18:01:55 #startmeeting Ubuntu TV discussion 18:01:55 Meeting started Fri Mar 8 18:01:55 2013 UTC. The chair is mhall119. Information about MeetBot at http://wiki.ubuntu.com/meetingology. 18:01:55 18:01:55 Available commands: #accept #accepted #action #agree #agreed #chair #commands #endmeeting #endvote #halp #help #idea #info #link #lurk #meetingname #meetingtopic #nick #progress #rejected #replay #restrictlogs #save #startmeeting #subtopic #topic #unchair #undo #unlurk #vote #voters #votesrequired 18:02:29 hi everybody, who is here for the Ubuntu TV meeting? 18:02:44 mhall119, let's try now :) 18:03:02 hi mhall119, here but just to follow (not participating :p) 18:03:37 YoBoY, same here. Would like to see how this is handled 18:04:09 <_danialjose> just to watch 18:04:11 I'm going to give a few more minutes, I hope some of the TV community guys can make it 18:04:45 hey tgm4883 18:04:53 o/ 18:05:22 Doesn't seem like much work has been done to Ubuntu TV for awhile now: http://bazaar.launchpad.net/~ubuntutv-dev-team/ubuntutv/trunk/files 18:05:33 Would be nice to see how far it has come 18:06:05 ok, I'm going to get started with some of the back story 18:06:10 #topic History 18:06:20 do we have more people here than normal? 18:06:35 tgm4883: yeah, I put out a call for more discussion 18:07:00 so, around this time last year Canonical debuted the Ubuntu TV at CES, and it was a pretty big hit 18:07:25 at the time, the TV UI was based on the Unity 2D code 18:07:32 which in early 2012 was still actively developed 18:08:03 quick correction, it was uds-P 18:08:04 we opened up all of the source code that had been worked on, and recruited interested developers from the community to participate in it's development 18:08:21 tgm4883: what was? 18:08:28 mhall119, oh wait, nm, I misread what you said 18:08:29 carry on 18:08:32 ok 18:09:18 so by this time last year the TV was open and we were getting great contributors like tgm4883 and bobweaver on board 18:10:12 then, at UDS-Q, there was a decision made by the Unity team that actively developing 2 separate Unity codebases (2D and 3D) was too much 18:10:39 and a big dicussion followed which ultimately resulted in Unity 3D being chosen 18:11:01 I wasnt a part of those conversations, so I can't tell you what the arguments were either way 18:11:49 so, with Unity 2D no longer being actively developed, we had to shift Ubuntu TV on to Nux and the Compiz plugin that powers Unity 3D 18:13:03 and it wasn't long after that when the Canonical engineers who had been working on the multi-media functionality for the TV were also tasked with getting that same multi-media functionality working on the new Phone/Tablet codebase 18:13:31 so for several months after we had started the transition to Unity 3D, we didn't have Canonical resources to help 18:14:00 and despite a large amount of effort from bobweaver, porting it from Qt (which he was familiar with) to Nux (which he wasn't) was no simple task 18:14:46 after months of this, those of us inside Canonical were frustrated that we couldn't contribute, and concerned that our absence was holding back to the community developers 18:15:24 so jhodapp and I spent some time gathering up all the internal documentation about design and implementation details that we could, and worked to get approval to open them up to the community 18:16:03 at this point, we were pretty much "throwing it over the wall", not because we wanted to, but because if we didn't it wouldn't get over the wall at all 18:16:39 what does throwing it over the wall mean ? 18:17:00 mr_man: it's a common term for making something open source but not supporting the community around it 18:17:22 like Google does with Android, or Sun/Oracle was known to do with their open source projects 18:17:27 sounds like the last year 18:17:46 cool so Ubuntu is droping the tv ? 18:17:48 mr_man, in-house code getting out to the public 18:17:55 Now, even though the phone shell was being written in Qt/QML at this time, as far as everybody inside Canonical knew the future was still Unity 3D, and the expectation was that the phone shell would eventually be integrated into Unity 3D as well 18:18:03 mr_man: no, this is the back story 18:18:34 so for a long time we kept encouraging the community TV hackers to port to Unity 3D, because as far as *we* knew, that was the right direction to take it 18:18:55 so you all got community members to do it ? 