Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Dave Crossland]

On 7 April 2016 at 10:15, Sebastian Silva  wrote:

> The problem with Sugarizer as a shell is that it can only run sugarizer
> Activities (Sugar Web Activities) and not the Android apps. These
> activities are very limited on a first experience.


There was a GSOC proposal to enable it as a shell that can launch other
Android apps
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Dave Crossland]

Hi Tony!

On 7 April 2016 at 19:14, Tony Anderson  wrote:

>
> There are a number of projects to put Linux on Android devices, e.g.
> http://linuxonandroid.org/.
>

Nice! I didn't know it :)


> I could see an openWRT scenario in which there would be an inventory of
> Android products which support a Linux install (possibly with specific
> install instructions and even images).
>

Perhaps could be done within https://h-node.org :)

Cheers
Dave
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Tony Anderson]

Hi, Dave

There are a number of projects to put Linux on Android devices, e.g. 
http://linuxonandroid.org/.


I could see an openWRT scenario in which there would be an inventory of 
Android products which support a Linux install (possibly with specific 
install instructions and even images).


Tony

On 04/07/2016 09:03 PM, Dave Crossland wrote:

Hi Tony

Don't get me wrong, what I am excited about is being able to buy up 
Amazon's subsidised hardware and rip out their Android distro and put 
in a clean one. I'm not sure about putting in a GNU/Linux, a clean 
Android/Linux distro seems more likely to go smoothly.


As I understood them, there were some GSOC proposals for making 
Sugarizer into an Android shell, that would be make this strategy even 
more viable.



On 6 April 2016 at 23:46, Tony Anderson > wrote:


Actually not. The problem is that I really believe the deployed
system needs to be free, meaning the deployment needs to pay no
subscription fees or royalties to continue to use the device.
Microsoft (and other vendors) suffer from the buy once model. All
of them are trying to find a way to get a stable revenue stream as
the telecoms do. Technically, the 'cloud' is a sham. There is no
way to provide a computer which is 50 times faster than the one in
your hand. So having users execute code on a server is a
non-starter. HTML is ok because the work is done on the client.
Cloud storage is ok because the server action is as a file server,
low processor overhead. Formatting Word documents on a server is
not economical - unless that is your source of revenue.

Durability yes, but the price premium is 2x. There will (and
probably are) a plethora of 'head start' computers with such
packages. However, with Sugar and a school server, we already put
them to shame.

Tony


On 04/07/2016 11:24 AM, Dave Crossland wrote:


On 6 April 2016 at 20:49, Tony Anderson > wrote:

Our need is a deployable device - one that can be purchased
in quantities of 30+.
If we develop a technique to install Sugar on such a device,
that can be done for all of them at
the time of deployment. So, if anyone can find a suitable
tablet with a manageable price (less than $100)
and can install Sugar on it from a usb drive - it would be a
boon.


:D

Sora Edwards-Thro is planning a deployment with the $50
Kindle Fire (an Android derivative). Her intent
is to use Sugarizer. I would recommend adding the GCompris
Android version. What she really needs is
a Sugarized version of the WriteBook activity. So far, no one
has stepped up to take that on. Naturally, the strategic
interest is how well the Kindle supports learning.


Fascinating!
I see

http://gizmodo.com/amazons-50-fire-tablet-is-the-impulse-buy-that-never-e-1731275123
from 2015-09-17:


If what you’re really looking for is durability, though, the
$100 Fire Kids Edition is the one you probably want. Big,
lifeproof rubber bumper, a 2-year no-questions-asked
replacement policy, a kid-friendly web browser you can turn
off or add whitelisted sites to, and 10,000 pre-approved
titles for junior to safely watch.



___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org

http://lists.sugarlabs.org/listinfo/sugar-devel



___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org

http://lists.sugarlabs.org/listinfo/sugar-devel




--
Cheers
Dave


___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel


___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Sebastian Silva]

El 07/04/16 a las 09:15, Sebastian Silva escribió:
> paint for instance (which lacks very important undo functionality for
> instance).

While looking at
http://server.sugarizer.org/activities/Paint.activity/index.html I
realized there is such functionality. Perhaps the version we got from
F-droid is old.
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Sebastian Silva]

I had a real interesting experience rooting my daugther's Fire Kids
Edition that grandma gave for her bday.
I rooted it once. But there was a program to auto update it and then it
had an android 5 further locked down device. Found another root hack out
of it and made sure to stop and completely remove the autoupdater. Most
of this process had to be done offline otherwise the tablet would
download updates and reboot locked.
We removed all of the Amazon apps and installed only free software,
including camera, gallery, file browser and home shell. There are nice
android educational apps from f-droid (the only library/app store we
have installed). The problem with Sugarizer as a shell is that it can
only run sugarizer Activities (Sugar Web Activities) and not the Android
apps. These activities are very limited on a first experience.

I'll try to file bugs as I observe the children play with paint for
instance (which lacks very important undo functionality for instance).
Mariana made a square and a house in TurtleJS but then it crashed. It
seems to do that often and she has not figured out the Journal yet which
is very well hidden.


