[Sugar-devel] [IAEP] Sugar oversight board meeting

[Walter Bender]

On Tue, Nov 5, 2013 at 4:44 PM, Flavio Danesse <fdanesse at gmail.com> wrote:
> My humble opinion (please stick to one):
>
> To put into perspective the opinion, I should remember that besides
> developing for sugar since 2009, I am also a teacher in high school, so I've
> been inside ceibal classrooms during this time.
>
> From the beginning, I said I saw the fate of sugar linked to the xo, the one
> without the other does not seem to make sense. Now, OLPC xo 4 and
> manufactures their away strip.
>
> For those who did the port to gtk3 last year, and we have also had to deal
> with the problems of arm processors, etc.. . ., We do not easily see how
> much time is lost in these "strategic decisions" while it ignores the
> feedback from deployments.

Just to set the record straight, Ceibal was very supportive of the
push to ARM. That doesn't mean it was the correct decision. But it
suggests that OLPC was not ignoring feedback from deployments
regarding that issue. As far as GTK3, the only feedback I heard was
after the fact from Peru that it had adverse performance impact on XO
1. This decision was made publicly by Sugar Labs for at least two
reasons: touch support and future-proofing. GTK2 is no longer
supported.

>
> I think this whole issue of android and html5, is a very grave mistake,
> probably the last.

I would be interested in hearing more on this topic.

>
> But hey, I'm just a teacher, probably the only one in this list.
>
>
> 2013/11/5 Daniel Narvaez <dwnarvaez at gmail.com>
>>
>> Oh, awesome, COPR seems to be exactly what we need.
>>
>>
>> On Tuesday, 5 November 2013, Peter Robinson wrote:
>>>
>>> On Tue, Nov 5, 2013 at 1:40 PM, Daniel Narvaez <dwnarvaez at gmail.com>
>>> wrote:
>>> > Going a bit off topic, but a pretty major issue I see in our workflow
>>> > with
>>> > Fedora is that we don't have a good way to develop unstable Sugar on a
>>> > stable Fedora. Rawhide is, or at least is perceived as, unstable. And
>>> > I'm
>>> > not sure what would be a good way to, for example, produce and
>>> > distribute
>>> > 0.100 rpms for Fedora 19. We can setup our custom automated build
>>> > system and
>>> > repository of course, but I'm not sure that's a good approach? Part of
>>> > the
>>> > problem here is that upstream tends to depend strongly on very recent
>>> > libraries which are not yet available in the stable fedora, though
>>> > maybe now
>>> > that the gi conversion is over we can avoid that.
>>>
>>> Actually a lot of that will be solved perfectly with COPR (similar in
>>> style to Ubuntu PPA) which is being worked upon at the moment and it
>>> should solve all the problems you see by enabling newer versions to be
>>> built for older releases while maintaining the stable shipped release
>>> in mainline.
>>>
>>> Peter
>>>
>>> > On Tuesday, 5 November 2013, Peter Robinson wrote:
>>> >>
>>> >> On Tue, Nov 5, 2013 at 2:14 AM, Walter Bender
>>> >> <walter.bender at gmail.com>
>>> >> wrote:
>>> >> > On Mon, Nov 4, 2013 at 7:42 PM, Daniel Narvaez <dwnarvaez at gmail.com>
>>> >> > wrote:
>>> >> >> On 4 November 2013 22:53, Sean DALY <sdaly.be at gmail.com> wrote:
>>> >> >>>
>>> >> >>> * It's not clear to me where we are going. The OLPC/Sugar
>>> >> >>> development
>>> >> >>> ecosystem seems to be at a crossroads. I am encouraged by the web
>>> >> >>> activity
>>> >> >>> work, but don't understand the path of transposing the value
>>> >> >>> proposition of
>>> >> >>> Sugar (interface, Journal, collaboration, Activities) to handheld
>>> >> >>> tactile
>>> >> >>> devices (tablets to smartphones). PCs (of any size) with keyboards
>>> >> >>> are
>>> >> >>> no
>>> >> >>> longer competitive with tablets for grade-school classroom use.
>>> >> >>> Perhaps the
>>> >> >>> XO-4 could still be in the running; there is no clear message from
>>> >> >>> OLPC.
>>> >> >>
>>> >> >>
>>> >> >>
>>> >> >> I'll try to express briefly my feelings about the directions the
>>> >> >> project
>>> >> >> could take. Note that I might be missing a lot of what is going on
>>> >> >> above the
>>> >> >> technical level.
>>> >> >>
>>> >> >> * The XO is not a viable hardware platform other than for existing
>>> >> >> deployments. OLPC is pretty clearly going in a different direction.
>>> >> >
>>> >> > I may be alone in thinking that there will be some runway left with
>>> >> > the XO. But deployments need alternatives regardless.
>>> >> >
>>> >> >> * Sugar web activities on the top of a full Android loses too much
>>> >> >> of
>>> >> >> the
>>> >> >> Sugar value proposition. It's great to have it in addition to
>>> >> >> Sugar-the-OS,
>>> >> >> but it's not enough alone.
>>> >> >
>>> >> > I agree.
>>> >> >
>>> >> >> * From the technical point of view there are several ways to get
>>> >> >> Sugar-the-OS running on tactile devices. Unfortunately it's not
>>> >> >> clear
>>> >> >> to me
>>> >> >> that any of these devices is open enough to be viable for
>>> >> >> deployments
>>> >> >> or
>>> >> >> "ordinary" users.
>>> >> >
>>> >> > We looked at ChromeOS a few years back, but at the time it was too
>>> >> > heavy for our hardware. Today, it is a different story. Might be a
>>> >> > viable option. Certainly running GNU/Linux/Sugar on a ChromeBook is
>>> >> > not a bad starting point.
>>> >>
>>> >> Given that ChromeOS is locked down I don't believe it's viable to ask
>>> >> a School to have to break/hack the HW to get it working OOTB.
>>> >>
>>> >> Having been involved in the OLPC OS side of things I believe you would
>>> >> be much better taking the work done by OLPC with things like
>>> >> olpc-os-builder and the work upstream with Fedora to use it to build
>>> >> out OS images that will work in a similar way across both XOs and
>>> >> other HW be it x86 netbook or cheap ARM devices rather than
>>> >> reinventing the wheel!
>>> >>
>>> >> Peter
>>> >
>>> >
>>> >
>>> > --
>>> > Daniel Narvaez
>>> >
>>
>>
>>
>> --
>> Daniel Narvaez
>>
>>
>> _______________________________________________
>> Sugar-devel mailing list
>> Sugar-devel at lists.sugarlabs.org
>> http://lists.sugarlabs.org/listinfo/sugar-devel
>>
>



-- 
Walter Bender
Sugar Labs
http://www.sugarlabs.org