18:19:05 mr_man: yes 18:19:18 port to 3d ? that is or are you going to tell the to do that then change your mind 5 times 18:19:29 will is on board with this ? 18:19:33 and while tgm4883 made good progress on a MythTV lens/scope that didn't depend on the UI toolkit, bobweaver was still struggling to make heads or tails of Nux and Compiz 18:19:36 * tgm4883 encourages mr_man to hold his questions until the end of the history section 18:19:51 mr_man: I'm explaining what happened, not what is going to happen 18:19:56 sorry 18:20:40 I thought that bobweaver quit because of many many reasons 18:20:49 perhaps we should open the second channel for questions? 18:20:59 mr_man: I'll get to some of that in a bit 18:21:14 finally, as the phone's releasing was approaching, the conversation inside Canonical went back to the Unity 2D vs. Unity 3D debate, and after proving how quickly and efficiently the Unity shell could be done in Qt/QML, the decision was reversed 18:21:56 a lack of communication seems to be the problem. You thought Unity 2d was dead so strating working on gettting Ubuntu TV working on Compiz, only to find out that up-stream was now using qt/Unity-2D. 18:22:08 which put us in a difficult position, we weren't allowed to talk about the phone before it was announced, but we didn't want to waste developer's time either 18:22:32 pinguy: that is not true pm me if yyou want the truth 18:23:06 but we did discretely let bobweaver know that the debate had restarted, and there was a distinct possibility that Unity 2D, or at least a Qt/QML version of Unity, would become the new target 18:23:35 really you did 18:24:50 Which brings us to where we are now, we have an old Unity 2D codebase, some regretably wasted time and effort on Unity 3D porting, and a new Qt/QML foundating that we want to converge all of our devices around 18:25:13 so, that's the history 18:25:21 #topic Future 18:25:51 maybe I should say my side of the history as it was me that wasted 3 months or more of programming 18:26:09 bobweaver, I think thats only fair 18:26:19 so we wasted a lot of development time, and burned through a non-significant amount of developer good will, but we still want to see an Ubuntu TV, and we still want it to be a community project 18:26:46 bobweaver: go ahead 18:26:47 #topic alternate history, community edition 18:26:55 #topic more history 18:27:07 tgm4883: feel free to add your input as well 18:27:19 what really happened . after the last non virtual uds I was told o make nux I was also told that 3 times before then I was told back to 2d then back to this then back to that. It seems like there needs to be better leadership 18:27:43 so it was a flip flop falure 18:27:57 bobweaver, nobody will argue with that point I think 18:28:04 meaning that I was lead along a fake road 18:28:46 this is no one fault but mine 18:29:19 but for the people in the future I do not want to see them have the same thing happen to them. so in other words you all need better leadership 18:29:38 bobweaver, was it ever at the back of your mind this may happen? Qt is really the only way to make apps fully cross platform. Compiz isn't really that great for this. From an outsider looking in. Ubuntu TV always seemed it was being made to work on many devices. 18:30:01 I knew that it was always going to be qt 18:30:02 pinguy: sure, he was pushing us to just stick with Qt/QML the whole time 18:30:12 he was a constant advocate for it 18:30:30 I fought that for months pinguy the point is I do not want to see you all have something like this happen to the next bobweaver 18:30:46 tgm4883: do you have anything you want to add? 18:31:12 just that I'm available for questions 18:31:17 oh 18:31:26 btw I have ubunntu tv running on my nexus 7 18:31:34 really? that's awesome 18:31:40 and that for all the help i've asked for (testing wise), nobody has actually come though 18:32:18 are we ready to talk about the future now? 18:33:11 bobweaver, how is the signal reception ? :P 18:33:18 listen you all are never going to get any where fighting about the code base. in this world it seems like it is about hardware and that is it. until the right people get the right tools it is not going to work 18:33:25 tgm4883, its a shame you didn't have that many testers. You probably could of done with the project being more open 