Re: [Sugar-devel] Journal files on regular desktop
On Tue, Jun 2, 2009 at 3:16 AM, Tomeu Vizoso wrote: > About a hierarchical file browser in Sugar, I think we need it anyway > for removable devices, but I'm not sure where we would put such a > window and how it would look like. > > Regards, > > Tomeu I have been using the character mode file browser Midnight Commander within Terminal. It's more than children need, but we could do a cut down version. http://www.midnight-commander.org/ http://wiki.sugarlabs.org/go/Activities#Midnight_Commander -- Silent Thunder (默雷/धर्ममेघशब्दगर्ज/دھرممیگھشبدگر ج) is my name And Children are my nation. The Cosmos is my dwelling place, The Truth my destination. http://earthtreasury.org/worknet (Edward Mokurai Cherlin) ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] I'm looking for a tree...
I am looking for a fast data structure with the following properties: Maintains an indexed list of arbitrary, non-ordered objects (like a python List or C array) Allows fast: Insertion at any location Deletion at any location Lookup of an object by its index Reverse lookup, to determine the index of an object Python's List has O(1) lookup, but O(N) insert, delete, and reverse-lookup. To make reverse lookup O(1) I could maintain a separate Dict mapping objects to indices, but this would cost an additional O(N) on every insertion and deletion. A linked list has O(1) insertion and deletion, but O(N) lookup and O(N) reverse lookup. I could maintain a separate Dict for the forward and reverse mappings, but this would cost O(N) on every insertion and deletion. A standard self-balancing tree cannot be used because the objects are not ordered, and self-balancing trees require ordered keys. I could use the index of an object as the sort key, but then insertion and deletion are O(N) because all subsequent keys must be altered. I could fabricate new sort keys to ensure that insertions occur at the desired location, but then the length of the keys will grow like O(N), making all operations at least O(N). I feel like there should be some kind of standard tree-like data structure that meets my requirements, but I can't find one. Do you know of one? Am I on a unicorn hunt? --Ben signature.asc Description: OpenPGP digital signature ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [GSoC] SSB creator
On Sat, Jun 6, 2009 at 9:57 PM, Lucian Branescu wrote: > For my project, I will extend Browse with the ability to create SSBs. > Read more here http://honeyweb.wordpress.com/2009/06/06/the-user-experience/ > > I have made a small script that is triggered by a button in the Browse > toolbar. It also works from Terminal. > > Later on this script will do more things, but I'd like some feedback > about it as early as possible. Hi Lucian, 2 things: 1) rainbow should prevent you from creating new activities in ~/Activities, so I don't think that approach would work on an XO with an OLPC build. Since I think what you want is to create an activity, zip it, and add it to the journal, you could probably do something like this: import tempfile ssb_path = tempfile.mkdtmp(dir=activity.get_activity_root()) 2) the bundle_id you create probably should begin with org.laptop , as that could cause collisions with other activities. maybe org.sugarlabs.ssb ? nice start so far! bobby ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] [GSoC] SSB creator
For my project, I will extend Browse with the ability to create SSBs. Read more here http://honeyweb.wordpress.com/2009/06/06/the-user-experience/ I have made a small script that is triggered by a button in the Browse toolbar. It also works from Terminal. Later on this script will do more things, but I'd like some feedback about it as early as possible. createssb.py Description: Binary data ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] OT: determining memory usage of short-lived processes
On Sat, Jun 6, 2009 at 9:40 PM, Sascha Silbe wrote: > I've examined memory usage of long running processes (i.e. daemons and > applications) in the past, no problems. If in the past you've used top, there's a new and more accurate way of measuring memory usage. Recent kernels export the 'smaps' of every process, and scripts such as 'ps_mem.py' provide very good summaries of that info. Perhaps you knew it already -- it's a good tool worthy of promotion so... > But for my VCS comparison I need to determine the peak memory usage of all > child processes (combined), which are rather short-lived. What's the best > way to do that? Performance is an issue as the benchmark already takes 13h > to run on a sample set of 100 Project Gutenberg files (originally I had a > sample size of ~800 files). That probably rules out valgrind (AFAIK it > incurs quite a performance penalty)... Run the whole thing under /usr/bin/time, which is different from the shell's 'time' built-in. Used it very often in git development to assess peak mem usage. The report is actually about the pagefaults, which for relatively short runs relate almost linearly to mem usage. For long lived processes that may release memory, and then allocate memory again, the relationship is a bit less clear. hth, m -- martin.langh...@gmail.com mar...@laptop.org -- School Server Architect - ask interesting questions - don't get distracted with shiny stuff - working code first - http://wiki.laptop.org/go/User:Martinlanghoff ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Can't compile mozilla on sugar-jhbuild
Sascha Silbe writes: > For most modules the directory name is the same as the module name (with > version appended in case of tarballs). For xulrunner it's different for > historical reasons. I had the hope we could get rid of xulrunner > completely (as a module of sugar-jhbuild), but as this has proven false > I'm going to fix the name soon. Thanks for the explanations. -- Bastien ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [math4] Activity Awards, Activity Alerts, Frame Alerts