El 07/04/16 a las 08:03, Dave Crossland escribió:
> Hi Tony
>
> Don't get me wrong, what I am excited about is being able to buy up
> Amazon's subsidised hardware and rip out their Android distro and put
> in a clean one. I'm not sure about putting in a GNU/Linux, a clean
> Android/Linux distro seems more likely to go smoothly. 
>
> As I understood them, there were some GSOC proposals for making
> Sugarizer into an Android shell, that would be make this strategy even
> more viable. 
>
>
> On 6 April 2016 at 23:46, Tony Anderson  > wrote:
>
> Actually not. The problem is that I really believe the deployed
> system needs to be free, meaning the deployment needs to pay no
> subscription fees or royalties to continue to use the device.
> Microsoft (and other vendors) suffer from the buy once model. All
> of them are trying to find a way to get a stable revenue stream as
> the telecoms do. Technically, the 'cloud' is a sham. There is no
> way to provide a computer which is 50 times faster than the one in
> your hand. So having users execute code on a server is a
> non-starter. HTML is ok because the work is done on the client.
> Cloud storage is ok because the server action is as a file server,
> low processor overhead. Formatting Word documents on a server is
> not economical - unless that is your source of revenue.
>
> Durability yes, but the price premium is 2x. There will (and
> probably are) a plethora of 'head start' computers with such
> packages. However, with Sugar and a school server, we already put
> them to shame.
>
> Tony
>
>
> On 04/07/2016 11:24 AM, Dave Crossland wrote:
>>
>> On 6 April 2016 at 20:49, Tony Anderson > > wrote:
>>
>> Our need is a deployable device - one that can be purchased
>> in quantities of 30+.
>> If we develop a technique to install Sugar on such a device,
>> that can be done for all of them at
>> the time of deployment. So, if anyone can find a suitable
>> tablet with a manageable price (less than $100)
>> and can install Sugar on it from a usb drive - it would be a
>> boon.
>>
>>
>> :D
>>  
>>
>> Sora Edwards-Thro is planning a deployment with the $50
>> Kindle Fire (an Android derivative). Her intent
>> is to use Sugarizer. I would recommend adding the GCompris
>> Android version. What she really needs is
>> a Sugarized version of the WriteBook activity. So far, no one
>> has stepped up to take that on. Naturally, the strategic
>> interest is how well the Kindle supports learning.
>>
>>
>> Fascinating!
>>  
>> I see
>> 
>> http://gizmodo.com/amazons-50-fire-tablet-is-the-impulse-buy-that-never-e-1731275123
>> from 2015-09-17:
>>
>>
>> If what you’re really looking for is durability, though, the
>> $100 Fire Kids Edition is the one you probably want. Big,
>> lifeproof rubber bumper, a 2-year no-questions-asked
>> replacement policy, a kid-friendly web browser you can turn
>> off or add whitelisted sites to, and 10,000 pre-approved
>> titles for junior to safely watch.
>>
>>
>>
>> ___
>> Sugar-devel mailing list
>> Sugar-devel@lists.sugarlabs.org
>> 
>> http://lists.sugarlabs.org/listinfo/sugar-devel
>
>
> ___
> Sugar-devel mailing list
> Sugar-devel@lists.sugarlabs.org
> 
> http://lists.sugarlabs.org/listinfo/sugar-devel
>
>
>
>
> -- 
> Cheers
> Dave
>
>
> 

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Dave Crossland]

Hi Tony

Don't get me wrong, what I am excited about is being able to buy up
Amazon's subsidised hardware and rip out their Android distro and put in a
clean one. I'm not sure about putting in a GNU/Linux, a clean Android/Linux
distro seems more likely to go smoothly.

As I understood them, there were some GSOC proposals for making Sugarizer
into an Android shell, that would be make this strategy even more viable.


On 6 April 2016 at 23:46, Tony Anderson  wrote:

> Actually not. The problem is that I really believe the deployed system
> needs to be free, meaning the deployment needs to pay no subscription fees
> or royalties to continue to use the device. Microsoft (and other vendors)
> suffer from the buy once model. All of them are trying to find a way to get
> a stable revenue stream as the telecoms do. Technically, the 'cloud' is a
> sham. There is no way to provide a computer which is 50 times faster than
> the one in your hand. So having users execute code on a server is a
> non-starter. HTML is ok because the work is done on the client. Cloud
> storage is ok because the server action is as a file server, low processor
> overhead. Formatting Word documents on a server is not economical - unless
> that is your source of revenue.
>
> Durability yes, but the price premium is 2x. There will (and probably are)
> a plethora of 'head start' computers with such packages. However, with
> Sugar and a school server, we already put them to shame.
>
> Tony
>
>
> On 04/07/2016 11:24 AM, Dave Crossland wrote:
>
>
> On 6 April 2016 at 20:49, Tony Anderson  wrote:
>
>> Our need is a deployable device - one that can be purchased in quantities
>> of 30+.
>> If we develop a technique to install Sugar on such a device, that can be
>> done for all of them at
>> the time of deployment. So, if anyone can find a suitable tablet with a
>> manageable price (less than $100)
>> and can install Sugar on it from a usb drive - it would be a boon.
>>
>
> :D
>
>
>> Sora Edwards-Thro is planning a deployment with the $50 Kindle Fire (an
>> Android derivative). Her intent
>> is to use Sugarizer. I would recommend adding the GCompris Android
>> version. What she really needs is
>> a Sugarized version of the WriteBook activity. So far, no one has stepped
>> up to take that on. Naturally, the strategic
>> interest is how well the Kindle supports learning.
>>
>
> Fascinating!
>
> I see
> http://gizmodo.com/amazons-50-fire-tablet-is-the-impulse-buy-that-never-e-1731275123
> from 2015-09-17:
>
>
> If what you’re really looking for is durability, though, the $100 Fire
> Kids Edition is the one you probably want. Big, lifeproof rubber bumper, a
> 2-year no-questions-asked replacement policy, a kid-friendly web browser
> you can turn off or add whitelisted sites to, and 10,000 pre-approved
> titles for junior to safely watch.
>
>
>
> ___
> Sugar-devel mailing 
> listSugar-devel@lists.sugarlabs.orghttp://lists.sugarlabs.org/listinfo/sugar-devel
>
>
>
> ___
> Sugar-devel mailing list
> Sugar-devel@lists.sugarlabs.org
> http://lists.sugarlabs.org/listinfo/sugar-devel
>
>


-- 
Cheers
Dave
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Tony Anderson]

Actually not. The problem is that I really believe the deployed system 
needs to be free, meaning the deployment needs to pay no subscription 
fees or royalties to continue to use the device. Microsoft (and other 
vendors) suffer from the buy once model. All of them are trying to find 
a way to get a stable revenue stream as the telecoms do. Technically, 
the 'cloud' is a sham. There is no way to provide a computer which is 50 
times faster than the one in your hand. So having users execute code on 
a server is a non-starter. HTML is ok because the work is done on the 
client. Cloud storage is ok because the server action is as a file 
server, low processor overhead. Formatting Word documents on a server is 
not economical - unless that is your source of revenue.


