[IAEP] Sugarized binaries? was Re: users doing python in XOs

[Yama Ploskonka]

Maybe there was a misunderstanding. I have no intention to rewrite Sugar 
activities.

Thank you for your concern.

No worry no more. I learned that my relevant target audience for 
Sugarizing RPMs, kids potential microcontroller developers in Uruguay, 
no longer are limited by the block Ceibal has put on sudo in XOs.

Because they don't use XOs anymore.
They use Ubuntu, on Classmates.


As to the background, my personal experience with compiled vs. 
interpreted is that compiled runs faster.

So, if you compile your code (I remember how excited we were when one of 
our pals discovered this compiler for Basic, back when) it would 
"improve" things.
No idea if there is such a thing as a compiler that will take Python as 
source, but nowadays it becomes a moot point, for me, to Sugarize things.

People free to use sudo, will use sudo.
The others use Ubuntu. They both use Terminal and don't even notice the 
silly overhead the GUI causes them, but since the binaries are so fast 
in compiling, it doesn't really matter.
A few XO users are lost - I don't have the skills to help them, sorry.

as to the specialized use, well, I really don't know, not being of that 
persuasion, if it is or is not.

As to real world use of accessing the code by kids to make modifications 
directly on their own XOs, I agree it was a *nice* idea (I loved it when 
I heard about it)

thank you again!


On 12/02/2012 02:33 PM, Martin Dengler wrote:
> On Sun, Dec 02, 2012 at 10:57:39AM -0600, Yama Ploskonka wrote:
>> On 12/02/2012 08:13 AM, Martin Dengler wrote:
>>> On Sat, Dec 01, 2012 at 10:08:07PM -0600, Yama Ploskonka wrote:
>>>> If it is defunct, can we use binaries finally to optimize and speed
>>>> up operation
>>> "Using binaries" is not what needs to be done.  "Rewriting each
>>> activity" (that you want to speed up) is what you're saying needs to
>>> be done.  That's a lot more work than just finding the bottlenecks in
>>> existing python applications and reducing them.
>> [no response]
> <shrug>
>
>> with lesser importance, my authorities were that C and variants are
>> faaaaar ahead [...] of anything else [...], as "the language for the
>> real world" right now [...] I'm surprised to hear otherwise
> Of course C is the most popular (by lines written, and usefulness as a
> "desert island language"), and nobody say anything but.
>
>> [...] but then there are specialized applications, I agree, and
>> opinions :-)
> Specialized applications like a constuctivist learning platform,
> perhaps?
>
>> I quote the Python link: “It’s still a relatively niche skill-set
>> and demand isn’t astronomical [...]"
> Your tiobe.com reference categorizes python as a "mainstream
> language".  If you're asking about re-writing major/all Sugar
> activities, we'd better have more to argue about than that.
>
> Martin