Yeah, somehow people have very clear memories of their today seemingly innocent, but personally quite serious at the time, achievements/awards/stars/buttons, or whatever it was. The journal seems ilke the clear place to store this information, and it should be done in such a way that it can be easily printed so as to put on the classroom wall, the family wall, or even just one's personal wall. The starting and ending of activities are awards in an of themselves... so these events can be triggered. For some activities, I'd imagine starting and ending an activity within a time frame (while getting things correct) would bring up yet another type of award... or being the last to complete could give you a thoroughness award... The ideas and possibilities are somehow endless here, and its really about creating the framework so we can come up with some fun concepts that can then be refined in real use cases. I'd imagine that being finished with an activity would trigger something like a beep (or some other sound, perhaps a personalised one... [that is another area that I've been thinking about a lot recently too, more personalisation of the user [[an avatar, personalised sound for event triggers, mood emblem, best friends, pet, etc]], which can be shared or even come from Moodle/Mahara which already has a lot of this stuff.] So going back to the original question, how to know when someone has finished... Well if every user can personalise their event trigger (like your mobile phone ringer) then u'd get an idea of who is finishing what when., or starting for that matter. kind Regards, David (nubae) On Sat, Jun 6, 2009 at 10:34 PM, Sean DALY wrote: > I don't know if this is related from a technical standpoint, but I was > wondering about the possibility of a mechanism for an Activity to > signal to a teacher (or the Neighborhood, or a Friend) when an > Activity task is completed. As a way of simplifying identification of > students who need more attention on a task than their peers. > > Back in the Stone Age when I was in school, this was accomplished > during a quiz or writing task by the leading students ostentatiously > throwing their pens down onto their desks and looking around smugly. > Maybe things have changed in the 21st century? > > Dell wants to differentiate their K-8 education netbook offer > (Latitude 2100) with a white LED bar in the lid which can light up via > a software call. Although in marketing materials they describe its > default setting as lighting up when there is any network activity > (they imply Internet access during study time), their documentation > states that any software can hook into it. I also read that if the > idea takes off, they will consider adding other LED colors to the bar, > so as to indicate simultaneously for example wireless activity / > Internet activity / application task state, or file transfer / school > filter hit / low battery condition. In each scenario, allowing the > teacher to see what is happening at a glance. > > Would such status indicators work for Sugar? Should they be visible > only to the teacher, or to everyone? (I'd vote for the teacher - it's > really a classroom context feature, unnecessary while the Learner is > enjoying her computer outside school hours). Could a Learner achieving > an Activity milestone show that right away, in such a way the teacher > could see it? Or notice it if the Learner achieved it outside school > hours? > > In 1968, when I was six, I wrote a little 4-line poem about the > astronauts going up into orbit. My first-grade teacher, Mrs. Menzi, > typed it up and put a gold star sticker on it, and a gold star sticker > on my forehead too. Like a star in outer space! I'm still proud of > that :D So yes, I think an award system could be a useful feature. > > Sean > > > > > On Sat, Jun 6, 2009 at 9:50 PM, David Van Assche > wrote: > > Yeah, due to the lack of resources and time, we didn't get to finalise > the > > end vision for such an award based system, but the idea came to me due > how > > successful the collection of these is in computer games. As Gary pointed > out > > to me, there are even games where the sole purpose is to unlock awards > > (which could be as silly as, dying by drowning 42 times.) The concept in > > gaming, as adopted by nintendo, sony, and Microsoft, has been a real > success > > and a means by which to get users 'hooked' to achieving them. If the same > > thing can be translated to Sugar, we can have a system which by its > nature > > pushes users to delve deep into the guts of activities to 'discover' > them, > > if you will, all the while getting recognition for that discovery. The > > coupling of this with collaboration brings together a whole world of > > possibilities, many of which have caused measurable excitement within > myself > > and Gary, as well as others listening/discussing the ideas. > > > > The inital activity we wanted to do this with was a quiz based activity, > > where the tea
Re: [Sugar-devel] [math4] Activity Awards, Activity Alerts, Frame Alerts
I don't know if this is related from a technical standpoint, but I was wondering about the possibility of a mechanism for an Activity to signal to a teacher (or the Neighborhood, or a Friend) when an Activity task is completed. As a way of simplifying identification of students who need more attention on a task than their peers. Back in the Stone Age when I was in school, this was accomplished during a quiz or writing task by the leading students ostentatiously throwing their pens down onto their desks and looking around smugly. Maybe things have changed in the 21st century? Dell wants to differentiate their K-8 education netbook offer (Latitude 2100) with a white LED bar in the lid which can light up via a software call. Although in marketing materials they describe its default setting as lighting up when there is any network activity (they imply Internet access during study time), their documentation states that any software can hook into it. I also read that if the idea takes off, they will consider adding other LED colors to the bar, so as to indicate simultaneously for example wireless activity / Internet activity / application task state, or file transfer / school filter hit / low battery condition. In each scenario, allowing the teacher to see what is happening at a glance. Would such