[Cuis-dev] Dynabook and education

[Thierry Goubier]

Le lun. 15 juin 2020 à 10:09, Hilaire Fernandes via Cuis-dev
<cuis-dev at lists.cuis.st> a écrit :
>
> Le 13/06/2020 à 17:10, Thierry Goubier via Cuis-dev a écrit :
> > This reminds me of the Hypermedia system I did in Parcplace Smalltalk
> > in 1993, to study navigation in hypermedias (one publication...) for
> > my MAS degree. Reusing HotDraw and MVC widgets embedding inside the
> > text editor (MVC widgets inside HotDraw, too, even if I think I
> > cheated on that one).
>
> Any paper to share? I am collecting materials for ideas.

I look if I can get a version of the paper (presented at ErgoIA'94).
It's in french, so that should be ok to you :). Does not describe how
it was implemented, if I remember correctly.

> > Morphic should allow that to be a lot easier, but the text support is
> > no better than it was in the 1990's (i.e. we don't have a text editor
> > component that is Morphic all the way down).
> >
> > Nowadays, I'd be interested by someone who has a concept for
> > hypermedia live programming (i.e. allowing a computation to start on a
> > page, and use that computation context for the following alternatives
> > pages in a wiki like setup) even if I think nowadays that teaching
> > programming is shortsighted...
>
> I think programming is something we are all interested, but we want it
> to be mixed in media as you described it

My view on programming is that we should be considering that:

- writing ourselves core algorithms means getting them wrong most of the time.
- writing ourselves core algorithms makes the optimising task of a
compiler exponentially hard
- writing ourselves core algorithms makes the parallelising task of a
compiler exponentially exponentially hard
- most of the code we write is a copy/paste of another code / with
often not even name changes
- good algorithms for our problems are non obvious and require a lot of research

Now, what should we teach?

> Teaching programming (in secondary) is a never ending story: in 70-80's
> it was the Holy Grail, then in the 90's it was considered irrelevant in
> favor of ICT, then toward the end of the 2000's it becomes again the
> Holy Grail thanks to the Scratch advent. So we have 15-20 years phase
> each time.

I believe Scratch made a very significant contribution (I still
remember what was "visual programming" before Scratch :( ). Smalltalk
is still a very significant contribution too.

> Likely the right position could be in between.

I agree...

Regards,

Thierry

> --
> GNU Dr. Geo
> http://drgeo.eu
>
> --
> Cuis-dev mailing list
> Cuis-dev at lists.cuis.st
> https://lists.cuis.st/mailman/listinfo/cuis-dev