18:33:34 why build something that is for the desktop when it is going on a tv 18:33:46 people have said that on the channel before 18:33:48 pinguy, the stuff I need testing doesn't have anything to do with the ubuntu tv code 18:33:50 pinguy: the project was open, tgm4883 and bobweaver are both community contributors 18:34:04 alright 18:34:06 so let me add a little here, what I actually am doing 18:34:29 I'm a Mythbuntu developer. What I bring to the table is the Ubuntu TV integration with a MythTV backend 18:34:45 This is done via Unity scopes, so it works on any platform that runs Unity 18:34:53 ogra_: I am using brodcom java drivers 18:35:03 The requests I've had for testing are 2 things 18:35:26 1) Testing the scopes. This admittedly is a little more difficult, as it requires you to have a MythTV backend in your environment 18:35:54 2) Testing of the MythTV backend quick start guide I wrote. This is far easier for people to test, as you can do it in a VM 18:36:24 https://docs.google.com/document/d/19knOlqz8cV5_8VQ1tCvEd8tjEk6U50KsSOJCROR60o4/edit?usp=sharing 18:36:31 Ironically, the few people that have tested the scope have given me feedback. The people that said they would test the QSG haven't given me any feedback yet 18:36:57 Comments are open on the QSG, which is where I'm looking for them 18:37:00 I'm guilty of that, it's still on my task list, but I've been quite behind lately 18:37:06 tgm4883: what is QSG? 18:37:12 Quick Start Guide 18:37:13 Quick Start Guide 18:37:15 the link I pasted above 18:37:30 The link above is the long version, that includes every screen in the backend setup 18:37:50 I would prefer to ship the smaller version, but it skips parts that aren't changed and may be more confusing 18:38:26 Short version https://docs.google.com/document/d/1eGJ8gGtRqFxQ8xVWlR6XgP6FAFJEVZ-Ntq_TT8XhWEs/edit?usp=sharing 18:38:28 tgm4883: I will read the guide and get back to you 18:38:48 tgm4883: bobweaver: I'd like to start talking about where we go from here, since we only have 20 minutes left 18:38:48 ok 18:38:56 mhall119, sounds good 18:39:02 #topic Future 18:39:23 alright, so now that everybody has the brief history of the TV project, I want to know how we progress from here 18:39:55 we still want a TV formfactor of Unity, and we still (miraculously) have community interest in it 18:40:49 if it is going to be part of the phablet then you all should just tell them to do it they wrote the stuff to start with 18:41:07 the goal is to have it all in one codebase 18:41:17 which we're evidently calling "Unity Next" 18:41:24 I think that they are re-writing libunity no one can do anything till then 18:41:32 which will start with the phablet code, and add in desktop and TV formfactors 18:41:35 I think "Unity Next" developers need to develop the interface 18:41:47 that is a HUGE trouble that there not communicating between teams that is why I am out 18:41:52 I hate to be the bad guy here, but Ubuntu TV is really only going to work with Ubuntu Touch. In the last 4 months XBMC has really come into its own. 18:41:55 well that and many many other reasons 18:42:33 pinguy, Unity brings more to the table than XBMC 18:42:41 pinguy: you wath live tv with xbmc ? 18:42:55 tgm4883: the "Unity Next" developers in Canonical have a short window to both polish the phablet UI for potential OEM partners, *and* port the desktop by 13.10 time-frame 18:42:55 bobweaver, yes. It was part of the last release 18:42:57 also manage all your apps 18:43:14 bobweaver, the android version also manages all of your apps 18:43:28 mhall119, it makes more sense for the Unity developers to develop the new layouts 18:43:29 if we wait for Canonical developers to port the TV UI, it's going to take a while 18:43:39 bobweaver, that's not true, it's just Qt5 and QML so what exactly do you think you need to wait on from the unity devs? 18:44:03 pinguy: so then you should use it and stop telling people what they should or should not do. you know that I could have xbmc in the dash right 18:44:12 because there is qtmediahub 18:44:24 tgm4883: they will be writing the code to support adding and switching to different formfacts, but I doubt they'll be able to port the TV UI unless we wait unti the end of this year 18:44:30 jhodapp: they are re-writing unity 18:44:35 libunity I should say 18:44:36 and I really don't want to wait another year on this 18:44:42 This is OT, but XBMC doesn't have TV capabilities built in. They farm it out and use PVR backends (such as MythTV) 18:45:06 bobweaver: yes, the libunity APIs are going to be changing, but the TV layout and what data it will need shouldn't be 18:45:26 bobweaver, sure, but we have a phablet UI today right? what's the difference between having the TV UI today and getting it working, vs the phablet one that already works? 