Web   ·   Wiki   ·   Activities   ·   Blog   ·   Lists   ·   Chat   ·   Meeting   ·   Bugs   ·   Git   ·   Translate   ·   Archive   ·   People   ·   Donate

#sugar-meeting meeting, 2014-07-18 13:29:08

Minutes | Index | Today     Channels | Search | Join

All times shown according to UTC.

Time Nick Message
13:29 meeting Meeting started Fri Jul 18 13:29:08 2014 UTC. The chair is tch__. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:29 Useful Commands: #action #agreed #help #info #idea #link #topic #endmeeting
13:29 tch__ gonzalo__, of course,
13:29 well, basically I have 3 main topics...
13:30 the first one is about, "what" should we do for Sugar 0.102... or what _can_ we do actually...
13:30 the second is about "how",
13:30 so we can discuss about the development process itself, deadlines, releases, etc
13:30 and lastly, but not less important, infrastructure needs
13:31 there missing things like moving repos, to bring pootle back to life..
13:31 shall we start by the first one then :)?
13:31 walterbender tch__, 102 or 104?
13:31 tch__ 104 :)
13:32 walterbender #topic goals for 104
13:32 tch__ I would like to share my view on this topic a bit
13:32 walterbender +1
13:33 tch__ 102 was a lot about adding some new features and lots of fixes, and I think we did an awesome work,
13:33 but I have the feeling that we are missing some mid to long term goals
13:33 walterbender *lots* of fixes
13:34 tch__ to give an example
13:34 if someone asked me: what would _you_ like to focus for 104
13:34 id say, fix collaboration, for example
13:34 because is something it will take time
13:35 and is extremely important for Sugar
13:35 but that is just an example,
13:36 I think we need some mid to long term goals, to guide us during the next development fase
13:36 walterbender tch__, would you like us to chime in with our similar wishes?
13:36 tch__ yes
13:37 walterbender tch__, fixing collaboration is a great goal
13:37 tch__, some other things I would like to see:
13:37 (1) there is an interesting trade-off to be made between providing eye candy and leaving Sugar incomplete for our users to complete themselves
13:38 I'd like us to take a close look at how to make Sugar more end-user customizable without overspecifying its look and feel
13:39 (2) there are still lots of places where we are overly dependent on text
13:39 tch__ walterbender, low floor, high ceiling? :)
13:39 walterbender I'd like to review many of the views -- samdroid is doing this in the Journal -- to make them more graphical
13:40 (3) I'd like to move all of the CP section into apps and made them as a separate homeview
13:40 ignacio, has done some of this work
13:41 (4) I'd like some way of customizing Sugar in the same way we can "duplicate" and customize apps
13:43 gonzalo__ walterbender, done?
13:43 walterbender yes :)
13:43 tch__ walterbender,  I see a pattern, as I see, overall, it is to make sugar more encouraging for user-based to create (from apps to sugar itself)
13:43 walterbender tch__, yes
13:44 gonzalo__ tch__, my wishlist:
13:44 * solve performance issues: one part is land the journal optimizations
13:45 this could be important to make easier to small deployments using the old xo-1 use newer sugar
13:45 tch__ gonzalo__, +1, in Py we still have lots of XO1
13:45 gonzalo__ * implement text to speech in the toolkit: remove a lot of duplicated code in activities, and make them start faster
13:46 tch__, and i was reviewing the feature pages, and looks like there are some low hanging fruits to take
13:46 finished
13:47 tch__ taking notes
13:47 bashintosh, what about you matthew?
13:47 btw, who else is joining us?
13:48 llaske I'm here
13:48 tch__ llaske, hola!!
13:48 llaske Hi !
13:49 bashintosh tch__: I would like to think about printing to be honest - a feature that is really missing and that would be great in schools
13:49 tch__: screensharing is another thing that always fascinates me
13:50 walterbender recalls his conversations with bashintosh, about VNC server :)
13:51 bashintosh :(
13:51 I mean :)
13:51 now I use nx :)
13:51 tch__ bashintosh, printing is definitely something deployments have been asking for, since a LONG time
13:51 bashintosh, I feel in fault with Caacupe in that regard,
13:53 bashintosh, when you say screensharing, what would you do differently from using vnc activity?
13:53 bashintosh tch__: yes, in AU deployment we had many schools even willing to wait, but really keen to tell kids "yes, you can print your doc"
13:53 walterbender tch__, do you want to discuss the details now?
