[Sugar-devel] Priorities and Ideas (for GSoC)

[Sayamindu Dasgupta]

On Sat, Mar 14, 2009 at 3:19 AM, Jameson Quinn <jameson.quinn at gmail.com> wrote:
> I will link to this thread (in IAEP) on the GSoC project ideas page. This
> page is the primary location where prospective GSoC students will come to
> learn about out project, and so I want them to get a feel for our community
> discussion of priorities. So please, in this thread, try to be a little bit
> more explicit and foot-noted than you would be otherwise, so they can
> understand what we're talking about.
>
> The primary purpose of GSoC, as others have pointed out, is NOT to do the
> things we're too busy to get around to. It is primarily a community-building
> exercise: to get students engaged in helping Sugar, and get mentors engaged
> in passing on knowledge to new community members. If somebody develops an
> educational game that only blind 3-year-olds use, but FINISHES it, has a
> great time doing it, and becomes a long-term contributing community member,
> then that would be a total GSoC success. However, that being said, we'd
> still prefer projects that help acheive our highest priorities for Sugar.
>
> There is no absolute ordering of Sugarlabs' priorities. Different members
> will not agree perfectly on what steps will do more to help our educational
> mission. So the list below is just my version. Community: Please respond
> with your thoughts. Students: I'll link what I can in the list, but I can't
> find good links, or even any links, for everything. If one of these ideas
> intrigues you, please, come ask in IRC (#sugar on freenode) - we'd love to
> try to point you in the right direction, and help you cut your ideas down to
> a reasonable GSoC size.
>
> My first priority is things that will have a strong effect on the long-term
> rate of development of Sugar. I'd put just 2 things in that category: easier
> sugarizing (primarily from AJAX, Flash, and legacy Linux); and a structure
> for sugar unit tests (IMO we will never get good enough software quality for
> wide adoption, running on multiple distribution without automated testing).
>
> My second priority is things that will improve on sugar's key promises. An
> easier and better way to handle files: versioned datastore, improvements in
> creating and using tags for the journal, file picker dialogs, and home view.
> A simpler and safer security model: getting Rainbow into the Sugar platform
> and improving it's coverage of the Bitfrost ideals. A simple and
> discoverable, yet powerful, UI overall: improved accessibility, discoverable
> keyboard shortcuts. Ubiquitous connectivity and collaboration: multi-pointer
> sharing, auto-collaborating data structures, viral/peer-to-peer activity
> distribution, shared journals. Useful in the classroom: a one-click workflow
> for getting AND turning in homework.
>
> My third priority is activities to better cover the core functions. Reading:
> an improved Read, which handles true ebook formats. (PDF is made for
> printing, and deployments have asked for this.)

Regarding support for more Ebook formats, in case it is relevant, I am
working on a sugarized FBReader[1] activity at the moment. I should be
able to do a preliminary release by tomorrow. (I was planning a
release tonight, but my main workstation seems to be incredibly messed
up, and won't boot, so I need to fix that first)

Screenshot at http://dev.laptop.org/~sayamindu/fbreader_sugar_v2.png
Thanks,
Sayamindu

[1] http://www.fbreader.org/
-- 
Sayamindu Dasgupta
[http://sayamindu.randomink.org/ramblings]