Hi, hearing the news that much of the Sugar dev team is fired I now wonder if there is any point in working on Sugar integration anymore. I have started on a JournalMorph where one could obtain pictures and sounds etc. from the Journal to link in or import into a project, but now I have slight doubts about if there is any use for such a thing ? Do you guys have any thoughts on the future ?
Karl
Keep your asset portable, whatever the host you are safe, and more importantly your user will be safe as well. This is why I like so much Smalltalk, so easily portable, that I want to keep away from the things underneath...
Hilaire
2009/1/10 Karl Ramberg karlramberg@gmail.com:
Hi, hearing the news that much of the Sugar dev team is fired I now wonder if there is any point in working on Sugar integration anymore. I have started on a JournalMorph where one could obtain pictures and sounds etc. from the Journal to link in or import into a project, but now I have slight doubts about if there is any use for such a thing ? Do you guys have any thoughts on the future ?
Karl _______________________________________________ Etoys mailing list Etoys@lists.laptop.org http://lists.laptop.org/listinfo/etoys
OLPC did layoff much of their engineering team, but that was almost entirely unrelated to Sugar. Sugar spun out of OLPC six months ago and is a growing: it is now available on most major GNU/Linux distributions and many more hardware platforms than just the OLPC-XO-1 laptop. (Almost anything that can run Linux can run Sugar/Etoys.)
-walter
On Sat, Jan 10, 2009 at 8:18 AM, Karl Ramberg karlramberg@gmail.com wrote:
Hi, hearing the news that much of the Sugar dev team is fired I now wonder if there is any point in working on Sugar integration anymore. I have started on a JournalMorph where one could obtain pictures and sounds etc. from the Journal to link in or import into a project, but now I have slight doubts about if there is any use for such a thing ? Do you guys have any thoughts on the future ?
Karl _______________________________________________ Etoys mailing list Etoys@lists.laptop.org http://lists.laptop.org/listinfo/etoys
<straightThoughMode>
I always feel sad how OLPC missed the Smalltalk/Morph/Etoys opportunity, both from the technical point of view but also and more importantly from the educative point of view.
If you look at the sugarized applications, the educative application are fragmented, unrelated and they can't be connected.
For examples, why can't I open the camera recorder in my abiword activity? Then also why can't I get data from the measure activity into an another unrelated activity to manipulate the data? Or why can't I control the measure activity from another activity? Why can't a teachers/developers create educative activities out of third party activities?
These drawback stop you to be creative when you use a computer. They stop teacher or developers to build easily educative activities. And given the intended goal of XO -- you all know it -- it is a serious design problem.
At this level Sugar is not better to the other UI framework around and it is disappointing.
But where it is even more disappointing it is when you are realizing all these features you want to have to be creative with computer are just siting there in Smalltalk/Morph/Etoys. It need polishing, some reengineering, but the design is there from the beginning with years of experience.
But sill we are forced to go the uncreative way.
Disappointing
</straightThoughMode>
Yes and no.
I agree with most of what you are saying, especially in regard to the merits of Etoys, but not about Sugar. While Sugar (and Python) is not as mature as Etoys (and Smalltalk), in fact we have given much consideration to user-level and activity-level collaboration, but we deliberately used a model of intermediation with the Sugar Journal mechanism. This was in large part driven by security needs, but also to help simplify development. Our goal with Sugar is to have many of the affordances of Etoys spread across the entire system, while accommodating to the extent we can more mainstream, legacy approaches.
We have a ways to go.
In regard to your specific examples, you can easily import the output of Record into Abiword and you can run the two Activities in parallel and share media objects between multiple users. What is the use case for a more tight coupling.
In regard to Measure, I think you have a stronger case to make. While Measure does export its data to the Journal or the clipboard, it is not apparent how to stream Measure data directly into another Activity in real time. I could image wanting to pipe Measure into SocialCalc for example. To date, these sorts of things have been done on an activity by activity basis, for example, Arjun Sarwal, the author of Measure, integrated data capture directly into Turtle Art. But that was a developer, not an end user. As I said, we have a ways to go.
-walter
On Sat, Jan 10, 2009 at 9:04 AM, Hilaire Fernandes hilaire@ofset.org wrote:
<straightThoughMode>
I always feel sad how OLPC missed the Smalltalk/Morph/Etoys opportunity, both from the technical point of view but also and more importantly from the educative point of view.
If you look at the sugarized applications, the educative application are fragmented, unrelated and they can't be connected.
For examples, why can't I open the camera recorder in my abiword activity? Then also why can't I get data from the measure activity into an another unrelated activity to manipulate the data? Or why can't I control the measure activity from another activity? Why can't a teachers/developers create educative activities out of third party activities?
These drawback stop you to be creative when you use a computer. They stop teacher or developers to build easily educative activities. And given the intended goal of XO -- you all know it -- it is a serious design problem.
