Sunday, July 26, 2009

Writing Sugar activities with PyQt

At David Boddie's suggestion, have dedicated part of this weekend to explore how well PyQt can be used to write Sugar activities. The results have been very satisfying, I would say thanks to python's flexibility and to the Sugar activity architecture, an excellent work of Marco Pesenti Gritti.

So I got something that looks and behaves similarly to activities compliant to the Sugar Human Interface Guidelines, with the following exceptions:

* No combobox for selecting the sharing mode.

* Some theme quirks were fixed with a custom stylesheet, I think these should be treated as bugs in either QGtkStyle or the Sugar Gtk theme. Or if QGtkStyle is not up to this job, we would need to implement a new QStyle. Couldn't workaround all of the issues in the stylesheet.

* No palettes at all. Hope implementing them in Qt will be much less work than it has been in Gtk+.

* The XCursor theme is not picked up, need to investigate further.

* Some Gtk stuff is still used, I think PyQt activities shouldn't use Gtk+ at all for better memory usage and startup performance.

Good news is that I have reimplemented good part of the activity API in PyQt quite quickly, but we may want to design a better API learning from the mistakes of the past.

Now, let's let the Sugar platform get flooded of KDE Edu activities :)




Friday, July 24, 2009

On Negroponte's omelettes

For the people who aren't following closely the Sugar project (specially my GNOME friends), if you read in certain media that Negroponte thinks that Sugar is OLPC's biggest mistake, you may want to know that some confusion has arisen from his badly-chosen words.

Negroponte is referring by the name Sugar to the whole OS image that ships in their machines, something different to what everybody else refers by Sugar, even OLPC's own Support FAQ.

No need to extend on this myself because others have explained better the gory details:

Something I can add to those posts though is that OLPC management tried once before to refer to the whole software they made as "Sugar" at one occasion before. It was when the community was outraged at the company's approximation to MS, they announced that their commitment to Sugar was still strong and proof of that was that they had doubled the resources available to Sugar.

What in fact happened was that they hired several people to work on the base system, but the people working on the real Sugar were still the same, sad, 2 and a half people.

Given that the Sugar community is working in the benefit of OLPC's customers, I wonder why Negroponte cannot think a bit more before making such "courageous" statements.

Telepathy Developer's Manual

I'm skimming through it right now and this is REALLY good stuff:

Ideally, 90% of Sugar activity authors shouldn't need to get to work much with telepathy directly, because they most likely only want to share a data structure and interchange some messages or signals.

But for the people doing activities such as Chat or VideoChat, implementing collaboration frameworks such as Groupthink or messing with the presence service in Sugar, this will be invaluable and will hopefully lower the floor for new contributors.

Thanks, Collabora!

Tuesday, July 21, 2009

Sugar, usando Software libre para aprender

Por casualidad tropece con este articulo excelentemente escrito por Maria del Pilar Saenz Rodriguez: Sugar, usando Software libre para aprender. Explica muy bien la importancia del software libre en educacion y como encaja Sugar ahi.

Desafortunadamente se necesita Flash para acceder al pdf, esta disponible en la caja "My shared files" en la misma pagina. Y claro, seria muy bueno que Pilar alias befana sindicara su blog en

Tambien seria estupendo tener una traduccion al ingles para la gente de Sugar Labs que lo promueve a nivel internacional, voluntarios?