[self-interest] macro systems

Thu Aug 22 17:24:09 UTC 2002

James McCartney wrote:

Dear James:
Special thanks!
At Enschede, during AOSD conference in April Robert Filman
suggested me to transform my ecodesign model into a domain
specific language. There Jeff Gray one of the organizers of the
Workshop on Early Aspects did his PHD thesis  in a domain
specific language. Later on I read Kiczales said that AOP
began as a domain specific language.
Then I ask Jeff Gray:

The earliest work on AOP languages, as far as I can tell, was done by
Cristina Lopes at Northeastern. Karl Lieberherr was her advisor, I think.
She created two languages, called COOL and RIDL, that were focused on
synchornization and distribution. I a not sure why those two concerns
were chosen, but perhaps those were the most evident concerns that
demonstrated crosscutting. From that, the AspectJ work has taken a more
general approach that also permits such concerns to be captured.
This was his answer.Furthermore:
 I am not sure  have I sent you my thesis yet? There is a section in there that
talks about
domain-specific languages, in particular, those that are visual.

BTW, there have been several workshops at OOPSLA on the idea of visual
domain-specific languages. In
case you would like a peek, they are at:

http://www.isis.vanderbilt.edu/oopsla2k1/

http://www.cis.uab.edu/info/OOPSLA-DSVL2/

When I say domain-specific language, I am really referring to a visual language that
can be used to create
models for a particular domain.

For example, think about an environment that would allow you to create models of
automobile plant
factories using a visual formalism.

Or, a chemical factory, or, any other kind of factory. The models for each factory
are peculiar to each
domain that they represent. That is, the models have a specific ontology depending
on the domain (they
have different entities, different associations among those entities, and even
different visual iconic
representations).

This is a little different than the traditional definition of a domain-specific
language. In fact, I think you
were totally correct to say that such languages when they become transdisciplinary
are just programming
languages.

Well this makes me wonder! My ecodesign model and its underlying geometric model
inspired by
the symmetry groups of the plane and the dotless plane from Escher enables me to
generate
all sorts of architectonic objects, from home to a cathedral!! and from the union of
architectonic
objects with the same reasoning you generate cities and hence can plan and design
sustainable cities.

The weaver of crosscutting concerns here is this geometric modeling!

Jim Coplien is trying to formalize design patterns as a pattern language through
symmetry groups.

Then my question is:

see below

> The best reference would probably be the book _On_Lisp_.
> (Which I just discovered you can download for free from
> http://www.paulgraham.com/onlisp.html  )
> A macro is basically a way to run language code at compile time to
> generate language code.
> With this you can do any kind of stitching together of concerns that
> you would want.
> If you wrote a set of macros to create your classes you could have them
> generate code for your methods,
> including any 'aspect' like concerns.
>

In my case I want first to create prototypes, responsible for crosscutting
concerns inspired on musical chords.
I suspect I have to create this on top of Self. This means using such
macros I can extend the Self compiler  and have my job done?
In this case it is already born transdisciplinary, because all one would have
to do is to substitute the notion of tones by the elements of the application
in focus. Then people do not know how to model mimicking my
ecodesign model. All I should do is to transform it into a programming
language implemented on top of Self for example?

Things sound simpler than I thought! I will take a look at the site
and at Jeff's PHD thesis. If you have some more suggestion, it is
welcome.
Best wishes
Albertina

>
> On Thursday, August 22, 2002, at 07:40 AM, Albertina Lourenci wrote:
>
> >
> >
> > James McCartney wrote:
> >
> > Dear James:
> > I studied CLOS and Common Lisp ten years ago. I forgot what
> > macro systems means. Would you mind gently explaining it to me?
> > Best wishes
> > Albertina
> >
> >> I have the same question. All of the examples given seemed to be
> >> something that having LISP like macros would solve.
> >> A call for better macro systems in languages would seem to me the
> >> appropriate response to the problem. But perhaps I missed something.
> >>
> >> On Wednesday, August 21, 2002, at 01:14 AM, Thorsten Dittmar wrote:
> >>
> >>> I'm asking because I followed the discussion about AOP over years and
> >>> I had
> >>> always the feeling that it solves problems that a good designed
> >>> language
> >>> does not have.
> >>
> >>
> >>
> >> Your use of Yahoo! Groups is subject to
> >> http://docs.yahoo.com/info/terms/
> >>
> >> ==========================================
> >> This email message has been verified against viruses and SPAM using
> >> MIMESweeper.
> >> Este email foi verificado quanto a SPAM e virus utilizando
> >> MIMESweeper.
> >> http://www.mimesweeper.com
> >> ==========================================
> >
> >
> >
> > Your use of Yahoo! Groups is subject to
> > http://docs.yahoo.com/info/terms/
> >
> > <lourenci.vcf>
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
> ==========================================
> This email message has been verified against viruses and SPAM using MIMESweeper.
> Este email foi verificado quanto a SPAM e virus utilizando MIMESweeper.
> http://www.mimesweeper.com
> ==========================================
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.selflanguage.org/pipermail/self-interest/attachments/20020822/592e4570/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: lourenci.vcf
Type: text/x-vcard
Size: 620 bytes
Desc: Card for Albertina Lourenci
URL: <http://lists.selflanguage.org/pipermail/self-interest/attachments/20020822/592e4570/attachment.vcf>