status indicators work for Sugar? Should they be visible only to the teacher, or to everyone? (I'd vote for the teacher - it's really a classroom context feature, unnecessary while the Learner is enjoying her computer outside school hours). Could a Learner achieving an Activity milestone show that right away, in such a way the teacher could see it? Or notice it if the Learner achieved it outside school hours? In 1968, when I was six, I wrote a little 4-line poem about the astronauts going up into orbit. My first-grade teacher, Mrs. Menzi, typed it up and put a gold star sticker on it, and a gold star sticker on my forehead too. Like a star in outer space! I'm still proud of that :D So yes, I think an award system could be a useful feature. Sean On Sat, Jun 6, 2009 at 9:50 PM, David Van Assche wrote: > Yeah, due to the lack of resources and time, we didn't get to finalise the > end vision for such an award based system, but the idea came to me due how > successful the collection of these is in computer games. As Gary pointed out > to me, there are even games where the sole purpose is to unlock awards > (which could be as silly as, dying by drowning 42 times.) The concept in > gaming, as adopted by nintendo, sony, and Microsoft, has been a real success > and a means by which to get users 'hooked' to achieving them. If the same > thing can be translated to Sugar, we can have a system which by its nature > pushes users to delve deep into the guts of activities to 'discover' them, > if you will, all the while getting recognition for that discovery. The > coupling of this with collaboration brings together a whole world of > possibilities, many of which have caused measurable excitement within myself > and Gary, as well as others listening/discussing the ideas. > > The inital activity we wanted to do this with was a quiz based activity, > where the teacher puts labels on the different parts of a picture s/he has > uploaded and lets the user then fill in the labels. We can measure a whole > host of things, such as, were the labels filled in while connected to the > internet (they could have used wikipedia to gather the information), and > should the user be encouraged to do that or not... > > What I guess I'm pointing out is, we need to delve deep into how this should > all work, but the initial concept rocks, and we should try and either adapt > an existing activity (Gary suggested using his moon activity to name the > various parts of it) or make an easy framwork that would allow a teacher to > upload a picture and then tag the various points in the picture together > with the possible answers. Either is fine, though the later is more > desirable as it would allow us to explore the possibilities a little more > deeply, and would allow for the immediate creation of content there for, > which could be easily stored in moodle or an activity meta bundle, or > whatever... So lets decide how we move forwards on this and get to it > then... > > kind regards, > David Van Assche > > On Sat, Jun 6, 2009 at 3:35 AM, Gary C Martin wrote: >> >> Hi Frederick, >> >> On 6 Jun 2009, at 00:30, Frederick Grose wrote: >> >> > On Fri, Jun 5, 2009 at 6:32 PM, Gary C Martin >> > wrote: >> > ... >> > >> > FWIW, David Van Assche raised some interesting Activity ideas at >> > SugarCamp Paris and I'm interested/active in getting us to at lease >> > 'demo' state in the Sugar 0.86 release timeframe. The idea is to focus >> > on an 'awards' mechanism/style to encourage exploration and provide >> > (sometimes) unexpected rewards. Idea is that Activity authors can >> > define a range of badges/medals/icons for certain behaviours/ >> > accompli
Re: [Sugar-devel] [math4] Activity Awards, Activity Alerts, Frame Alerts
Yeah, due to the lack of resources and time, we didn't get to finalise the end vision for such an award based system, but the idea came to me due how successful the collection of these is in computer games. As Gary pointed out to me, there are even games where the sole purpose is to unlock awards (which could be as silly as, dying by drowning 42 times.) The concept in gaming, as adopted by nintendo, sony, and Microsoft, has been a real success and a means by which to get users 'hooked' to achieving them. If the same thing can be translated to Sugar, we can have a system which by its nature pushes users to delve deep into the guts of activities to 'discover' them, if you will, all the while getting recognition for that discovery. The coupling of this with collaboration brings together a whole world of possibilities, many of which have caused measurable excitement within myself and Gary, as well as others listening/discussing the ideas. The inital activity we wanted to do this with was a quiz based activity, where the teacher puts labels on the different parts of a picture s/he has uploaded and lets the user then fill in the labels. We can measure a whole host of things, such as, were the labels filled in while connected to the internet (they could have used wikipedia to gather the information), and should the user be encouraged to do that or not... What I guess I'm pointing out is, we need to delve deep into how this should all work, but the initial concept rocks, and we should try and either adapt an existing activity (Gary suggested using his moon activity to name the various parts of it) or make an easy framwork that would allow a teacher to upload a picture and then tag the various points in the picture together with the possible answers. Either is fine, though the later is more desirable as it would allow us to explore the possibilities a little more deeply, and would allow for the immediate creation of content there for, which could be easily stored in moodle or an activity meta bundle, or whatever... So lets decide how we move forwards on this and get to it then... kind regards, David Van Assche On Sat, Jun 6, 2009 at 3:35 AM, Gary C Martin wrote: > Hi Frederick, > > On 6 Jun 2009, at 00:30, Frederick Grose wrote: > > > On Fri, Jun 5, 2009 at 6:32 PM, Gary C Martin > > wrote: > > ... > > > > FWIW, David Van Assche raised some interesting Activity ideas at > > SugarCamp Paris and I'm