At this level Sugar is not better to the other UI framework around and it is disappointing.
But where it is even more disappointing it is when you are realizing all these features you want to have to be creative with computer are just siting there in Smalltalk/Morph/Etoys. It need polishing, some reengineering, but the design is there from the beginning with years of experience.
But sill we are forced to go the uncreative way.
Disappointing
</straightThoughMode> _______________________________________________ Etoys mailing list Etoys@lists.laptop.org http://lists.laptop.org/listinfo/etoys
2009/1/10 Walter Bender walter.bender@gmail.com:
In regard to your specific examples, you can easily import the output of Record into Abiword and you can run the two Activities in parallel and share media objects between multiple users. What is the use case for a more tight coupling.
That's the point, we don't know exactly what is the point of more tight coupling. And it is from this point we can have creative way of using the computer. Now, in this previous example, if I mention an interactive geometry activity instead of the Record activity, the coupling will be more obvious to you as an educated adult: coupling text activities and interactive geometry canvas may look more obvious. Nevertheless can we pretend coupling tightly Abiword and Record is (will be) irrelevant? This is not how I perceive been creative with computer.
I will take another example. A few years ago when I implemented user programming in DrGeo1.1, I had no idea of the use of such features. I only know this feature will be an open door for creative users. Indeed several months later, a math teacher from Italy started to use this feature in unique way useful to him to expose historical math problems (a few examples in http://www.ofset.org/uploadfiles/64/download/libro.pdf), another one from Taiwan where modeling mechanic problem with it. All these uses were unthinkable in their exact instantiation, but I was clearly expecting this instrumentation process .
activity by activity basis, for example, Arjun Sarwal, the author of Measure, integrated data capture directly into Turtle Art. But that was a developer, not an end user. As I said, we have a ways to go.
Yes, it is where the Smalltalk/Morph/Etoys has excelled for years. It pushed this capability in the user land... That's indeed a huge contribution. Not only Smalltalk/Morph/Etoys makes it possible for user, but moreover it makes it possible at two levels: - at the Etoys level, to interlink artifacts - at a lower Smalltalk level (however still a very higher level compare to other solution) to design additional artifacts for example.
And yes all this in a very user-friendly environment. Well may be this one is really the downside of Smalltalk/Morph/Etoys. However project like Smalltalk Pharo with much fewer resources than Etoys/Sugar/OLPC shows it can be easily updated and improved.
However all this creative capability in the user land got a price, the documentation. You really need good documentation to take a chance to get it used. You need to provide the key to play with the lock, or it will be show stopper. And better provide such documentation directly in the laptop.
Hilaire
Hilaire Fernandes wrote:
<straightThoughMode>
I always feel sad how OLPC missed the Smalltalk/Morph/Etoys opportunity, both from the technical point of view but also and more importantly from the educative point of view.
I hope you are aware that I have been developing a Smalltalk-only computer for education for many years (way before OLPC was started). Until a couple of months ago it was just me and I was working on my own Neo Smalltalk design, but currently there is a growing group around the project and the focus is now Squeak.
Please note that the Sugar project has my full support, as do the OLPC hardware projects (both the current one and the future XO-2). No solution is perfect and we need to offer the teachers and children alternatives.
You have already mentioned several advantages of a Squeak machine and I agree, which is why I have been working on it. Another advantage is the possibility of costing less - while some people are feeling contrained by the 256MB in the XO-1 it is easy to imagine SqueakNOS running on 64MB or less, for example. A custom Squeak processor should use up less energy to get the same result as a x86 with a software VM.
The most obvious problem of a Squeak-only machine is that you wouldn't have Firefox nor OpenOffice. I am typing this in the Squeak email client, Celeste, and it is pretty crude compared to Kmail and others. And Etoys itself is "the demo that wouldn't die" which we hope can be improved considerably. So we will need to invest heavily to clean things up before this can be an acceptable alternative to Sugar/"normal" Linux/Mac/Windows. It is part of our plan to get funds to allow us to do so.
-- Jecel
On 10.01.2009, at 14:18, Karl Ramberg wrote:
Hi, hearing the news that much of the Sugar dev team is fired I now wonder if there is any point in working on Sugar integration anymore. I have started on a JournalMorph where one could obtain pictures and sounds etc. from the Journal to link in or import into a project, but now I have slight doubts about if there is any use for such a thing ? Do you guys have any thoughts on the future ?
OLPC is *not* abandoning Sugar. On the contrary, Sugar is becoming available not only on the XO but on more Linux distributions, too.
For the time being of course the more than 500 thousand kids (soon to be a million) are more than enough motivation. Etoys is one of the more versatile activities available to them. Great journal access would make it quite a bit more useful.