Durability yes, but the price premium is 2x. There will (and probably 
are) a plethora of 'head start' computers with such packages. However, 
with Sugar and a school server, we already put them to shame.


Tony

On 04/07/2016 11:24 AM, Dave Crossland wrote:


On 6 April 2016 at 20:49, Tony Anderson > wrote:


Our need is a deployable device - one that can be purchased in
quantities of 30+.
If we develop a technique to install Sugar on such a device, that
can be done for all of them at
the time of deployment. So, if anyone can find a suitable tablet
with a manageable price (less than $100)
and can install Sugar on it from a usb drive - it would be a boon.


:D

Sora Edwards-Thro is planning a deployment with the $50 Kindle
Fire (an Android derivative). Her intent
is to use Sugarizer. I would recommend adding the GCompris Android
version. What she really needs is
a Sugarized version of the WriteBook activity. So far, no one has
stepped up to take that on. Naturally, the strategic
interest is how well the Kindle supports learning.


Fascinating!
I see 
http://gizmodo.com/amazons-50-fire-tablet-is-the-impulse-buy-that-never-e-1731275123 
from 2015-09-17:



If what you’re really looking for is durability, though, the $100
Fire Kids Edition is the one you probably want. Big, lifeproof
rubber bumper, a 2-year no-questions-asked replacement policy, a
kid-friendly web browser you can turn off or add whitelisted sites
to, and 10,000 pre-approved titles for junior to safely watch.



___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel


___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Dave Crossland]

On 6 April 2016 at 20:49, Tony Anderson  wrote:

> Our need is a deployable device - one that can be purchased in quantities
> of 30+.
> If we develop a technique to install Sugar on such a device, that can be
> done for all of them at
> the time of deployment. So, if anyone can find a suitable tablet with a
> manageable price (less than $100)
> and can install Sugar on it from a usb drive - it would be a boon.
>

:D


> Sora Edwards-Thro is planning a deployment with the $50 Kindle Fire (an
> Android derivative). Her intent
> is to use Sugarizer. I would recommend adding the GCompris Android
> version. What she really needs is
> a Sugarized version of the WriteBook activity. So far, no one has stepped
> up to take that on. Naturally, the strategic
> interest is how well the Kindle supports learning.
>

Fascinating!

I see
http://gizmodo.com/amazons-50-fire-tablet-is-the-impulse-buy-that-never-e-1731275123
from 2015-09-17:


If what you’re really looking for is durability, though, the $100 Fire Kids
Edition is the one you probably want. Big, lifeproof rubber bumper, a
2-year no-questions-asked replacement policy, a kid-friendly web browser
you can turn off or add whitelisted sites to, and 10,000 pre-approved
titles for junior to safely watch.
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Dave Crossland]

On 6 April 2016 at 20:53, Tony Anderson  wrote:

> One approach is for the 'cloud' server to emulate the school server.
>

100% agree
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Tony Anderson]

One approach is for the 'cloud' server to emulate the school server.

Tony

On 04/06/2016 10:41 PM, Walter Bender wrote:



On Wed, Apr 6, 2016 at 10:09 AM, Dave Crossland > wrote:



On 6 April 2016 at 09:17, Walter Bender > wrote:

the service model that is inherent to the web is really
problematic from the point of view of children's privacy,
security, and freedom.


Naw, it ain't inherent to the model! :)


There is the model and there is how the model is being implemented in 
practice.



Off the top of my head, here are 3 projects that adapt the
software freedom movement to the model:

http://sandstorm.io
https://owncloud.org
https://unhosted.org


UnHosted is pretty interesting.



"There is no cloud, just other people's computers," says the FSFE.

"There is no cloud, just other people's computers - that you can
rent," I say.

Some people on this list surely own their own home, free and clear.

Many more people on this list surely rent a home.

People who rent a container instance on someone else's computer
and run libre software on it are enjoying privacy, security and
freedom.


I agree. If we design for the Cloud as if it were simply an extension 
of the local file system we are "renting" then it is not much 
different than what we have right now. But as soon as we start relying 
on remote services, most of which as not free/libre, we run into the 
problems I was alluding to. So it behooves us to show how it can be 
done in a way that respects freedom.


-walter

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



___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel


___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Tony Anderson]

Our need is a deployable device - one that can be purchased in 
quantities of 30+.
If we develop a technique to install Sugar on such a device, that can be 
done for all of them at
the time of deployment. So, if anyone can find a suitable tablet with a 
manageable price (less than $100)

and can install Sugar on it from a usb drive - it would be a boon.

Sora Edwards-Thro is planning a deployment with the $50 Kindle Fire (an 
Android derivative). Her intent
is to use Sugarizer. I would recommend adding the GCompris Android 
version. What she really needs is
a Sugarized version of the WriteBook activity. So far, no one has 
stepped up to take that on. Naturally, the strategic

interest is how well the Kindle supports learning.

Tony

On 04/06/2016 10:03 PM, Dave Crossland wrote:


Hey Jonas

A nice surprise to see you here :)

On 6 April 2016 at 09:29, Jonas Smedegaard > wrote:


Quoting Dave Crossland (2016-04-06 14:42:57)
> On 6 April 2016 at 02:59,  wrote:
>
>> You can install GNU on a chromebook, you can install GNU on
computer,
>> you can install GNU on some tablets.  Those are the pedagogic
devices
>> of now and the future.  Those run faster with Gtk than with WebKit.
>
>
> Will devices running GNU and Sugar desktops get into the hands
of most
> of the world's poor children in the next 10 years?