18:45:44 Umm... that is not possible it listens to libunity it uses nux all of that 18:46:01 Can't we just blow up the tablet interface and use that for TV? 18:46:03 bobweaver: the converged codebase will all be on the same libunity API 18:46:14 tgm4883: that would be a good place ot start I think 18:46:27 the phablet is not using all the parts of unity also the main thing would be dconf-qt that right there is huge 18:46:28 but the tablet is still geared to touch input 18:46:32 tgm4883, yes, I think that's a good place to start too 18:46:49 mhall119, IMHO, the work that we need to do is ensure that you can control the interface via remote control 18:46:55 also qx11info as you all are making new server one can not do anything that is why I made into standalone app that runs on android 18:47:08 ok, things are starting to fall apart into technical implementation here, I'd like to keep this more high-level 18:47:23 well, that and ensure you can use it from 10' away, but I think that is going to deal more with font sizes 18:47:32 mhall119: you ant tell people to waste there time that is not fair to them 18:47:34 mhall119, +1 18:47:36 bottom line ^^ 18:47:37 so let's re-cap what we have and what we don't 18:48:13 spinning wheels in the mud just gets you more stuck 18:48:16 we have the phablet codebase, we have a design spec for the TV interface, we have some existing Qt4/QML1 code that may or may not be reusable 18:48:48 we don't have: Canonical resources devoted exclusively to porting TV to the new codebase 18:48:59 mhall119: have you looked at the code you can not use it 18:49:14 bobweaver: I haven't but I'm not as technically proficient as you when it comes to code 18:49:17 mhall119, perhaps a better angle would be to list all the needs (even stuff currently done), then list what is done 18:49:32 tgm4883: ok 18:49:32 esp because of the things that are happening with the server and libunity you are telling people to waste there time 18:49:37 eg. what is needed to ship something the community would deem "Ubuntu TV" 18:50:00 * We need the converged Unity foundation to support multiple formfactors (Canonical resources are workign on this, it doesn't exist yet) 18:50:25 libhbris ?? 18:50:34 * We need a stable set of APIs for building the interface (libunity, indicator services, hud services, etc) 18:50:35 er wrong place so sorry 18:50:47 some of that exists, some doesn't 18:50:52 phablet has almost all that ^^ 18:50:57 Canonical resources have already been allocated to it 18:51:17 bobweaver: on the surface yes, I don't know how much is mocked and how much is subject to change 18:51:31 at the very least, the phablet's use of libunity is going to change when libunity changes 18:51:38 for lack of a better way to mark those, I'm going to put those as actions 18:51:44 but you all are digging your selfs deeper until the things that I have stated change then nothing can be done. OH HOW I whis that I could have talked about this at UDS 18:51:48 tgm4883: good idea 18:52:02 #action * We need the converged Unity foundation to support multiple formfactors (Canonical resources are workign on this, it doesn't exist yet) 18:52:02 * meetingology * We need the converged Unity foundation to support multiple formfactors (Canonical resources are workign on this, it doesn't exist yet) 18:52:23 #action * We need a stable set of APIs for building the interface (libunity, indicator services, hud services, etc) Canonical resources are already allocated to this 18:52:23 * meetingology * We need a stable set of APIs for building the interface (libunity, indicator services, hud services, etc) Canonical resources are already allocated to this 18:52:46 you all need to know how this is going to work on mer 18:52:52 Woops I mean mir 18:53:06 IMO, it's way to early to think about mir 18:53:11 #action we need to re-create the TV shell layout and components on top of Unity Next 18:53:11 * meetingology we need to