interested/active in getting us to at lease > > 'demo' state in the Sugar 0.86 release timeframe. The idea is to focus > > on an 'awards' mechanism/style to encourage exploration and provide > > (sometimes) unexpected rewards. Idea is that Activity authors can > > define a range of badges/medals/icons for certain behaviours/ > > accomplishment in an attempt to get students to dig deeper (mix of > > 'easter eggs' and specific goals). It's mainly Activity side work (a > > demo activity to start with) but perhaps could find a home in the > > Journal (through an ability of Activity to set some private entry tag > > and for Journal to display that in a user appealing graphical form). > > > > Even for something as hard to measure as the Write Activity, there > > could be 'awards' (hidden or hinted at) for things like "found 10 or > > more collaborators for one document", "gained at least 100 words each > > from 5 or more collaborators", "wrote more than 1,000 words", "you > > used the word entomology!". The idea is many would be hidden > > ("surprise, you did something cool!") and that some initial more > > obvious and visible 'awards' would hint that others were there for > > discovery. > > > > Regards, > > --Gary > > > > P.S. Mechanisms for 'awards' could hook into services like Moodle, the > > Journal, or via collaboration (so perhaps a shared Write session would > > show awards gained by the collaborators). Having a view to show all > > Activity Awards would also be a good driver (could be an activity, or > > ideally at some point part of Journal). The general idea for awards > > drifts in from the gaming environment, where awards are used to > > increase re-playability and tempt folks to try some other possible > > path. > > ... > > > > Nice concept. > > > > Some design and code integration with Activity Alerts, > http://wiki.sugarlabs.org/go/Development_Team/Almanac/sugar.graphics.alert > > , > > and Frame alerts, > http://wiki.sugarlabs.org/go/Design_Team/Designs/Frame#12 > > , http://wiki.sugarlabs.org/go/Design_Team/Designs/Frame#13, etc. > > may be appropriate. > > Yep, I imagine it's being up to the activity how it reveals a new > 'award' being reached, but it could be as simple as showing an > activity alert message (perhaps as a default design guideline for > awards, if it is accepted), or some fancy splash/animation if the > activity deems it appropriate. > > > http://dev.sugarlabs.org/ticket/722 has some discussion. I would > > like to see more noticeable messages for chat invitation alerts, for > > examp
[Sugar-devel] OT: determining memory usage of short-lived processes
Sorry for the OT post, at least it's for my GSoC project. :) I've examined memory usage of long running processes (i.e. daemons and applications) in the past, no problems. But for my VCS comparison I need to determine the peak memory usage of all child processes (combined), which are rather short-lived. What's the best way to do that? Performance is an issue as the benchmark already takes 13h to run on a sample set of 100 Project Gutenberg files (originally I had a sample size of ~800 files). That probably rules out valgrind (AFAIK it incurs quite a performance penalty)... CU Sascha -- http://sascha.silbe.org/ http://www.infra-silbe.de/ signature.asc Description: Digital signature ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Journal activity release 101 not tagged in Git
Maybe if you list some of the problems you are running into, we can help, as we must have gone through the same issues in order to get it working under openSUSE, Mandriva, et al. kind regards, David (nubae) On Sat, Jun 6, 2009 at 5:40 PM, Jonas Smedegaard wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: RIPEMD160 > > On Sat, Jun 06, 2009 at 04:52:08PM +0200, David Van Assche wrote: > >And though packaging debian is quite different from doing rpms, I just > >finished getting all activities working for openSUSE. There were lots > >of little glitches that prevented activities from working,so if u run > >into any issues with particular activities, ping me and I might be able > >to tell u how to get stuff working. Activities like infoslicer, colors, > >browse, record, and Maze come to mind... Generally I've been working > >directly off the latest source in gitorious unless there was a required > >downgrade to another source (storybuilder)... > > Thanks. > > Not sure if you know, but I have been packaging for some time (since > Sugar 0.81) and Sugar 0.82 worked correctly. > > Also, the current trouble is with core Sugar environment, not > activities. > > That said, I certainly appreciate the offer, and generally I like the > growth in teams attacking the Sugar code from different angles. That > really helps improve code quality for all of us. :-D > > > - Jonas > > - -- > * Jonas Smedegaard - idealist og Internet-arkitekt > * Tlf.: +45 40843136 Website: http://dr.jones.dk/ > > [x] quote me freely [ ] ask before reusing [ ] keep private > -BEGIN PGP SIGNATURE- > Version: GnuPG v1.4.9 (GNU/Linux) > > iEYEAREDAAYFAkoqjf4ACgkQn7DbMsAkQLg41ACeLapB6MhAX4GK2TuJpS7rz5Sb > NkAAn2PmymU7T3ZiOe8IV9TmEjTm1fAy > =OSXG > -END PGP SIGNATURE- > ___ > Sugar-devel mailing list > Sugar-devel@lists.sugarlabs.org > http://lists.sugarlabs.org/listinfo/sugar-devel > ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Can't compile mozilla on sugar-jhbuild