There was an OLPC developers meeting yesterday, and the plan of action is to not do a full operating system release anymore (which is hardly possible with two engineers left), but instead ship regular Fedora Linux. Sugar is already available in Fedora as a UI option, and it would be made default on a particular Fedora "spin". IIUC countries can choose what system to have pre-installed on their machines and this "Fedora-Sugar" system would be the choice recommended by OLPC.
The Etoys rpms are on their way into Fedora thanks to Gavin so I don't see reason to despair :) If any it's going to be an exciting year for Etoys, there is a great announcement imminent.
- Bert -
2009/1/10 Bert Freudenberg bert@freudenbergs.de:
On 10.01.2009, at 14:18, Karl Ramberg wrote:
Hi, hearing the news that much of the Sugar dev team is fired I now wonder if there is any point in working on Sugar integration anymore. I have started on a JournalMorph where one could obtain pictures and sounds etc. from the Journal to link in or import into a project, but now I have slight doubts about if there is any use for such a thing ? Do you guys have any thoughts on the future ?
OLPC is *not* abandoning Sugar. On the contrary, Sugar is becoming available not only on the XO but on more Linux distributions, too.
I don't buy on that think Sugar on Linux. Not only to develop educative software for linux you have to be very motivated (tiny user base), and I know that for years of personal involvement as a developer to promote linux for education, but then you will still have to be even more motivited to port to Sugar because you will have to adapt your software for an even smaller user base in linux, and for benefice not obvious for me, as an educator. From a developer point of view it looks to me as fragmentation and not exciting at all. I prefer the very comforable Smalltalk based world to write application I know can communicate easily with each other. Don´t forget that if you want developer to buy and invest on it, they need something back in return. Developers do not grow like mushroom. A few years ago, I personally invest resources on Smalltalk because I saw great possibilities of innovation, this is what I got back in return.
May be Etoys was not capable to sell itself as this underneath thing to develop on top of it educative activities. See for example how ungood is the DrGeo shipment for XO, an interactive artifact developped on Smalltalk.
Karl, I think your contribution for kids around the world could be very valuable if you could develop interactive artifacts as the ones I mentioned earlier http://blog.ofset.org/hilaire/index.php?post/2008/05/01/Operational-thinking
Whatever the host, whatever the UI, they can use it, and it is what matter first.
Hilaire
On 10.01.2009, at 16:49, Hilaire Fernandes wrote:
2009/1/10 Bert Freudenberg bert@freudenbergs.de:
On 10.01.2009, at 14:18, Karl Ramberg wrote:
Hi, hearing the news that much of the Sugar dev team is fired I now wonder if there is any point in working on Sugar integration anymore. I have started on a JournalMorph where one could obtain pictures and sounds etc. from the Journal to link in or import into a project, but now I have slight doubts about if there is any use for such a thing ? Do you guys have any thoughts on the future ?
OLPC is *not* abandoning Sugar. On the contrary, Sugar is becoming available not only on the XO but on more Linux distributions, too.
I don't buy on that think Sugar on Linux. Not only to develop educative software for linux you have to be very motivated (tiny user base), and I know that for years of personal involvement as a developer to promote linux for education, but then you will still have to be even more motivited to port to Sugar because you will have to adapt your software for an even smaller user base in linux, and for benefice not obvious for me, as an educator. From a developer point of view it looks to me as fragmentation and not exciting at all. I prefer the very comforable Smalltalk based world to write application I know can communicate easily with each other. Don´t forget that if you want developer to buy and invest on it, they need something back in return. Developers do not grow like mushroom. A few years ago, I personally invest resources on Smalltalk because I saw great possibilities of innovation, this is what I got back in return.
May be Etoys was not capable to sell itself as this underneath thing to develop on top of it educative activities. See for example how ungood is the DrGeo shipment for XO, an interactive artifact developped on Smalltalk.
Karl, I think your contribution for kids around the world could be very valuable if you could develop interactive artifacts as the ones I mentioned earlier http://blog.ofset.org/hilaire/index.php?post/2008/05/01/Operational-thinking
Whatever the host, whatever the UI, they can use it, and it is what matter first.
I think your ideas would be well appreciated on the Sugar education list:
http://lists.sugarlabs.org/listinfo/iaep
- Bert -
Thanks guys for the feedback.
I'll go on with the Journal morph at my slow pace for now. And I think it can be some potable aspects of it as well, I been thinking of making a file reference tile so media items can be loaded when told to.
As for the portability issue, I agree with you Hilaire, platform specific code tend to draw focus from what I think is fun and interesting.
The Sugar activity based approach can feel a little restrictive and would maybe not be my preferred first choice. Then again many feel lost in the multiple possibilities and unrestictive environment of Squeak/Etoys
And Bert got me really curious mentioning a upcoming announcement ...