Devices *not* running GNU (and therefore Sugar desktop) are likely
devices not in the control of its owner.

Underlying question is therefore if it matters that the device "in the
hands" of those kids is in their own control or not.

Only if it does not matter is it relevant to consider throwing
away the
work done developing Sugar-atop-a-POSIX-system and instead invest only
in developing Sugar-atop-a-web-browser.


It would be nice if the devices came with GNU preinstalled. Sure! :)

But, they won't.

So, how will GNU get on them?

If the owners learn to love computers, as you and I love computers, 
then they will care to assert control and run GNU.


That is, afterall, how we first came to be running GNU. Not sure about 
you old boy, but I was 16 when I installed my first distro. The 
learning experience changed my life.


Sugarizer presents a viable way to help them learn to love computers 
with the computers they have.


GNU/Sugar is the platform kids deserves, but not the one they need 
right now.


The way I see it, the OLPC strategy of shipping devices with GNU on 
them isn't reaching most kids.


Sugar Labs first press release stated an intention to court other 
hardware vendors. AFAIK there were and are none, and if Sugar Labs got 
into selling hardware it still wouldn't matter, and would just compete 
with what is left of OLPC. The Infinity laptop being built in 
Australia is great, all power to them, but it shouldn't be the leading 
strategy.


The SOAS strategy of booting devices into GNU, and offering Activities 
from ASLO, also fails to reach most kids, cause in MOST cases their 
devices can't boot SOAS anyway. And as I said the other day, the cold 
hard fact is that ASLO traffic is down 90% over the last 3 years.


So look, don't get me wrong. For this year, and the next year, and the 
next year, the way Sugarizer Activities are integrated into GNU/Sugar 
is GREAT. I am NOT advocating that we "throw out the great 
technologies we have now," and I'm sorry if I was unclear in my 
admittedly provocative thinkpiece; that is simply not how Sugar + 
Sugarizer works today, so its nothing to be worried about :)


What I _am_ advocating for is a vision, a mission, and a strategy for 
what Sugar Labs will do over the next 1, 3, 5, 10 years with the mission.


As clear as I can distill it, it is this:

Sugar is high quality software for children to learn with,
especially younger children.


If we don't meet kids where they are at, we will not convert as many 
of them into computer lovers as we could. Focusing on the development 
of GNU/Sugar as the primary platform for the next 10 years is not 
meeting kids where they are at. Dropping development of GNU/Sugar 
would be immoral, given 100,000s of kids using it daily. But to get to 
1,000,000,000 kids, it can not be the primary focus.


Only web based software can reach a billion kids.

So Sugar has to become high quality web based software for children to 
learn with.




___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel


___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Tony Anderson]

Thanks, Walter

Beautifully said.

Tony

On 04/06/2016 09:17 PM, Walter Bender wrote:

I'm going to land squarely in the middle on this issue.

I agree with Sam that what we have to offer in the world of a 
GNU/Linux desktop is far far better than any alternatives I have seen. 
The opportunity for growth there is demonstrably great. We have pretty 
decent offerings in Fedora, Debian, and its relatively popular Ubuntu 
instance. And I don't think the GNU desktop is going to disappear as 
rapidly as the pundits predict, despite the popularity of Android and 
iStuff,  (And I think there are some serious problems of pedagogy in 
the solutions offered in the smartphone space.) The GNU desktop is 
going to be a relatively small market for the foreseeable future, but 
one where we can show thought leadership, reach some kids directly, 
and influence the rest of the ed tech industry through the tangible 
demonstration of our ideas. One spark of hope is that the Maker 
Movement -- the ed tech idea de jour -- is to a large extent Linux 
based. Might make sense to revisit improving the Sugar experience on 
RPi and other platforms popular with makers.


Chromebooks are interesting in that (a) they can run GNU and 
consequently native Sugar quite well -- but I doubt too many schools 
will go down that path; and (b) you can almost treat them like 
computers in that the form factor is bit more friendly to programming, 
word processing, and other tool-oriented activities. That said, I hear 
rumors that Chrome OS will be subsumed by Android, so it is not 
obvious that it is a long-term viable solution any more than GNU. And 
the service model that is inherent to the web is really problematic 
from the point of view of children's privacy, security, and freedom.


That said, there is something to be said for trying to meet people 
halfway. The browser is ubiquitous. If we can develop within the 
context of Sugar desktop and the browser, it is to a large extent a 
win-win. This is why I have been wrestling with JavaScript in my newer 
activities. (For similar reasons, I have tried to make most of my 
activities run in GNOME as well as Sugar.) It opens some doors. While 
not perfect, the Sugar JS activity experience is decent. And hopefully 
Lionel's effort will help us reach kids we would not have otherwise 
reached, even with a lesser solution than GNU. It is important that as 
we develop in this space we keep in mind some principles, such as 
making our source code readable, making sure things can run locally, 
focusing on tools rather than apps, providing explicit mechanisms for 
reflect, etc.


-walter

On Wed, Apr 6, 2016 at 2:59 AM, > 
wrote:


Hi Dave,

On Wed, Apr 6, 2016 at 10:46 AM, Dave Crossland > wrote:

I would be happy if by 2020 the "classic" Sugar desktop was
totally gone. Zero Python! In its place could be a laptop OS
derived from ChromiumOS, plus a nodejs web server serving on
localhost that is stuffed full of activities and content.