-BEGIN PGP SIGNED MESSAGE- Hash: RIPEMD160 On Sat, Jun 06, 2009 at 06:11:11PM +0200, Sascha Silbe wrote: > On Sat, Jun 06, 2009 at 05:55:22PM +0200, Jonas Smedegaard wrote: > >> You wrote in ticket 137 that *hulahop* is broken in *Debian*: > Ah, sorry if that wasn't clear enough. I was talking about hulahop in > sugar-jhbuild, which doesn't build against the Debian-shipped xulrunner. > >> (hint: maybe _I_ can help _you_ here - if what breaks is jhbuild and I >> might know about some build options required for proper linking >> against Debian xulrunner) > What I need is some recipe that works on Debian _and_ other distros, > because sugar-jhbuild doesn't support setting different options and/or > patches depending on the distro (the sysdeps stuff was introduced by us, > upstream jhbuild doesn't detect the distribution). Ah, ok. I guess I can't help you then. I can suggest that you file a bugreport against Debian xulrunner, and not the bugnumber in the Sugar ticket system when done. If you believe the Debian package is in error, that is. Else you should obviously do the same with both Ubuntu and Fedora. Or with jhbuild, if you (like me) suspect that build options are there because the world is not a monoculture). Kind regards, - Jonas - -- * Jonas Smedegaard - idealist og Internet-arkitekt * Tlf.: +45 40843136 Website: http://dr.jones.dk/ [x] quote me freely [ ] ask before reusing [ ] keep private -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (GNU/Linux) iEYEAREDAAYFAkoqnS8ACgkQn7DbMsAkQLhwdgCfRFKNECICN9OPO9pFWNhKAJBp OaMAn2q984i8ehkPD4Tr5SWQA0wf0Ymg =H2Vm -END PGP SIGNATURE- ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Can't compile mozilla on sugar-jhbuild
On Sat, Jun 06, 2009 at 05:55:22PM +0200, Jonas Smedegaard wrote: You wrote in ticket 137 that *hulahop* is broken in *Debian*: Ah, sorry if that wasn't clear enough. I was talking about hulahop in sugar-jhbuild, which doesn't build against the Debian-shipped xulrunner. (hint: maybe _I_ can help _you_ here - if what breaks is jhbuild and I might know about some build options required for proper linking against Debian xulrunner) What I need is some recipe that works on Debian _and_ other distros, because sugar-jhbuild doesn't support setting different options and/or patches depending on the distro (the sysdeps stuff was introduced by us, upstream jhbuild doesn't detect the distribution). CU Sascha -- http://sascha.silbe.org/ http://www.infra-silbe.de/ signature.asc Description: Digital signature ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Can't compile mozilla on sugar-jhbuild
-BEGIN PGP SIGNED MESSAGE- Hash: RIPEMD160 On Sat, Jun 06, 2009 at 04:59:26PM +0200, Sascha Silbe wrote: > On Sat, Jun 06, 2009 at 03:37:34PM +0200, Jonas Smedegaard wrote: > >> Do you mean that this is the case only for sugar-jhbuild or also for >> packaged sugar-xulrunner? > I don't know anything about the sugar-xulrunner package, you have to > ask the Ubuntu team about that. I only maintain sugar-jhbuild, no > distro packages. Ahem, no, I use Debian not Ubuntu, and I meant sugar-hulahop. You wrote in ticket 137 that *hulahop* is broken in *Debian*: >hulahop in git seems to work the other way around now: using the full >version for the path instead of major.minor only. This works on Fedora >and Ubuntu, but breaks on Debian. Any chance we could support both >schemes? Could you please clarify what breaks on *Debian* ? (hint: maybe _I_ can help _you_ here - if what breaks is jhbuild and I might know about some build options required for proper linking against Debian xulrunner) Kind regards, - Jonas - -- * Jonas Smedegaard - idealist og Internet-arkitekt * Tlf.: +45 40843136 Website: http://dr.jones.dk/ [x] quote me freely [ ] ask before reusing [ ] keep private -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (GNU/Linux) iEYEAREDAAYFAkoqkWkACgkQn7DbMsAkQLhaCgCgjalt59I+YXdpmifkVBuzms3p ++cAnR/8zC9WmBk6NHJBxsHeBZtOhIe9 =7uya -END PGP SIGNATURE- ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Journal activity release 101 not tagged in Git
-BEGIN PGP SIGNED MESSAGE- Hash: RIPEMD160 On Sat, Jun 06, 2009 at 04:52:08PM +0200, David Van Assche wrote: >And though packaging debian is quite different from doing rpms, I just >finished getting all activities working for openSUSE. There were lots >of little glitches that prevented activities from working,so if u run >into any issues with particular activities, ping me and I might be able >to tell u how to get stuff working. Activities like infoslicer, colors, >browse, record, and Maze come to mind... Generally I've been working >directly off the latest source in gitorious unless there was a required >downgrade to another source (storybuilder)... Thanks. Not sure if you know, but I have been packaging for some time (since Sugar 0.81) and Sugar 0.82 worked correctly. Also, the current trouble is with core Sugar environment, not activities. That said, I certainly appreciate the offer, and generally I like the growth in teams attacking the Sugar code from different angles. That really helps improve code quality for all of us. :-D - Jonas - -- * Jonas Smedegaard - idealist og Internet-arkitekt * Tlf.: +45 40843136 Website: http://dr.jones.dk/ [x] quote me freely [ ] ask before reusing [ ] keep private -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (GNU/Linux) iEYEAREDAAYFAkoqjf4ACgkQn7DbMsAkQLg41ACeLapB6MhAX4GK2TuJpS7rz5Sb NkAAn2PmymU7T3ZiOe8IV9TmEjTm1fAy =OSXG -END PGP SIGNATURE- ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Can't compile mozilla on sugar-jhbuild
On Sat, Jun 06, 2009 at 03:37:34PM +0200, Jonas Smedegaard wrote: Do you mean that this is the case only for sugar-jhbuild or also for packaged sugar-xulrunner? I don't know anything about the sugar-xulrunner package, you have to ask the Ubuntu team about that. I only maintain sugar-jhbuild, no distro packages. CU Sascha -- http://sascha.silbe.org/ http://www.infra-silbe.de/ signature.asc Description: Digital signature ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Journal activity release 101 not tagged in Git
