[Mozart Oz Users] Mozart documentation -- proposal to simplifyit

[Seif Haridi]

I think your proposal is quite nice.
I also think that the book we are writing is not a replacement for
tutorials.
While the book is using Oz and Mozart I do not see the book as a Mozart
book.
My goal is the book succeeds if it is not viewed as a specific language
book.
Tutorials should concentrate on the language and the system.

The problem as Peter discussed is that I am not able to improve the tutorial
right
now since I am quite busy with other things including the book.

On the other hand if any one want to improve the tutorials or write other
materials
as suggested here I am willing to assist.

--- Seif

Torbjörn Lager wrote:

> I personally don't think the Tutorial is a very good entry point to Oz.
> It is too long already, and contains a lot of material that is difficult
> for many readers/students, and perhaps also some bad advice (like the
> use of the dis-construct which could scare away a bench-marking Prolog
> programmer very quickly -- almost did for me until I realized 'choice'
> was a much better choice...). Also, since it is far from finished, it
> can only become longer, and then it will soon start to 'compete' with
> the Models book.
>
> I would like to suggest another approach. Oz is a multi-paradigm
> programming language, so there should really be multiple entry points
> into Oz. (Really, Oz is unique in that there *can* be multiple entry
> points!) These could be small documents, around 20-30 pages each:
>
> - Oz for Java programmers
> - Oz for Prolog programmers
> - Oz for Python programmers
> - Oz for Smalltalk programmers
> - Oz for Ruby programmers
> - Oz for Lisp programmers
> - 'Oz for Dummies'
> - etc.
>
> Each 'Oz for XXX programmers' document could contain a gentle
> introduction to the Oz language by way of comparing with the XXX
> programming language. Besides, many of them documents would be able to
> share some text and example code, so the effort of writing them would
> perhaps not be that great.
>
> The user community could have a role to play here, as I'm sure there are
> competent XXX programmers in the crowd, that would be capable (with a
> little help from the Oz developers) of comparing XXX and Oz on a level
> which is interesting and useful to the typical XXX programmer.
>
> Each 'Oz for XXX programmers' document could also contain a Beyond-XXX
> section ('Beyond Prolog', 'Beyond Python', etc.) that shows the power of
> the *other* paradigms (which are different depending on where you are
> coming from...) and the power of combining paradigms. Each document
> could provide a pointer to the Models book, for those who want to learn
> more.
>
> This would be a way to steal away lots of people from other languages
> :-) Just send the documents to the relevant comp.lang.XXX groups
> (comp.lang.prolog, comp.lang.java, etc.) and wait. In fact, they could
> be posted there on an regular basis, since they are (in a way) relevant
> to the user groups in question.
>
> Regarding the Models book. I think it's a truly wonderful book and I
> wish Peter and Seif would never stop adding interesting ideas,
> comparisons and examples to it. I will certainly buy a hard copy when it
> gets out. In fact, if they add another 500 pages to it, I will buy two
> copies, because then it will become to heavy to carry around :-)
>
> And you guys writing about NLP in Oz, please don't stop. I will buy your
> book too, I promise :-)
>
> Cheers,
> Torbjörn
> --
> ----------------------------------------------------------------
> Torbjörn Lager                  Computational Linguist
> Department of Linguistics       Uppsala University
> S-751 20 Uppsala                Sweden
> Phone: +46 18 471 7860          http://stp.ling.uu.se/~torbjorn/
> ----------------------------------------------------------------
> -
> Please send submissions to users at mozart-oz.org
> and administriva mail to users-request at mozart-oz.org.
> The Mozart Oz web site is at http://www.mozart-oz.org/.

--
Seif Haridi
Professor of Computer Systems,
Department of Teleinformatics, Royal Institute of Technology.

Chief Scientific Advisor, Swedish Institute of Computer Science.
Web: http://www.sics.se/~seif

Phone (office): +46 8 7521351
Mobile: +46 70 5121540
Mail: SICS, Box 1263, SE-164 29 Kista, Sweden


-
Please send submissions to users at mozart-oz.org
and administriva mail to users-request at mozart-oz.org.
The Mozart Oz web site is at http://www.mozart-oz.org/.