Why?  Why do we throw out the great technologies we have now?  Why
do we waste out time replacing Telepathy (amazing back end for
collaboration)?  Why do we waste our time replacing
GtkSourceView?  AbiWord?  WebKitGtk?  Gtk? These are great
technologies.  Sure they are not the current trend.  But unlike
your proposed nodejs server, we don't end up in callback hell.

We have a technology stack that we have used for over 10 years. 
Those people who OLPC paid to start writing sugar made good

choices; they left us with a great foundation even as OLPC down
sizes.  It works great across keyboard, mouse and touch (can
sugarizer even show a tooltip on long touch?).  It works great on
slow computers (my trusty old Core2Duo laptop runs sugar faster
than Sugarizer/webkit).  It works great off-line (collaboration
over salut doesn't need a centeral server). And all the activities
are written in python now.

Why waste time to javashit it?  You can install GNU on a
chromebook, you can install GNU on computer, you can install GNU
on some tablets.  Those are the pedagogic devices of now and the
future.  Those run faster with Gtk than with WebKit.

Don't waste time.

Thanks,
Sam

[GNU in this post refers to GNU/Linux]




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



___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel


___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[James Cameron]

Back in 2006, when choice was to use C++, C, Qt or Gtk, or even
php-gtk (!), one side said use C because we had it now, it would work,
and work really fast.

Python was chosen because it worked, and was easily accessible for
modification by the learner.

Silly, thought I, 'cause who uses Python for system stuff?  Would it
scale?  Can we afford to lose that much memory?  Why compile on the
fly?  Can it have predictable interactive response times in the face
of garbage collection?

But Python could run everywhere, we were told.  And they were right.
It did run on Windows, Mac OS X, and the various Unix.

Since, the dependencies of the Python Sugar have grown to fit Linux
only, with most not available on other platforms.

Since, new platforms have arisen without the dependencies.

Horse has bolted; the effort required to add those dependencies
cleanly to the new platforms exceeds the effort to switch language.

I'm glad the new language was not Java.  I'm okay with it being
JavaScript; it feels like ten years ago all over again.  I do hope it
won't depend on npm or left-pad!

-- 
James Cameron
http://quozl.netrek.org/
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Lionel Laské]

Waooo, what amazing discussion in this thread !
I don't remember a such rich debate since few month. Sorry to say that but
it's probably more interesting for lot of people than our SLOB internal
debate :-)

BTW I'm happy to be one of the guy that launched the debate here with my
"Vision proposal". Thanks a lot Dave to be the guy with the outside view on
Sugar. You're always asking the good questions !

Now, because I'm the author of Sugarizer, I just want to add my two cents
on the thread to explain my point of view:

1) I don't hate Python ! I've written few activities in Python myself.
2) I didn't choose JavaScript/HTML because it's better than Python/Gtk.
I've just decided to use JavaScript/HTML because it could run everywhere.
3) I don't want to replace Sugar Desktop by Sugarizer. In my mind Sugar is
just a platform for activities. I just want that all activities could run
both on Sugar Desktop and Sugarizer. Remind you that all new Sugarizer
activity is a new Sugar activity: we don't waste time: we're investing time.

My only objective with Sugarizer is to attract more and more people to
Sugar.
I will be fully happy if Sugarizer is forever a "Sugar lite" and that you
need Sugar Desktop to have the better Sugar experience.
I will be fully happy if a child that discovered Sugarizer on an Android
tablet decide few months later to explore a bit and install Sugar Desktop
on a PC.
But without Sugarizer on Android this could never happen.
Every child has the right to benefit to Sugar, even a child with an iPad
(shocking !)

Best regards from France.

   Lionel


Date: Wed, 06 Apr 2016 16:59:29 +1000
> From: sam@sam.today
> To: Dave Crossland 
> Cc: Sugar-dev Devel , Walter Bender
> 
> Subject: [Sugar-devel] Python is good - don't waste time (was Re: The
> future of Sugar on XO-1s)
> Message-ID: <1459925969.172...@smtp.gmail.com>
> Content-Type: text/plain; charset="utf-8"; Format="flowed"
>
> Hi Dave,
>
> On Wed, Apr 6, 2016 at 10:46 AM, Dave Crossland  wrote:
> > I would be happy if by 2020 the "classic" Sugar desktop was totally
> > gone. Zero Python! In its place could be a laptop OS derived from
> > ChromiumOS, plus a nodejs web server serving on localhost that is
> > stuffed full of activities and content.
>
> Why?  Why do we throw out the great technologies we have now?  Why do
> we waste out time replacing Telepathy (amazing back end for
> collaboration)?  Why do we waste our time replacing GtkSourceView?
> AbiWord?  WebKitGtk?  Gtk?  These are great technologies.  Sure they
> are not the current trend.  But unlike your proposed nodejs server, we
> don't end up in callback hell.
>
> We have a technology stack that we have used for over 10 years.  Those
> people who OLPC paid to start writing sugar made good choices; they
> left us with a great foundation even as OLPC down sizes.  It works
> great across keyboard, mouse and touch (can sugarizer even show a
> tooltip on long touch?).  It works great on slow computers (my trusty
> old Core2Duo laptop runs sugar faster than Sugarizer/webkit).  It works
> great off-line (collaboration over salut doesn't need a centeral
> server). And all the activities are written in python now.
>
> Why waste time to javashit it?  You can install GNU on a chromebook,
> you can install GNU on computer, you can install GNU on some tablets.
> Those are the pedagogic devices of now and the future.  Those run
> faster with Gtk than with WebKit.
>
> Don't waste time.
>
> Thanks,
> Sam
>
> [GNU in this post refers to GNU/Linux]
> -- next part --
> An HTML attachment was scrubbed...
> URL: <
> http://lists.sugarlabs.org/archive/sugar-devel/attachments/20160406/1d59b03e/attachment.html
> >
>
> --
>
> Subject: Digest Footer
>
> ___
> Sugar-devel mailing list
> Sugar-devel@lists.sugarlabs.org
> http://lists.sugarlabs.org/listinfo/sugar-devel
>
>
> --
>
> End of Sugar-devel Digest, Vol 90, Issue 30
> ***
>
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Dave Crossland]