And though packaging debian is quite different from doing rpms, I just finished getting all activities working for openSUSE. There were lots of little glitches that prevented activities from working,so if u run into any issues with particular activities, ping me and I might be able to tell u how to get stuff working. Activities like infoslicer, colors, browse, record, and Maze come to mind... Generally I've been working directly off the latest source in gitorious unless there was a required downgrade to another source (storybuilder)... kind Regards, David (nubae) On Sat, Jun 6, 2009 at 10:40 AM, Jonas Smedegaard wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: RIPEMD160 > > On Sat, Jun 06, 2009 at 10:16:29AM +0200, Tomeu Vizoso wrote: > >On Sat, Jun 6, 2009 at 00:19, Jonas Smedegaard wrote: > >> On Fri, Jun 05, 2009 at 04:37:32PM -0400, Frederick Grose wrote: > >>>You are already on it by virtue of your work and contributions. > >>> Please edit http://wiki.sugarlabs.org/go/Packaging_Team and its > >>>pending links/subpages as appropriate. > >> > >> Ok. I'll look into that. Later - I am on my way to bed now (spent > >> all of last night finalizing a major ghostscript packaging update). > >> > >> > >>>Thank you for contributing so much already! --Fred > >> > >> Ah, well - my work is just silently absorbed by Ubuntu, and the core > >> 0.84 packages that I finally got packaged last week does not work at > >> all. Segmentation fault somewhere initially if using Xephyr, and if > >> cheating and first running sugar non-emulated (just to create the > >> initial account), activities won't start - no debug log messages, no > >> nothing. > > > >I'm going to be offline the weekend, but we can look at these issues > >together next week. > > That'd be awesome. Ping me when you are ready. > > > - Jonas > > - -- > * Jonas Smedegaard - idealist og Internet-arkitekt > * Tlf.: +45 40843136 Website: http://dr.jones.dk/ > > [x] quote me freely [ ] ask before reusing [ ] keep private > -BEGIN PGP SIGNATURE- > Version: GnuPG v1.4.9 (GNU/Linux) > > iEYEAREDAAYFAkoqK2sACgkQn7DbMsAkQLgXHwCbB3HYWoh8f0Z/fMITxEkHkwCk > g7kAn10W91Ha+sNZB1Cqd+KCGkGFS/Ew > =n2Qj > -END PGP SIGNATURE- > ___ > Sugar-devel mailing list > Sugar-devel@lists.sugarlabs.org > http://lists.sugarlabs.org/listinfo/sugar-devel > ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Can't compile mozilla on sugar-jhbuild
-BEGIN PGP SIGNED MESSAGE- Hash: RIPEMD160 On Sat, Jun 06, 2009 at 11:50:54AM +0200, Sascha Silbe wrote: > On Sat, Jun 06, 2009 at 10:04:37AM +0200, Jonas Smedegaard wrote: > >> As for your problem, I would not recommend compiling xulrunner >> yourself (if jhbuild even supports that) - you should probably >> instead install xulrunner-dev. > Unfortunately we have to do that (compiling xulrunner ourselves in > sugar-jhbuild) because Debian changed the xulrunner location [1]. > While the new location (...-1.9) is way better (and should be changed > upstream as well), hulahop currently only supports the upstream way > (...-1.9.0.x). > > > [1] http://dev.sugarlabs.org/ticket/137 Do you mean that this is the case only for sugar-jhbuild or also for packaged sugar-xulrunner? That bugreport does not clarify what changed in sugar-xulrunner, and no recent changes seem obious to me. Could you please elaborate a bit? (Browse fails for me currently with recently built sugar-xulrunner, but for different reasons than "cannot open shared object file" as described in above bugreport, so I cannot verify at the moment) - Jonas - -- * Jonas Smedegaard - idealist og Internet-arkitekt * Tlf.: +45 40843136 Website: http://dr.jones.dk/ [x] quote me freely [ ] ask before reusing [ ] keep private -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (GNU/Linux) iEYEAREDAAYFAkoqcR4ACgkQn7DbMsAkQLh02wCgjPHOhbYK7g2SGd2veXnK1kd8 ir0An3o7SvE3RBNc2WMAgGnNcv2BpeL9 =An03 -END PGP SIGNATURE- ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Can't compile mozilla on sugar-jhbuild
On Sat, Jun 06, 2009 at 03:46:38PM +0200, Bastien wrote: So why did I need to "rm -rf source/mozilla source/hulahop" before rebuilding? Politically correct answer: because xulrunner often breaks after updates and rebuilding it from scratch fixes it. Maybe I miss something here, but the hard part is not to "rm -rf " the hard part is to know what dir to delete... For most modules the directory name is the same as the module name (with version appended in case of tarballs). For xulrunner it's different for historical reasons. I had the hope we could get rid of xulrunner completely (as a module of sugar-jhbuild), but as this has proven false I'm going to fix the name soon. CU Sascha -- http://sascha.silbe.org/ http://www.infra-silbe.de/ signature.asc Description: Digital signature ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Can't compile mozilla on sugar-jhbuild
Sascha Silbe writes: > On Sat, Jun 06, 2009 at 03:19:04PM +0200, Bastien wrote: > >> Maybe ./sugar-jhbuild could make sure that directories are clean? > It's by design. Only changed portions of the code should get > rebuilt. Doing a full rebuild each time would be a huge waste of > resources. So why did I need to "rm -rf source/mozilla source/hulahop" before rebuilding? >> Or maybe we could have ./sugar-jhbuild clean (similar to make clean)? > Actually there is, but it just invokes "make clean" which usually isn't > enough (we especially want autoconf / autogen.sh to be run again). What > would be more useful is a "--force-checkout" to updateone (like > sugar-jhbuild offers interactively upon encountering an error), but as > "rm -rf " is easy enough... Maybe I miss something here, but the hard part is not to "rm -rf " the hard part is to know what dir to delete... -- Bastien ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Can't compile mozilla on sugar-jhbuild
On Sat, Jun 06, 2009 at 03:19:04PM +0200, Bastien wrote: Maybe ./sugar-jhbuild could make sure that directories are clean? It's by design. Only changed portions of the code should get rebuilt. Doing a full rebuild each time would be a huge waste of resources. Or maybe we could have ./sugar-jhbuild clean (similar to make clean)? Actually there is, but it just invokes "make clean" which usually isn't enough (we especially want autoconf / autogen.sh to be run again). What would be more useful is a "--force-checkout" to updateone (like sugar-jhbuild offers interactively upon encountering an error), but as "rm -rf " is easy enough... CU Sascha -- http://sascha.silbe.org/ http://www.infra-silbe.de/ signature.asc Description: Digital signature ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
[Sugar-devel] sugar-jhbuild broken on Debian squeeze (was: Re: Can't compile mozilla on sugar-jhbuild)