13:54 or still solicit ideas?
13:54 gonzalo__ probably is better don't start discussing details now
13:54 tch__ I think we should gather ideas for now, only the details to make sure we understand what people is suggesting
13:54 bashintosh tch__: it's probably some kind of better integration really, VNC activity feels somewhat not fully "embedded" in Sugar yet
13:54 tch__ bashintosh, ok
13:55 llaske, how about you :) with your amazing sugar to the web effort, I am sure you have your view !
13:55 llaske Sure.
13:55 You now my mission, my crusade:
13:55 Sugar everywhere: laptop, tablet, smartphone, browser, …
13:56 Of course it's a far target
13:56 But my wish is to ensure that every new Sugar feature could be think to be ported easily on a other platform.
13:56 When you told about collaboration for example, I think we need to fix (rethink ?) it to ensure we could collaborate with something in a browser.
13:57 So, may be a web socket collaboration based API could be an interesting solution here.
13:57 I would love too to have a roadmap about migration of main Sugar activity into web activities
13:58 gonzalo__ websocket libraries fro python are available now
13:58 llaske It will ensure us a way to prepare the future
13:58 walterbender llaske, "main Sugar activity"?
13:58 tch__ llaske, I still thinking about it, from one POV i feel like we should completely re-work collab, but then, I see we have some many great activities depending on the current model.. and being realistically we won't be able to port all of them to something new
13:59 llaske we could think to a portability layer
13:59 tch__ llaske, what I am trying to say is that we need think it this carefully, and being very conscious of the pros and cons for whatever we decide
13:59 llaske, that could be an option,
14:00 llaske I don't want to rethink Sugar from scratch. My point of view is that we need to think about a transition layer
14:00 and sugar-web is a good way to think about this transition
14:00 all my work on Sugarizer is based on this idea
14:01 main Sugar activity = core Sugar activities
14:01 Write, Paint, TurtleArt, ...
14:01 walterbender llaske, gotit :)
14:01 llaske, some progress on Turtle :)
14:02 llaske TurtleJS is great. I would like to have more time to contribute on this
14:02 I've talked to Bert about EToys too
14:03 That's all for my wish. Sorry to be disturbing :-)
14:03 tch__ llaske, great stuff :)
14:03 walterbender anyone else with ideas for tch__ 's topic #1?
14:03 gonzalo__ llaske, good points
14:04 tch__ In general terms, I think we 3 general goals (1) make Sugar core ideas work (collab, customizations), (2) improve performance and (3) help deployments with the day to day needs
14:05 as always, is not possible to do everything, but we can have these goal in mind to prioritize
14:07 another important things is to have people (invidually or in groups) wanting to take the responsibility for specific goals
14:08 not sure if you would like to discuss that now? ;)
14:08 probably not, since it requires talking about details
14:08 walterbender thinks it is interesting that no one brought up assessment :)
14:08 tch__ but we could assign working groups?
14:08 walterbender tch__, "how" is your second topic :P
14:08 are we ready for how?
14:09 gonzalo__ walterbender, 0.102 assesment?
14:09 walterbender gonzalo__, no... I mean student assessment
14:09 Harvest ++
14:09 gonzalo__ walterbender, harvest == statistics?
14:09 tch__ to bad sameer is not around!
14:10 walterbender seems as though there is lots of effort in that space, especially in the school-server space
14:10 and there are lots of half-baked ideas
14:10 but maybe Sugar could take the lead in providing a positive model for the learner
14:11 gonzalo__ may be if we have a better implementation of collaboration in 0.104, we can focus in sugar <-> schoolserver integration n 0.106
14:12 tch__ it would depend on the extend for what "collab" stand for
14:12 gonzalo__ in general we have a lot of starts with services (or cloud services) but not a well rounded solution
14:12 tch__ I did not think about that way
14:12 walterbender brb
14:13 tch__ gonzalo__, the webservice effort is still not very mature,
14:13 gonzalo__ i agree
14:13 tch__ gonzalo__, and we still don't have a strong use case
14:14 gonzalo__ tch__, what about your journal integration stuff?
