[Cuis-dev] Dynabook and education

Juan Vuletich juan at jvuletich.org
Sun Jun 14 13:30:05 PDT 2020


On 6/12/2020 6:21 PM, Hilaire Fernandes via Cuis-dev wrote:
> Le 12/06/2020 à 16:06, Juan Vuletich a écrit :
>
>> I can not help wondering what is still missing for that very article, 
>> together with the examples, to be an active essay itself, running in 
>> the Smalltalk system. Erudite, Glamorous Toolkit, StyleTextEditor, 
>> Morphic Projects are all pieces in that space. What else is needed? 
>> How can we move forward?
>
> IMO, getting right the backbone where dynamic media will fit in place 
> is the tuff part. Since 5000 years, our culture is text based, the 
> *hand* and the *eyes* are its tools. Looking at what is happening in 
> school (primary up to senior high school) is interesting because it is 
> where our culture is transmitted to our young.
>
> The way we teach at school is articulated around texts: teachers 
> provide a lot of text documents (book, exercises book), students 
> produce a lot of text too (mainly exercises resolution, drawing - on 
> the desk too -, notes). As a teacher, I only know and speak about this 
> context. But clearly we are not where we could be, I fell daily this 
> frustration. Why be limited to physical text contents, why not digital 
> dynamic media as text, sound, video, lesson, exercises circulating 
> between the teachers and students, annotated, re-annotated, cut in 
> piece for exploration. Sometime I fell like we are stuck in the XIXe 
> century although I use daily computer in front of the students. So yes 
> now we can use computer, but it's like dragging around a cinderblock, 
> so many obstacles for so little outcomes.
>
>> The Dynabook and Active Essays ideas are extremely important for me. 
>> Cuis is becoming mature enough to be a good base for them. (Although 
>> I think that vector graphics is a must).
>
> Cuis is mature and even more importantly it is *agile like a 
> ballerina*. I use your recommendation: start the morning with a fresh 
> image, load my code, never save the image, only the packages. It is 
> very easy to do with a minimal, easy to remember syntax, very fast and 
> Cuis never crash (ok, just once)
>
> Agree about vector graphics, otherwise how will you do state of the 
> art stylus handwriting/sketching on the dynabook?
>
>>
>> I think that, as you suggest, writing active essays will require, in 
>> addition to some natural language, and perhaps topic specific jargon, 
>> good Smalltalk skills. But I believe that a casual reader of an 
>> active essay should be able to get most of the ideas without needing 
>> to understand the code, in the same way we can get a good part of 
>> what a technical book is talking about, without looking at the formulas.
>
> I wrote another small article discussing about these questions, based 
> on A. Kay writing.
>
> https://blog.drgeo.eu/2018/07/the-dynabook-concept.html
>
>

Very nice writeup, thanks. I fully agree with what you say.

I don't think that Etoys (or Scratch) are the answer. Today, it is 
possible to build interactive multi-media / hyper-media learning 
contents. It can even be done very convenient to access, in the form of 
html + js + canvas. The problem is that it is extremely hard to build 
such content, and it takes way too much work. And the result is not user 
editable.

What we need is a tool that allows both the consumption and the creation 
of such contents. That's why I don't think that Etoys or Scratch go in 
this directions. Something closer to Erudite, perhaps a bit of 
StyledTextEditor,of course DrGeo, dynamically created vector graphics 
morphs (code part of the authored content), dynamic simulations (model 
code, animation code, datasets all part of the authored content), etc, 
all integrated with good looking text. Untill you interact with the 
stuff, it should look as good as a .pdf of a professionally edited and 
typeset book. Then, it would react and be alive. And the, you'd be able 
to edit it, comment it, re-publish, etc.

An integration of Erudite and DrGeo would be a great next step from 
where we are now.

If you haven't yet, read SqueakNews. It is not only really enjoyable, 
but also the best example of Dynabook content I've seen. And it was done 
20 years ago! All the published issues are available at 
https://github.com/tansel/SqueakNews . As it is done in an old Squeak, 
using an older image format, you can't run them with a recent VM. If you 
have a Windows system, they run perfectly fine. If you have a Mac or 
Linux PC, it is really worth to install a Windows VM, using VMWare or 
similar, just to read this amazing stuff. It is really worth it.

>>
>> People are trying to do OpenScience / ReproducibleResearch with 
>> Python Jupyter Notebooks. We can do much better than that!
>
> I don't know much about this. Experimental science teacher in school 
> could benefit of a content with media of heterogeneous forms: text 
> with some handsketched writing, photography/video, table of data 
> collected from the device input sensors, etc.
>
> The mental image I like is to imagine a *student A4 textbook*, with 
> contents on the forms of handwritten text, an interactive diagram he 
> can play with, a vocal note of a Spanish language exercises, and some 
> data he collected from the temperature sensors plugged to the 
> textbook. For a meaningfull use of IT in education, it would be a good 
> start.
>
> Of course student will receive his dynabook at the primary school and 
> keep it at least until senior high.
>
> Are we dreaming? ;-)
>
> Hilaire
>
> -- 
> GNU Dr. Geo
> http://drgeo.eu

Most school kids already have the required hardware. It is us who need 
to build the missing software parts!

Cheers,

-- 
Juan Vuletich
www.cuis-smalltalk.org
https://github.com/Cuis-Smalltalk/Cuis-Smalltalk-Dev
https://github.com/jvuletich
https://www.linkedin.com/in/juan-vuletich-75611b3
@JuanVuletich

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cuis.st/mailman/archives/cuis-dev/attachments/20200614/d0f512f9/attachment-0001.htm>


More information about the Cuis-dev mailing list