On Sat, Jun 06, 2009 at 11:55:49AM +0200, Sascha Silbe wrote: Please check you've installed all dependencies (i.e. ./sugar-jhbuild depscheck returns nothing), remove old build results (rm -rf source/mozilla source/hulahop) and try again (./sugar-jhbuild build). Just noticed sugar-jhbuild recently got broken on Debian squeeze [1] because lsb-release now outputs "unstable/testing" instead of either "testing" or "unstable". I'll commit a change to sugar-jhbuild I've worked on last week that will fix this, but I'd like to get a second pair of eyes on it first (it's a partial rewrite of the sysdeps config system). [1] http://dev.sugarlabs.org/ticket/939 CU Sascha -- http://sascha.silbe.org/ http://www.infra-silbe.de/ signature.asc Description: Digital signature ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Can't compile mozilla on sugar-jhbuild
Sascha Silbe writes: > Please check you've installed all dependencies (i.e. ./sugar-jhbuild > depscheck returns nothing), remove old build results (rm -rf > source/mozilla source/hulahop) and try again (./sugar-jhbuild build). Maybe ./sugar-jhbuild could make sure that directories are clean? Or maybe we could have ./sugar-jhbuild clean (similar to make clean)? -- Bastien ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Can't compile mozilla on sugar-jhbuild
Sascha Silbe writes: > Unfortunately we have to do that (compiling xulrunner ourselves in > sugar-jhbuild) because Debian changed the xulrunner location [1]. While > the new location (...-1.9) is way better (and should be changed upstream > as well), hulahop currently only supports the upstream way > (...-1.9.0.x). xulrunner-1.9 was already installed but I installed xulrunner and xulrunner-dev and was able to build sugar-jhbuild. Thanks! 1. ./sugar-jhbuild depscheck did not advertize it 2. I added this information here: http://wiki.sugarlabs.org/go/Development_Team/Jhbuild/Debian One remaining problem with Etoys: , | + cp po/zh_TW/etoys.po Content/po/etoys/zh_TW.po | cp: ne peut créer le fichier régulier `Content/po/etoys/zh_TW.po': Aucun fichier ou dossier de ce type | make: *** [Content/po] Erreur 1 ` I remember it's been suggested not to include Etoys in sugar-jhbuild, I second that suggestion! -- Bastien ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Can't compile mozilla on sugar-jhbuild
On Sat, Jun 06, 2009 at 06:03:03AM +0200, Bastien wrote: | ../../staticlib/libxpcom_core.a: member ../../staticlib/libxpcom_core.a(nsDeque.o) in archive is not an object | collect2: ld returned 1 exit status | make[3]: *** [libxul.so] Erreur 1 Please check you've installed all dependencies (i.e. ./sugar-jhbuild depscheck returns nothing), remove old build results (rm -rf source/mozilla source/hulahop) and try again (./sugar-jhbuild build). CU Sascha -- http://sascha.silbe.org/ http://www.infra-silbe.de/ signature.asc Description: Digital signature ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Can't compile mozilla on sugar-jhbuild
On Sat, Jun 06, 2009 at 10:04:37AM +0200, Jonas Smedegaard wrote: As for your problem, I would not recommend compiling xulrunner yourself (if jhbuild even supports that) - you should probably instead install xulrunner-dev. Unfortunately we have to do that (compiling xulrunner ourselves in sugar-jhbuild) because Debian changed the xulrunner location [1]. While the new location (...-1.9) is way better (and should be changed upstream as well), hulahop currently only supports the upstream way (...-1.9.0.x). [1] http://dev.sugarlabs.org/ticket/137 CU Sascha -- http://sascha.silbe.org/ http://www.infra-silbe.de/ signature.asc Description: Digital signature ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [RELEASE] Library-1
On Mon, Jun 1, 2009 at 11:39, Aleksey Lim wrote: > Hi all, > > This is in initial release of Library activity. I decided to release it > before implementing collab features to get some feedback about UI and basic > ideas of activity. So, it can browse only local objects. Some related work in the GNOME 3.0 side of things: http://seilo.geekyogre.com/2009/06/some-zeitgeist-visuals/ http://seilo.geekyogre.com/2009/06/the-db-candidate-for-zeitgeist/ Would be awesome if someone could spend some time talking with them and seeing how much work we could share. Regards, Tomeu > == Bundle == > > http://activities.sugarlabs.org/en-US/sugar/addon/4089 > > == NEWS == > > * Add filter to ObjectChooser invoking > * Browse local objects > > == TODO fot v2 == > > * Run in 0.82 environment > * Implement collab features > > -- > Aleksey > ___ > Sugar-devel mailing list > Sugar-devel@lists.sugarlabs.org > http://lists.sugarlabs.org/listinfo/sugar-devel > ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] [Marketing] [IAEP] ASLO - rebranding
On Fri, Jun 05, 2009 at 12:29:45PM -0700, Josh Williams wrote: > Thanks everyone, > > Gary mentioned > > "Hey, fab, looking much cleaner/simpler/Sugar-like :-) One quick > thought (will try to take a closer look later), would double > thickness borders work better, make the colour theme pop a little > more, feel more robust? > > Regards, > --Gary " > > > I agree that adding two or three pixels to the borders would help the > design. However, if there aren't any major objections to the design, I > think we should push it to the live ASLO, and make minor changes later. > Any thoughts on this? I guess we can push it to activities.sugarlabs.org and continue to improve it Josh, after commiting your pathces to master branch, let me know and I'll rebase production branch -- Aleksey ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] More information on font scaling problem
On Fri, Jun 05, 2009 at 05:30:13PM +0200, Tomeu Vizoso wrote: > On Fri, Jun 5, 2009 at 17:11, James Simmons wrote: > > Tomeu, > > > > The thing is, in the code below I *am* using style.zoom. It seems to have > > no effect at all. > > What I would have supposed that would work is using in 0.84 a fixed > font size of 10 and on the XO style.zoom(X). I didn't wanted to imply > that X could be 10, I think we were using a smaller font size in the > Sugar UI. > > Maybe Aleksey could add a zoom() function to sugar.port that on 0.82 > would do some scaling based on the screen resolution and on 0.84 would > return the value unchanged? done, http://people.sugarlabs.org/~alsroot/sugar-port/port.style-module.html -- Aleksey ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Journal activity release 101 not tagged in Git