14:14 tch__ gonzalo__, yeah, I think the journal backend stuff could be improved too
14:14 gonzalo__, to allow different backends
14:14 gonzalo__, the current "model.py" does not make this easy
14:15 gonzalo__, but that is a big task too, and we need a real use case to understand what we need
14:16 gonzalo__ tch__, finishing the particular features. i would like we follow more the feature process in 0.104
14:16 tch__ gonzalo__, you mean the protocol for adding features?
14:17 gonzalo__ tch__, in 0.102, we published the feature pages, and started worked without discussing
14:17 tch__ gonzalo__, yes, and a people did not like that..
14:17 gonzalo__, it was chaotic too, from the development POV
14:17 gonzalo__ count as people :)
14:17 tch__ wonders (haha)
14:18 thats why this kind of meetings are important
14:18 there will always be emergent needs
14:18 we but need some guidance
14:18 gonzalo__ tch__, yes, and we need these meeting periodicallly
14:18 tch__ something to aim for
14:18 I don't mind doing this weekly or every 15 days
14:18 gonzalo__ maybe once by month is enugh
14:19 or two :)
14:19 tch__ it will depend on how we schedule the release phases
14:19 two months could be too late for some discussiongs
14:19 it should happen in ML
14:19 but taking "face to face" sometimes helps
14:20 gonzalo__ tch__, well, that is a qood point, what about the schedule?
14:20 tch__ gonzalo__, that is the next topic :0
14:20 :)
14:20 bashintosh suggests more meetings == shorter meetings
14:20 tch__ is there anyone else to wants to step in?
14:22 ok, lets move on
14:22 #topic roadmap "the how"
14:23 I think 0.100, was the last time we had formal roadmap like http://wiki.sugarlabs.org/go/0.100/Roadmap
14:24 persoanlly, I do like the current structure,
14:24 is important to have phases and deadlines
14:24 walterbender +1
14:24 and it forces us to be realistic
14:24 gonzalo__ +1
14:25 bashintosh +1
14:25 tch__ but, is there anything we would like to suggest?
14:25 during 0.102 we had many delays
14:25 and the string freeze did not make much sense, because pootle was dead
14:26 I am trying to imagine if there are things to reoconsider like.. how long should this process be?
14:26 we keep doing 6 months releases?
14:27 walterbender tch__, I think the merit of 6 months remains
14:27 it is commiserate with Fedora and Ubuntu
14:28 and it is not so long that we get overambitious
14:28 gonzalo__ sadly, we are disaligned now with fedora and gnome
14:28 walterbender gonzalo__, we could resync
14:28 gonzalo__ gnome 3.14 will be released on Sept 24
14:29 and F21 on Oct 14
14:29 tch__ gonzalo__, we might fall short for that one,
14:29 for the goals we mentioned at least,
14:29 we could re-release 0.102 until then :P
14:29 haha
14:29 ie., by bringing pootle back to like and fixing strings
14:30 gonzalo__ yes
14:30 tch__ s/like/life
14:30 walterbender tch__ we cannot achieve all of our medium- and long-term goals in one release cycle
14:30 tch__ walterbender, true
14:30 walterbender but it is important to keep those goals in mind when pursuing short-term goals
14:30 gonzalo__ if we define a 6 months cycle, could finish on December
14:31 tch__ walterbender, 6 months should be sufficient for things like printing or making the journal more responsive
14:31 it really depends on how big is the bite
14:32 walterbender tch__,  we should consider a code sprint per release cycle for one or two bigger goals
14:32 tch__ walterbender, +100
14:32 walterbender tch__, we talked about Oct with rgs looking at Collaboration
14:32 tch__ walterbender, yes, that will be a big push
14:32 gonzalo__ walterbender, tch__, we have the event in Uy (Sept) too
14:33 tch__ yes, we can put a goal for each one
14:33 collab is already for oct in SF
14:33 walterbender let's take advantage of these opportunitiies
14:33 what should we target for .UY?
14:35 tch__ I pushed forward collab (i know llaske would like it), maybe gonzalo__ bashintosh should propose it
14:36 llaske :-)
14:36 gonzalo__ tch__, i would like in the performance issues
14:36 tch__ gonzalo__, think we can convince manuq to participate?
14:36 gonzalo__, he was part of your the first session
14:36 gonzalo__ tch__, i hope
14:36 tch__ i agree on performance,
14:37 even if we _only_ get collab to work, and improve performance for 104
14:37 it would be a GREAT release already
14:38 bashintosh tch__: really agree
14:38 walterbender +1
14:39 tch__ so, regarding the the relase deadline, does december sounds reasonable?