On 6 April 2016 at 10:41, Walter Bender  wrote:

>
>
> On Wed, Apr 6, 2016 at 10:09 AM, Dave Crossland  wrote:
>
>>
>> On 6 April 2016 at 09:17, Walter Bender  wrote:
>>
>>> the service model that is inherent to the web is really problematic from
>>> the point of view of children's privacy, security, and freedom.
>>
>>
>> Naw, it ain't inherent to the model! :)
>>
>
> There is the model and there is how the model is being implemented in
> practice.
>

Sure! I 100% agree that the service model that is _common_ to the web is
really problematic from the point of view of children's privacy, security,
and freedom.

But the model itself is ubiquitous because it has many practical
advantages, and for the sfm, since it can be adapted, it should be instead
of rejected.



> People who rent a container instance on someone else's computer and run
>> libre software on it are enjoying privacy, security and freedom.
>>
>
> I agree. If we design for the Cloud as if it were simply an extension of
> the local file system we are "renting" then it is not much different than
> what we have right now. But as soon as we start relying on remote services,
> most of which as not free/libre, we run into the problems I was alluding
> to. So it behooves us to show how it can be done in a way that respects
> freedom.
>

100% agree

Its awesome that the systems committee has already been moving SL
self-hosted systems to a container based approach :)

-- 
Cheers
Dave
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Jonas Smedegaard]

Quoting Walter Bender (2016-04-06 16:41:07)
> On Wed, Apr 6, 2016 at 10:09 AM, Dave Crossland  wrote:
>> Off the top of my head, here are 3 projects that adapt the software 
>> freedom movement to the model:
>>
>> http://sandstorm.io
>> https://owncloud.org
>> https://unhosted.org
>>
>
> UnHosted is pretty interesting.

Unhosted is rather vague.

I find SoLiD quite interesting: 
https://github.com/solid/solid-spec#overview

 - Jonas

-- 
 * Jonas Smedegaard - idealist & Internet-arkitekt
 * Tlf.: +45 40843136  Website: http://dr.jones.dk/

 [x] quote me freely  [ ] ask before reusing  [ ] keep private


signature.asc
Description: signature
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Walter Bender]

On Wed, Apr 6, 2016 at 10:09 AM, Dave Crossland  wrote:

>
> On 6 April 2016 at 09:17, Walter Bender  wrote:
>
>> the service model that is inherent to the web is really problematic from
>> the point of view of children's privacy, security, and freedom.
>
>
> Naw, it ain't inherent to the model! :)
>

There is the model and there is how the model is being implemented in
practice.

>
> Off the top of my head, here are 3 projects that adapt the software
> freedom movement to the model:
>
> http://sandstorm.io
> https://owncloud.org
> https://unhosted.org
>

UnHosted is pretty interesting.

>
>
> "There is no cloud, just other people's computers," says the FSFE.
>
> "There is no cloud, just other people's computers - that you can rent," I
> say.
>
> Some people on this list surely own their own home, free and clear.
>
> Many more people on this list surely rent a home.
>
> People who rent a container instance on someone else's computer and run
> libre software on it are enjoying privacy, security and freedom.
>

I agree. If we design for the Cloud as if it were simply an extension of
the local file system we are "renting" then it is not much different than
what we have right now. But as soon as we start relying on remote services,
most of which as not free/libre, we run into the problems I was alluding
to. So it behooves us to show how it can be done in a way that respects
freedom.

-walter

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

___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Dave Crossland]

On 6 April 2016 at 09:17, Walter Bender  wrote:

> If we can develop within the context of Sugar desktop and the browser, it
> is to a large extent a win-win.


That is the 'great hack' I referred to earlier - I think its copyleft-level
clever stuff :)

I'm not so familiar with GNOME 3, but its JS scripted. Nor do I have an
XO-4. How well does GNOME 3 run on XO-4s?
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Dave Crossland]

On 6 April 2016 at 09:17, Walter Bender  wrote:

> the service model that is inherent to the web is really problematic from
> the point of view of children's privacy, security, and freedom.


Naw, it ain't inherent to the model! :)

Off the top of my head, here are 3 projects that adapt the software freedom
movement to the model:

http://sandstorm.io
https://owncloud.org
https://unhosted.org

"There is no cloud, just other people's computers," says the FSFE.

"There is no cloud, just other people's computers - that you can rent," I
say.

Some people on this list surely own their own home, free and clear.

Many more people on this list surely rent a home.

People who rent a container instance on someone else's computer and run
libre software on it are enjoying privacy, security and freedom.
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Dave Crossland]

Hey Jonas

A nice surprise to see you here :)

On 6 April 2016 at 09:29, Jonas Smedegaard  wrote:

> Quoting Dave Crossland (2016-04-06 14:42:57)
> > On 6 April 2016 at 02:59,  wrote:
> >
> >> You can install GNU on a chromebook, you can install GNU on computer,
> >> you can install GNU on some tablets.  Those are the pedagogic devices
> >> of now and the future.  Those run faster with Gtk than with WebKit.
> >
> >
> > Will devices running GNU and Sugar desktops get into the hands of most
> > of the world's poor children in the next 10 years?
>
> Devices *not* running GNU (and therefore Sugar desktop) are likely
> devices not in the control of its owner.
>
> Underlying question is therefore if it matters that the device "in the
> hands" of those kids is in their own control or not.
>
> Only if it does not matter is it relevant to consider throwing away the
> work done developing Sugar-atop-a-POSIX-system and instead invest only
> in developing Sugar-atop-a-web-browser.
>

It would be nice if the devices came with GNU preinstalled. Sure! :)

But, they won't.