Karl
On 1/10/09, Hilaire Fernandes hilaire@ofset.org wrote:
2009/1/10 Bert Freudenberg bert@freudenbergs.de:
On 10.01.2009, at 14:18, Karl Ramberg wrote:
Hi, hearing the news that much of the Sugar dev team is fired I now wonder if there is any point in working on Sugar integration anymore. I have started on a JournalMorph where one could obtain pictures and sounds etc. from the Journal to link in or import into a project, but now I have slight doubts about if there is any use for such a thing ? Do you guys have any thoughts on the future ?
OLPC is *not* abandoning Sugar. On the contrary, Sugar is becoming available not only on the XO but on more Linux distributions, too.
I don't buy on that think Sugar on Linux. Not only to develop educative software for linux you have to be very motivated (tiny user base), and I know that for years of personal involvement as a developer to promote linux for education, but then you will still have to be even more motivited to port to Sugar because you will have to adapt your software for an even smaller user base in linux, and for benefice not obvious for me, as an educator. From a developer point of view it looks to me as fragmentation and not exciting at all. I prefer the very comforable Smalltalk based world to write application I know can communicate easily with each other. Don´t forget that if you want developer to buy and invest on it, they need something back in return. Developers do not grow like mushroom. A few years ago, I personally invest resources on Smalltalk because I saw great possibilities of innovation, this is what I got back in return.
May be Etoys was not capable to sell itself as this underneath thing to develop on top of it educative activities. See for example how ungood is the DrGeo shipment for XO, an interactive artifact developped on Smalltalk.
Karl, I think your contribution for kids around the world could be very valuable if you could develop interactive artifacts as the ones I mentioned earlier http://blog.ofset.org/hilaire/index.php?post/2008/05/01/Operational-thinking
Whatever the host, whatever the UI, they can use it, and it is what matter first.
Hilaire _______________________________________________ Etoys mailing list Etoys@lists.laptop.org http://lists.laptop.org/listinfo/etoys
Greg posted a bit more detail about what's happening with OLPC and Fedora:
http://gregdek.livejournal.com/43698.html
He links to the todo-list of packages:
https://fedoraproject.org/wiki/OLPC/Packages_for_F11
... which includes a short discussion on the "etoys" and "squeak-vm" packages, hinting that there might be a packaging problem.
- Bert -
Hilaire Fernandes wrote:
2009/1/10 Bert Freudenberg bert@freudenbergs.de:
On 10.01.2009, at 14:18, Karl Ramberg wrote:
Hi, hearing the news that much of the Sugar dev team is fired I now wonder if there is any point in working on Sugar integration anymore. I have started on a JournalMorph where one could obtain pictures and sounds etc. from the Journal to link in or import into a project, but now I have slight doubts about if there is any use for such a thing ? Do you guys have any thoughts on the future ?
OLPC is *not* abandoning Sugar. On the contrary, Sugar is becoming available not only on the XO but on more Linux distributions, too.
I don't buy on that think Sugar on Linux. Not only to develop educative software for linux you have to be very motivated (tiny user base), and I know that for years of personal involvement as a developer to promote linux for education, but then you will still have to be even more motivited to port to Sugar because you will have to adapt your software for an even smaller user base in linux, and for benefice not obvious for me, as an educator. From a developer point of view it looks to me as fragmentation and not exciting at all. I prefer the very comforable Smalltalk based world to write application I know can communicate easily with each other. Don´t forget that if you want developer to buy and invest on it, they need something back in return. Developers do not grow like mushroom. A few years ago, I personally invest resources on Smalltalk because I saw great possibilities of innovation, this is what I got back in return.
May be Etoys was not capable to sell itself as this underneath thing to develop on top of it educative activities. See for example how ungood is the DrGeo shipment for XO, an interactive artifact developped on Smalltalk.
Karl, I think your contribution for kids around the world could be very valuable if you could develop interactive artifacts as the ones I mentioned earlier http://blog.ofset.org/hilaire/index.php?post/2008/05/01/Operational-thinking
Whatever the host, whatever the UI, they can use it, and it is what matter first.
One of my main issues with Logo was that the artifacts created could not be used further on, they where drawings, not objects. I would like to design a cool spaceship in Logo and then use that in a space game, so that the Logo drawing would be a beginning of a bigger project, not the end of one. Squeak can do this. This is what make it exiting for me. Squeak have quite a few rough edges but most of them are approachable and solvable.
As for learning environments I started out on C64 computers and found them totally captivating and spent countless hours on them. But I realize I'm not a average student and a C64 would be a brick for most people. They where quite perfect for me and the possibilities seemed countless and when people got all exited over IBM pc's I could not understand them, PC seemed dull and unimaginative.
Karl
Karl
I watched a
etoys-dev@lists.squeakfoundation.org
-
Bert Freudenberg -
Hilaire Fernandes -
Jecel Assumpcao Jr -
karl ramberg -
Karl Ramberg
-
Walter Bender