14:39 needs to study fedora and gnome deadlines
14:40 gonzalo__ http://fedoraproject.org/wiki/[…]eases/20/Schedule
14:40 http://fedoraproject.org/wiki/[…]eases/21/Schedule
14:40 https://wiki.gnome.org/ThreePointThirteen
14:40 the first is wrong
14:41 tch__ gonzalo__, thanks
14:42 gp94 <gp94!uid25529@gateway/web/irccloud.com/x-ncmseojpwgawbcst> has joined #sugar-meeting
14:42 tch__ we are definitely late for these, but what we could do is to make 104 cycle longer (?)
14:42 haha shorter
14:43 satellit FYI f21 sugar 0.102 (soas) is branched and rawhide is now f22
14:45 tch__ satellit, thanks for the info,
14:45 gonzalo__ tch__, if we want work on collab, may be is better make it longer
14:45 tch__ I really need to understand these cycles
14:45 gonzalo__ tch__, if not, we can make a shorter cycle, and work only in easier features
14:45 bashintosh tch__: not sure if of any interest here but collaboration over scatternet/piconet could be an interesting step in a more reliable "low level" platform
14:45 puneet_kaur <puneet_kaur!~puneet_ka@> has joined #sugar-meeting
14:46 walterbender maybe we can have an interim release with some bug fixes to lock us back into the cycle
14:46 bashintosh tch__: take note perhaps then we talk..
14:46 walterbender bashintosh, we have been looking at a few other possibities too
14:46 bashintosh, we need to get Rangan to send you to our Oct meeting in SF :)
14:47 gonzalo__ tch__, maybe do a short cycle and finish in Oct 30?
14:47 bashintosh walterbender: please? :)
14:47 walterbender gonzalo__, we have a lot queued up for review we could land in Oct
14:47 tch__ walterbender, you mean 104 for 30 Oct?
14:48 gonzalo__, ^
14:48 honestly, I need to look carefully at the fedora and gnome stuff, to understand exaclty when is the best moment to have these tarballs ready
14:48 gonzalo__ walterbender, we can select a part for this cycle
14:49 tch__ maybe satellit can help me :)
14:49 satellit tch_ I do a lot of the #fedora-qa testing...glad to help if I can
14:50 tch__ we don't need to decide today, lets use some days to think whats best?
14:50 walterbender tch__, we used to try to land some weeks in advance so we had time for bug fixes
14:51 gonzalo__ tch__, we can discuss this on the mailing list
14:51 tch__ yeah, I will send a summary of the topics, so we can continue there
14:51 I think we expoosed the main issues already
14:51 lets move on to the next topic?
14:52 gonzalo__ ok
14:53 walterbender #topic infrastructure
14:53 tch__ pootle is dead
14:53 seriously speaking, pootle was in the middle of a migration process...
14:53 but
14:53 walterbender is it just a server issue or something deeper?
14:53 tch__ it was never finishes
14:54 bernie sent me a lot of stuff
14:54 bashintosh tch__: define migration?
14:54 tch__ starting from "update the new VM for hosting pootle"
14:54 bashintosh tch__: to what, from where, on what? :)
14:54 walterbender tch__, maybe we need a mini sprint with cjl, you, me, and bernie?
14:55 tch__ we had pootle in server called "sunjammer", then someone created  a new one exclusively for pootle, and started moving the platform there
14:55 for some reason, the pootle in sunjammer went down
14:55 and the new one neve rgot up
14:55 at least is easy to say we are stuck in the middle
14:55 haha
14:55 walterbender, +1
14:56 walterbender tch__,  should not be rock science
14:56 it is just software after all :)
14:56 rock^rocket
14:56 tch__ I know, but we need to make sure we know what we are doing,
14:56 like where are the backups
14:57 and how to integrate pootle with our repos etc
14:57 bashintosh tch__: if not the right time to get into details, you know where to find me if I can help :)
14:57 tch__ there was a lot of stuff in the wiki, noisy
14:57 walterbender, gonzalo__ bashintosh can join our sysadmin team :)
14:57 gonzalo__ rock science is geology?