So, how will GNU get on them?

If the owners learn to love computers, as you and I love computers, then
they will care to assert control and run GNU.

That is, afterall, how we first came to be running GNU. Not sure about you
old boy, but I was 16 when I installed my first distro. The learning
experience changed my life.

Sugarizer presents a viable way to help them learn to love computers with
the computers they have.

GNU/Sugar is the platform kids deserves, but not the one they need right
now.

The way I see it, the OLPC strategy of shipping devices with GNU on them
isn't reaching most kids.

Sugar Labs first press release stated an intention to court other hardware
vendors. AFAIK there were and are none, and if Sugar Labs got into selling
hardware it still wouldn't matter, and would just compete with what is left
of OLPC. The Infinity laptop being built in Australia is great, all power
to them, but it shouldn't be the leading strategy.

The SOAS strategy of booting devices into GNU, and offering Activities from
ASLO, also fails to reach most kids, cause in MOST cases their devices
can't boot SOAS anyway. And as I said the other day, the cold hard fact is
that ASLO traffic is down 90% over the last 3 years.

So look, don't get me wrong. For this year, and the next year, and the next
year, the way Sugarizer Activities are integrated into GNU/Sugar is GREAT.
I am NOT advocating that we "throw out the great technologies we have now,"
and I'm sorry if I was unclear in my admittedly provocative thinkpiece;
that is simply not how Sugar + Sugarizer works today, so its nothing to be
worried about :)

What I _am_ advocating for is a vision, a mission, and a strategy for what
Sugar Labs will do over the next 1, 3, 5, 10 years with the mission.

As clear as I can distill it, it is this:

Sugar is high quality software for children to learn with, especially
younger children.


If we don't meet kids where they are at, we will not convert as many of
them into computer lovers as we could. Focusing on the development of
GNU/Sugar as the primary platform for the next 10 years is not meeting kids
where they are at. Dropping development of GNU/Sugar would be immoral,
given 100,000s of kids using it daily. But to get to 1,000,000,000 kids, it
can not be the primary focus.

Only web based software can reach a billion kids.

So Sugar has to become high quality web based software for children to
learn with.
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Dave Crossland]

ChromeOS is a GNU distribution :)
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Jonas Smedegaard]

Quoting Dave Crossland (2016-04-06 14:42:57)
> On 6 April 2016 at 02:59,  wrote:
>
>> You can install GNU on a chromebook, you can install GNU on computer, 
>> you can install GNU on some tablets.  Those are the pedagogic devices 
>> of now and the future.  Those run faster with Gtk than with WebKit.
>
>
> Will devices running GNU and Sugar desktops get into the hands of most 
> of the world's poor children in the next 10 years?

Devices *not* running GNU (and therefore Sugar desktop) are likely 
devices not in the control of its owner.

Underlying question is therefore if it matters that the device "in the 
hands" of those kids is in their own control or not.

Only if it does not matter is it relevant to consider throwing away the 
work done developing Sugar-atop-a-POSIX-system and instead invest only 
in developing Sugar-atop-a-web-browser.

 - Jonas

-- 
 * Jonas Smedegaard - idealist & Internet-arkitekt
 * Tlf.: +45 40843136  Website: http://dr.jones.dk/

 [x] quote me freely  [ ] ask before reusing  [ ] keep private


signature.asc
Description: signature
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Walter Bender]

I'm going to land squarely in the middle on this issue.

I agree with Sam that what we have to offer in the world of a GNU/Linux
desktop is far far better than any alternatives I have seen. The
opportunity for growth there is demonstrably great. We have pretty decent
offerings in Fedora, Debian, and its relatively popular Ubuntu instance.
And I don't think the GNU desktop is going to disappear as rapidly as the
pundits predict, despite the popularity of Android and iStuff,  (And I
think there are some serious problems of pedagogy in the solutions offered
in the smartphone space.) The GNU desktop is going to be a relatively small
market for the foreseeable future, but one where we can show thought
leadership, reach some kids directly, and influence the rest of the ed tech
industry through the tangible demonstration of our ideas. One spark of hope
is that the Maker Movement -- the ed tech idea de jour -- is to a large
extent Linux based. Might make sense to revisit improving the Sugar
experience on RPi and other platforms popular with makers.

Chromebooks are interesting in that (a) they can run GNU and consequently
native Sugar quite well -- but I doubt too many schools will go down that
path; and (b) you can almost treat them like computers in that the form
factor is bit more friendly to programming, word processing, and other
tool-oriented activities. That said, I hear rumors that Chrome OS will be
subsumed by Android, so it is not obvious that it is a long-term viable
solution any more than GNU. And the service model that is inherent to the
web is really problematic from the point of view of children's privacy,
security, and freedom.

That said, there is something to be said for trying to meet people halfway.
The browser is ubiquitous. If we can develop within the context of Sugar
desktop and the browser, it is to a large extent a win-win. This is why I
have been wrestling with JavaScript in my newer activities. (For similar
reasons, I have tried to make most of my activities run in GNOME as well as
Sugar.) It opens some doors. While not perfect, the Sugar JS activity
experience is decent. And hopefully Lionel's effort will help us reach kids
we would not have otherwise reached, even with a lesser solution than GNU.
It is important that as we develop in this space we keep in mind some
principles, such as making our source code readable, making sure things can
run locally, focusing on tools rather than apps, providing explicit
mechanisms for reflect, etc.