re-create the TV shell layout and components on top of Unity Next 18:53:32 mhall119, we need remote control functionality of Unity Next 18:53:43 bobweaver: that's something that should be done in the common Unity Next code, there shouldn't be anything TV specific about it 18:53:45 tgm4883, agreed 18:53:52 if that is action then you are going to need all the libs that are used some are not open as of yet 18:54:07 mhall119, I'm leaving you to add the actions so we don't duplicate them 18:54:08 #action We need Remote control/navigation in the TV shell 18:54:08 * meetingology We need Remote control/navigation in the TV shell 18:54:18 mhall119: what the heck are you talking about 18:54:19 mhall119, we need a TV backend 18:54:23 that is bull 18:54:44 *action We need to make sure that all of the libs necessary to write apps and the shell are available and open source 18:54:46 how are you going to handle windowing how are you going to handle spread ect ect 18:54:54 alt+tab so many things 18:55:15 bobweaver, don't get buried in the details for now 18:55:18 *action we need to specify a data model for passing TV data between Scopes and the Shell 18:55:20 this was all done on X before 18:55:24 with qx11info 18:55:30 #action We need to make sure that all of the libs necessary to write apps and the shell are available and open source 18:55:30 * meetingology We need to make sure that all of the libs necessary to write apps and the shell are available and open source 18:55:35 #action we need to specify a data model for passing TV data between Scopes and the Shell 18:55:35 * meetingology we need to specify a data model for passing TV data between Scopes and the Shell 18:55:55 so if you are going to have all the things that the TV had then you are going to have to implant all that 18:56:02 #action we need scopes that pull TV content from various backends and services that use the datamodel and new API 18:56:02 * meetingology we need scopes that pull TV content from various backends and services that use the datamodel and new API 18:56:08 which I hear hush hush words that canonical is doing that 18:56:26 bobweaver: doing what? 18:56:38 there is no work for the community because canonical can change anything at any second giving the community the eff bomb 18:57:03 bobweaver: that's what I'm actively working to change 18:57:14 and jhodapp has been pushing for that to change too 18:57:21 * jhodapp will back mhall119 on that 18:57:35 mhall119: good for you I am glade to see that 18:57:39 so we're almost out of time, tgm4883 any more actions you want me to record? 18:58:03 nope 18:58:05 Listen the only reason that I am hear is because I have to tell you all what happend to me and how you can NOT make it happen to the next me 18:58:14 I'd like to end this by asking how you guys, on the community, want us in Canonical to work with you 18:58:34 you know what you all can do stop laughing at people 18:58:39 take them for real 18:59:01 bobweaver: none of us has been laughing at all about this 18:59:03 bobweaver, come on man, that's not fair 18:59:06 listen to them and dont clang them up to be just people make them part of the team 18:59:26 mhall119: should I take out the video of uds and saviq laughing ? 18:59:42 life is not fair 19:00:24 listen if you seee some like me again just tell them that they need to contact saviq and the other people that are too cool to show up to these meeting 19:00:27 I certainly hope that didn't happen, but if it did I apologize on their behalf 19:00:32 even though they said that they would make it 19:00:42 member at the last real UDS 19:00:44 bobweaver: Saviq doesn't work on TV anymore, he hasn't for some time 19:01:00 neither does jhodapp really, he's here on his own volition 19:01:15 that is not the point the point is these are the people that CONTROL things 19:01:25 you must make them work with people 19:01:26 bobweaver: no, they don't 19:01:32 bobweaver, no that's not right 19:01:35 if they dont want to then this is doomed 19:01:41 they've been yanked around and re-directed as much as you have (more probably) 19:01:50 * jhodapp can attest to that 19:01:54 if you can not work together then what do you have 19:01:55 ?> 19:02:00 people wasting there time 19:02:10 alright, our time is up, if anybody has any other questions, concerns or comments, you can find us in #ubuntu-tv 19:02:21 #endmeeting