and write files in ~/workspace?" 14:43:50 stickster: or I supose we could have an Eclipse flatpak that did everything within the app directory in ~/.local, but do people really want gigs of stuff there? 14:44:00 right 14:44:09 they don't see it == they might not back it up ;-) 14:44:43 otherwise, might need a packager effort to standardize somewhere like ~/Projects or ~/workspace 14:45:02 stickster: But yes, homedir access is one problem, and the second problem is that we have little development influence on these projects and there's no way we can do anyting that requires comprehensive code changes 14:45:05 so yeah... punting is also an option I guess, although we might need to explain why 14:45:17 "you guys said flatpaks are secure" 14:45:53 A: "well, in this case, it's more important for developers to be able to do what they need to and we can't predict the needs for every app they're working on" 14:46:34 I think we probably wouldn't present them as "not sandboxed" but rather "having unrestricted access to your files" 14:46:50 well we do have some influence on Eclipse, so if the changes we want are not to major I wouldn't think that would be impossible 14:46:53 may be a worthwhile experiment to make an eclipse flatpak 14:46:59 and see how it fares 14:48:24 *nod 14:52:01 OK, well, we appear to have a couple things to act on 14:52:40 #action mclasen provide update on installer story for rpm-ostree for next meeting (July 20) 14:53:23 #action otaylor share more about Flock talk on list or at next meeting, whenever first opportunity arises 14:54:14 #action cschalle stickster pursue approval of 3rd party software proposal so we can make progress on more flatpak content such as IDEs 14:54:39 anything I missed? 14:55:44 * otaylor notes that his talk is scheduled against langdon's modularity talk, so he probably needs to get his content out separately if he wants anybody to see it 14:55:46 at some point we should revisit default apps for f25; eg shotwell -> photos 14:56:18 modularity, now thats a sexy topic 14:57:16 mclasen: OK, if I want people who are interested in things like modularity to see it 14:58:02 * mclasen just commenting on the general idea that people could get excited about modularity... 14:59:59 * otaylor notes to call his next flock talk, regardless of the subject, "Fedora, Base jumping, and wingsuits" 15:00:15 otaylor, i have that one submitted already ;) 15:00:28 * langdon was apparently half lurking 15:00:34 only if you do it live onstage 15:00:46 :-) 15:00:52 And with that... 15:01:00 #endmeeting