-walter

On Wed, Apr 6, 2016 at 2:59 AM,  wrote:

> Hi Dave,
>
> On Wed, Apr 6, 2016 at 10:46 AM, Dave Crossland  wrote:
>
> I would be happy if by 2020 the "classic" Sugar desktop was totally gone.
> Zero Python! In its place could be a laptop OS derived from ChromiumOS,
> plus a nodejs web server serving on localhost that is stuffed full of
> activities and content.
>
>
> Why?  Why do we throw out the great technologies we have now?  Why do we
> waste out time replacing Telepathy (amazing back end for collaboration)?
> Why do we waste our time replacing GtkSourceView?  AbiWord?  WebKitGtk?
> Gtk?  These are great technologies.  Sure they are not the current trend.
> But unlike your proposed nodejs server, we don't end up in callback hell.
>
> We have a technology stack that we have used for over 10 years.  Those
> people who OLPC paid to start writing sugar made good choices; they left us
> with a great foundation even as OLPC down sizes.  It works great across
> keyboard, mouse and touch (can sugarizer even show a tooltip on long
> touch?).  It works great on slow computers (my trusty old Core2Duo laptop
> runs sugar faster than Sugarizer/webkit).  It works great off-line
> (collaboration over salut doesn't need a centeral server). And all the
> activities are written in python now.
>
> Why waste time to javashit it?  You can install GNU on a chromebook, you
> can install GNU on computer, you can install GNU on some tablets.  Those
> are the pedagogic devices of now and the future.  Those run faster with Gtk
> than with WebKit.
>
> Don't waste time.
>
> Thanks,
> Sam
>
> [GNU in this post refers to GNU/Linux]
>



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

___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

Re: [Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[Tony Anderson]

My dream is that someone ports gnu/linux to run on low-cost tablets. 
Rabi Karmacharya, with lots of experience, believes we need at least a 
10in screen.
I think the tablet needs to be in a leather (or substantially 
ruggedized) case with a keyboard (wired usb or wireless).


Suppose someone approached you with the idea of sponsoring a one laptop 
per child deployment in the developing world. This would

require 200 laptops. What platform would you recommend?

How would an Android tablet work without any possibility of connection 
to the internet?


I believe that a deployment is a system. Tediously, I have claimed the 
essential need of a school server to provide the content which
is not available without the internet. The deployment must have a way to 
charge the laptops and to power the server and its local network.


For me, the most difficult thing has been to find a way to inform 
non-technical users on how to take good advantage of the laptops in the 
absence of an on-site system adminstrator. Addressing the problem was 
why I was so intrigued by the 'onboarding' idea. Typically, I get 1-2 
weeks at a deployment per year. My model is a ball which needs a shove 
periodically to counter friction and keep it moving.


Luckily there is no pressure to move away from standard software on the 
server side (CentOS).


Tony

On 04/06/2016 02:59 PM, sam@sam.today wrote:

Hi Dave,

On Wed, Apr 6, 2016 at 10:46 AM, Dave Crossland  wrote:
I would be happy if by 2020 the "classic" Sugar desktop was totally 
gone. Zero Python! In its place could be a laptop OS derived from 
ChromiumOS, plus a nodejs web server serving on localhost that is 
stuffed full of activities and content.


Why?  Why do we throw out the great technologies we have now?  Why do 
we waste out time replacing Telepathy (amazing back end for 
collaboration)?  Why do we waste our time replacing GtkSourceView? 
 AbiWord?  WebKitGtk?  Gtk?  These are great technologies.  Sure they 
are not the current trend.  But unlike your proposed nodejs server, we 
don't end up in callback hell.


We have a technology stack that we have used for over 10 years.  Those 
people who OLPC paid to start writing sugar made good choices; they 
left us with a great foundation even as OLPC down sizes.  It works 
great across keyboard, mouse and touch (can sugarizer even show a 
tooltip on long touch?).  It works great on slow computers (my trusty 
old Core2Duo laptop runs sugar faster than Sugarizer/webkit).  It 
works great off-line (collaboration over salut doesn't need a centeral 
server). And all the activities are written in python now.


Why waste time to javashit it?  You can install GNU on a chromebook, 
you can install GNU on computer, you can install GNU on some tablets. 
 Those are the pedagogic devices of now and the future.  Those run 
faster with Gtk than with WebKit.


Don't waste time.

Thanks,
Sam

[GNU in this post refers to GNU/Linux]


___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel


___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel

[Sugar-devel] Python is good - don't waste time (was Re: The future of Sugar on XO-1s)

[sam]

Hi Dave,

On Wed, Apr 6, 2016 at 10:46 AM, Dave Crossland  wrote:
I would be happy if by 2020 the "classic" Sugar desktop was totally 
gone. Zero Python! In its place could be a laptop OS derived from 
ChromiumOS, plus a nodejs web server serving on localhost that is 
stuffed full of activities and content.


Why?  Why do we throw out the great technologies we have now?  Why do 
we waste out time replacing Telepathy (amazing back end for 
collaboration)?  Why do we waste our time replacing GtkSourceView?  
AbiWord?  WebKitGtk?  Gtk?  These are great technologies.  Sure they 
are not the current trend.  But unlike your proposed nodejs server, we 
don't end up in callback hell.


We have a technology stack that we have used for over 10 years.  Those 
people who OLPC paid to start writing sugar made good choices; they 
left us with a great foundation even as OLPC down sizes.  It works 
great across keyboard, mouse and touch (can sugarizer even show a 
tooltip on long touch?).  It works great on slow computers (my trusty 
old Core2Duo laptop runs sugar faster than Sugarizer/webkit).  It works 
great off-line (collaboration over salut doesn't need a centeral 
server). And all the activities are written in python now.


Why waste time to javashit it?  You can install GNU on a chromebook, 
you can install GNU on computer, you can install GNU on some tablets.  
Those are the pedagogic devices of now and the future.  Those run 
faster with Gtk than with WebKit.


Don't waste time.

Thanks,
Sam

[GNU in this post refers to GNU/Linux]
___
Sugar-devel mailing list
Sugar-devel@lists.sugarlabs.org
http://lists.sugarlabs.org/listinfo/sugar-devel