14:57 tch__ maybe there is a science for rock stars
14:58 gonzalo__ tch__, i did some help with the pootle server in the past
14:58 tch__, is not the thing I enjoyed most in my live anyway
14:59 bashintosh walterbender: somehow it is not just software: a hard drive could have gone down, a NIC could have died etc.. :) :)
14:59 tch__ gonzalo__, I will call for a meeting regarding this specific topic for next week
14:59 bashintosh, I am pretty sure the job was not finished,
14:59 bashintosh, it could range from, there backup was never moved to the new server, to simply turn on ngix,
14:59 we dont know
14:59 haha
14:59 bashintosh tch__: is a VM of physical?
14:59 tch__ bashintosh, VM
15:00 bernie, should be able to tell
15:00 bashintosh, I will include you in the loop
15:00 bashintosh thx
15:00 puneet_kaur has quit IRC
15:00 tch__ in terms of infrastructure, anything else you guys can think of?
15:00 i only worried about pootle right now
15:01 (specially for the up-coming sugar in Guarani translation)
15:01 gonzalo__ i think is the most important piece right now
15:01 tch__ ;)
15:01 agreed,
15:01 gonzalo__ what about the work from sam on aslo2?
15:02 i am worried is not guided
15:02 tch__ I haven't have the time to follow it
15:02 gonzalo__ ahh! and talking about that
15:02 i remembered now, we lost our design team probably now
15:02 tch__ we have discourse GSoC project too,
15:03 gonzalo__ tch__, that will be a problem
15:03 tch__ gonzalo__, yes, that is REALLY sad
15:04 people with general + GUI design skills, and, the understanding  of sugar
15:05 gonzalo__ the only solution I see right now, is we take a collective responsibility right now, but would be good if we can involve to Paul Cotton by example
15:05 tch__ paul would be very helpful, we can't speak in his behalf though,
15:05 gonzalo__, can you talk to him?
15:05 walterbender paul has taken some interest
15:05 gonzalo__ tch__, yes, I can
15:06 walterbender I have pointed him at samdroid's work
15:06 and he is involved with the GSoC project
15:06 tch__ we gotta make the best with what have now :)
15:06 gonzalo__ about missing people, the next one who leave, please be open about that, I don't know how, bu we need change this
15:07 tch__ gonzalo__, not sure I understand !
15:07 gonzalo__, you mean to consider the fact that people come and go?
15:08 gonzalo__ tch__, i think if I leave a project where I was deeply involved, I would say goodbay
15:08 tch__ well, not sure what we can do about that...
15:08 gonzalo__ maybe we should elaborate about the responsabilities of maintainers
15:08 tch__ put a timer?
15:08 haha
15:08 gonzalo__ a watchdog :)
15:09 tch__ would be more effective
15:09 yes, I mean, this is a communication problem
15:09 if we had regular meetings
15:09 it would be diferently
15:09 gonzalo__ more seriously, a page "About maintainership" could help
15:10 yes, regular meetings would be a great help I think
15:10 tch__ ok, i will include that in the summary
15:10 so we can seriously consider these meetings,
15:11 gonzalo__ tch__, we need think about the health of the project
15:11 tch__ yes, that is a broader discussion I think, the project is people who make it at the end,
15:11 we need to make we sure WE stay healthy too
15:11 aha
15:12 gonzalo__ true
15:12 tch__ it would be really great if people could lend a couple hours per week at least with guidance for example
15:13 puneet_kaur <puneet_kaur!~puneet_ka@> has joined #sugar-meeting
15:13 tch__ people like dsd and other former devels
15:14 gonzalo__ :)
15:14 tch__ is a just a happy wish haha
15:14 anyway, I think we are done for today
15:14 gonzalo__ +1, a good start
15:15 tch__ I will make a summary of the topics discussed and send it to the ML
15:15 thanks everyone for coming :)
15:15 gonzalo__ great have llaske and satellit here
15:15 tch__ of course :)
15:15 llaske great to be here :-)
15:15 tch__ gonzalo__, walterbender bashintosh llaske satellit many thanks!
15:15 gonzalo__ i hope we have more people in next meetings
15:16 bashintosh pleasure!
15:16 tch__ #endmeeting
15:16 meeting Meeting ended Fri Jul 18 15:16:41 2014 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot. (v 0.1.4)
15:16 Minutes: http://meeting.sugarlabs.org/s[…]-18T13:29:08.html
15:16 Log:     http://meeting.sugarlabs.org/s[…]14-07-18T13:29:08

Minutes | Index | Today     Channels | Search | Join

Powered by ilbot/Modified.