[Sugar-devel] XO Infinity?

[Tony Anderson]

Hi, Sebastian

Thanks for your input.

My perspective is somewhat limited. I am looking at a deployment where 
(a) internet access is unavailable or limited in capacity so that online 
browsing by the laptop is not feasible, (b) a school server is available 
to provide access to educational resources beyond what can be stored on 
1-4GB of local storage, (c) institutional staff control access to the 
laptops (users taking them home, a rare exception), and (d) all laptops 
on the local school server network should, in so far as possible, behave 
in the same way. This last is increasingly difficult: the XO-1 does not 
have enough storage for Libre Office, the ARM based XOs do not execute 
many of the Sugar activities with binary components (e.g. GCompris), the 
move to support touch-screen has apparently put the XO-1 at software 
end-of-life (while the XO-1 probably represents 80+% of the XO laptops 
in current use).

Someone who comes to us with hopes to deploy the XO to a community 
school, library, or orphanage has limited opportunities to acquire XO 
laptops.

There is surprisingly limited useful advice on things to consider in 
planning such a deployment: solar or grid power, battery charging, 
capacity of routers to support concurrent access to the school server, 
good performance studies on the load placed on the school server by 
40-100 XOs connected concurrently, how well sharing works with (or 
without) ejabberd in an environment of 40+ laptops or what 
teachers/staff need to know to support such a deployment. Strangely, a 
much of this has been addressed in scattered emails and web pages (e.g 
Richard Smith's contributions to understanding the electrical power 
requirements).

There is no clear consensus within the community on what the laptop 
brings to the educational experience: collaborative learning using Sugar 
activities, access to information (Wikipedia, Gutenberg books, Open 
Street Maps), opportunity to learn to use a computer effectively, 
opportunity to learn programming through Turtle Blocks, Scratch, Etoys, 
Pippy, or View Source,  and/or interactive support for meeting the 
learning objectives of the school's curriculum. The 'commissioned' 
studies on the impact of OLPC deployments invariably are focused on 
observed improvements in student performance on standard tests - that, 
at least, is the official consensus on what we are about.

Tony

On 02/19/2015 10:32 AM, Sebastian Silva wrote:
> Hi Tony
>
> I applaud you for asking some important questions going forward in 
> such a concise way.
> Even though we're officially, as a software community, 
> hardware-agnostic, it is very good to discuss and find solutions to 
> common problems (such as availability of appropriate hardware).
>
> It seems from here as if consumer product manufacturers, telecoms, and 
> software vendors were coalesced, and wished communities such as 
> ourselves did not exist.
>
> I'll try to answer bellow, but I am just a volunteer such as yourself 
> and don't represent Sugar Labs.
>
> El 17/02/15 a las 21:05, Tony Anderson escibió:
>> Hi,
>>
>> I am sceptical that the XO market will ever be able to sustain 
>> manufacture of an XO-specific product. I hope and wish I am wrong. 
>> However, I think we need to look for alternatives. Possibly the most 
>> serious impediment to success of the OLPC initiative has been the 
>> lack of laptop available for purchase by a deployment in small 
>> quantities.
> +1
> I like initiatives like the Kit Kano, for instance. I've grown to 
> expect revolutions to happen slowly. I know eventually the killer 
> durable, attractive, affordable, usable and truly libre solution will 
> arrive. Part of that task is in our hands (the software, this /is/ the 
> Sugar community after all). Attractive, usable, truly libre. Simple, 
> collaborative, reflexive.
>>
>> As Wayan Vota said, 'Would you recommend a new deployment with the 
>> XO?'. My answer would be yes, provided the deployment had a reliable 
>> source of XO laptops for under $200 (and spare parts).
>> The XO-1 is still viable provided that there is a source of ongoing 
>> support.
> I do all my work in a year old $199 Chromebook with Parabola 
> GNU/Linux. It is about as libre as it gets, and whooping fast with a 
> Haswell processor.
> However Google seems intent to only allow such a setup for "developer 
> mode" (which can wipe your drive at a wrong keystroke on boot).
>>
>> I must applaud Samuel Greenfeld's initiative to create community 
>> builds of the XO software. As Bernie Innocenti pointed out at the 
>> Malaysia summit, finding a community that can sustain support for the 
>> XO builds going forward is probably the number one problem facing the 
>> community.
> +1 I've failed to respond but do plan to resume work on XO (and 
> regular PC) builds soon. I count on picking up from Samuel's work and 
> as usual, have our develop, build and publish process be 
> community-oriented.
>>
>> There are Android tablets on the market which can be purchased with a 
>> case and keyboard:
>>
>> http://www.sears.com/proscan-7inch-internet-tablet-with-8-gb-and/p-020W006276292001P
>>
>>
>> Rabi Karmacharya believes we need a minimum 10" screen, the above has 
>> 7" and is quoted at $69.
>> Is there a comparable device with a 10" screen at 1200x900 or better 
>> for under $100, under 200$?
>>
>> The CTL Education Chromebook is available for $279 
>> (http://ctl.net/ctl-education-chromebook). Does it offer the 
>> capabilities we need? Can alternate software (such as Fedora/Sugar) 
>> be installed?
>
> We must raise awareness at the level of the people who consult with 
> us, with regard to the importance of devices respecting users (and 
> deployments). It is not acceptable to be tied to one "software store". 
> It is not acceptable to have our governments procure machines that 
> require non-free software to work, or worse, have no support at all.
>>
>> Is our future to go away from Fedora/Sugar (Linux/Gnu/Sugar) and to 
>> base our deployments on Android?
> Fedora/Sugar need not be the only option. I'f like to see more 
> Sugar-like initiatives, i.e. tailored desktops.
>> Does this mean we must abandon our insistence on open source and open 
>> educational resources? 
> Never!
>> Is it possible to deploy an Android system without access to the 
>> internet?
> Probably. Not too interesting.
>>
>> There are currently 200+ educational activities available for Sugar. 
>> Must we give these up?
> No! We should strive to mantain some compatibility for as long as 
> possible.
>> Do we need to reprogram them in javascript?
> Rather create new ones, support and fix old ones! It's not easy as 
> there are few hands.
>>
>> If we are to continue with Fedora/Sugar, can this software be 
>> installed on Android tablets?
> Most likely nope.
>>
>> The world's professional programmers are now either (or both) 
>> programming for the javascript/html market or the Java Android 
>> market. Does this mean we need to jump on those bandwagons? Should we 
>> shift our 'view source' initiative to Java or to javascript/html?
>>
> Nope.
>> In summary, I believe that the future of the olpc initiative depends 
>> on the skills, commitment, and hard work of the community. I think it 
>> very unlikely that our challenges will be met by a 'magic wand' waved 
>> by OLPC reborn, OLPC Australia, or other agency. It's up to us.
>>
> Much as yourself, I feel the loss of the momentum the OLPC community 
> had. But I am profoundly aware that, in all, this story is still just 
> beginning, and there are plenty of worthy initiatives around us. And 
> we ourselves keep pushing forward for Sugar, the vision, not 
> necessarily a specific implementation.
>
> I have a short saying about this last bit. I offer it to you. Call it 
> Sebastian's Razor. "Si no es libre, no existe".
>
> :-)
> Regards,
> Sebastian

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sugarlabs.org/archive/sugar-devel/attachments/20150219/b8454be3/attachment.html>