-BEGIN PGP SIGNED MESSAGE- Hash: RIPEMD160 On Sat, Jun 06, 2009 at 10:16:29AM +0200, Tomeu Vizoso wrote: >On Sat, Jun 6, 2009 at 00:19, Jonas Smedegaard wrote: >> On Fri, Jun 05, 2009 at 04:37:32PM -0400, Frederick Grose wrote: >>>You are already on it by virtue of your work and contributions. >>> Please edit http://wiki.sugarlabs.org/go/Packaging_Team and its >>>pending links/subpages as appropriate. >> >> Ok. I'll look into that. Later - I am on my way to bed now (spent >> all of last night finalizing a major ghostscript packaging update). >> >> >>>Thank you for contributing so much already! --Fred >> >> Ah, well - my work is just silently absorbed by Ubuntu, and the core >> 0.84 packages that I finally got packaged last week does not work at >> all. Segmentation fault somewhere initially if using Xephyr, and if >> cheating and first running sugar non-emulated (just to create the >> initial account), activities won't start - no debug log messages, no >> nothing. > >I'm going to be offline the weekend, but we can look at these issues >together next week. That'd be awesome. Ping me when you are ready. - Jonas - -- * Jonas Smedegaard - idealist og Internet-arkitekt * Tlf.: +45 40843136 Website: http://dr.jones.dk/ [x] quote me freely [ ] ask before reusing [ ] keep private -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (GNU/Linux) iEYEAREDAAYFAkoqK2sACgkQn7DbMsAkQLgXHwCbB3HYWoh8f0Z/fMITxEkHkwCk g7kAn10W91Ha+sNZB1Cqd+KCGkGFS/Ew =n2Qj -END PGP SIGNATURE- ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Journal activity release 101 not tagged in Git
On Sat, Jun 6, 2009 at 00:19, Jonas Smedegaard wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: RIPEMD160 > > On Fri, Jun 05, 2009 at 04:37:32PM -0400, Frederick Grose wrote: >>You are already on it by virtue of your work and contributions. Please >>edit http://wiki.sugarlabs.org/go/Packaging_Team and its pending >>links/subpages as appropriate. > > Ok. I'll look into that. Later - I am on my way to bed now (spent all > of last night finalizing a major ghostscript packaging update). > > >>Thank you for contributing so much already! --Fred > > Ah, well - my work is just silently absorbed by Ubuntu, and the core > 0.84 packages that I finally got packaged last week does not work at > all. Segmentation fault somewhere initially if using Xephyr, and if > cheating and first running sugar non-emulated (just to create the > initial account), activities won't start - no debug log messages, no > nothing. I'm going to be offline the weekend, but we can look at these issues together next week. Cheers, Tomeu > Oh well. I should sleep now :-/ > > > > - Jonas > > - -- > * Jonas Smedegaard - idealist og Internet-arkitekt > * Tlf.: +45 40843136 Website: http://dr.jones.dk/ > > [x] quote me freely [ ] ask before reusing [ ] keep private > -BEGIN PGP SIGNATURE- > Version: GnuPG v1.4.9 (GNU/Linux) > > iEYEAREDAAYFAkopmf0ACgkQn7DbMsAkQLjW7wCeIwpGdpEagK0q7Z/9TJxYLFUd > Xr0AoIG/iRZ565cKBPCW7MEzJl+IBMOT > =r9Fi > -END PGP SIGNATURE- > ___ > Sugar-devel mailing list > Sugar-devel@lists.sugarlabs.org > http://lists.sugarlabs.org/listinfo/sugar-devel > ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] Can't compile mozilla on sugar-jhbuild
-BEGIN PGP SIGNED MESSAGE- Hash: RIPEMD160 On Sat, Jun 06, 2009 at 06:03:03AM +0200, Bastien wrote: >Still failing to compile sugar-jhbuild: > >, >| ../../staticlib/libxpcom_core.a: member >../../staticlib/libxpcom_core.a(nsDeque.o) in archive is not an object >| collect2: ld returned 1 exit status >| make[3]: *** [libxul.so] Erreur 1 >` > >I'm running GNU/Linux version 2.6.26-2-686 (Debian 2.6.26-15). Above is only your _Linux_ version. For info on your userspace, please use this command: lsb_release -a As for your problem, I would not recommend compiling xulrunner yourself (if jhbuild even supports that) - you should probably instead install xulrunner-dev. - Jonas Not a jhbuild user - -- * Jonas Smedegaard - idealist og Internet-arkitekt * Tlf.: +45 40843136 Website: http://dr.jones.dk/ [x] quote me freely [ ] ask before reusing [ ] keep private -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (GNU/Linux) iEYEAREDAAYFAkoqIxUACgkQn7DbMsAkQLjhPACeOSdGrC6iCgS1jcIqjk3LnWKe uUoAn3aUSQtGi8EPVbYi1XKnW5cHrfMC =CCcK -END PGP SIGNATURE- ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] ASLO - rebranding
On Fri, Jun 05, 2009 at 11:42:36AM -0500, James Simmons wrote: > Josh, > > I attempted to upload Read Etexts to your site. The upload seemed to go OK > but there is no Activity found yet. Maybe it needs to be promoted from the > sandbox? thats because you didn't even complete it so, you need to complete it, since that time activity will appear on the list as a experimental(but only for logined users) and nominate it -- Aleksey ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel
Re: [Sugar-devel] More information on font scaling problem
On Fri, Jun 05, 2009 at 11:16:44AM +0200, Tomeu Vizoso wrote: > On Wed, Jun 3, 2009 at 16:40, James Simmons wrote: > > I mentioned before that I'm having a font sizing problem on the XO > > running .82. When I set a font on a textviewer it looks OK running > > under sugar-devel or SoaS by is *much* larger on the XO. > > Sorry, in an earlier email i replied to you what would work in 0.84, > didn't realized you were talking about 0.82. > > In 0.84 Pango knows the resolution of the screen and scales all fonts > accordingly. This means that by using, say 10, the font will always > look good in all hw. > > In 0.82 didn't knew about the screen, so we were scaling all fonts by > using style.zoom(). > > Aleksey, do you have an idea about what can do activity authors that > wish to provide a single bundle that looks good both in 0.82 and 0.84? Well, situation with font sizes in 0.82 was for me a bit messy so, I just didn't use any custom font sizes(I didn't need any special ones) But I see that James uses custom font sizes by design. James, could you use sugar.graphics.style.FONT_SIZE for initial font size instead of hardcoded value. On 0.84 it will be 10 all time, on 0.82 it should be scaled. -- Aleksey ___ Sugar-devel mailing list Sugar-devel@lists.sugarlabs.org http://lists.sugarlabs.org/listinfo/sugar-devel