Harald Gliebe wrote:

Hi Selfers:

This sounds as great news!!!From the e-mails I have been reading
it seems that the version for Mac does not work well! How can I
be sure that the  Selfversion for Linux work as well as the one for Sun?

My Scientific Report is ready. It is about 3,5 Mbytes.
I show it is necessary a multi-paradigm design to make programming languages
really expressive.

As multi-paradigm design I imply the inclusion of reflective architecture as
well as aspect oriented programming in Self and separation of concerns.
Of course I believe that delegation and Kansas are great achievements that
should be maintained!!!

Greg Kiczales, the founder of aspect-oriented programming and a pioner of
reflective architecture finds my undertaking interesting.

However of course really revolutionary innovations nobody has time to
pay time!~It is easier to deal with Aspect J Aspect C and so on....

Anyhow I have been delving deeper into semiotics. This time I am analysing
the work by the semiotician Algirdas Julien Greimas. What's new about him?

However to situate semiology (now semiotics) within linguistics, as BArthes
    and Greimas do, requires the use of the terminology of linguistics to explore
    the larger semiological phenomena: it requires what Greimas calls the
    transposition of the methodological procedures of the plane of the signifier
    to that of the signified.
For example: to the smallest unit of signification called a pheme there is
a corresponding one in meaning called a seme!! A phoneme is a sum of
phemes = a syllable and a collection of semes is a lexeme = a word or a sign!

    This is precisely Greimas's semiotic project, the theoretical mediation
    between narrative forms and linguistic forms of sentential dimensions what
    he calls an attempt to shed a little light upon the relations which can exist
    between discourse and the sentence, between discursive linguistics and
    sentential linguistics.

    For this reason Greimas defines semiotics in terms of actants, actantial
    roles and the structure of the narrative functions of discourse in the terms
    of linguistic analysis.

Well I believe it is well known separation of concerns  approach looks for
entities smaller than the class or the object to model reality such as layers
and views (although layers and views can also be entities bigger than classes
and objects!!!)

Yet component based languages are also of research interest exactly to
model entities greater than the class and the objects. Currently design
patterns are the best entitites to do so.

What is interesting in Greimas's semiotics is that he deals with the whole
range of granularity of entities that make up sentences and discourse.
He says:

actants + predicats = messages

It seems to me that multi-agent systems were intended to be actantial
models (based on the subject) and functional models and qualificative
(predicats)  models fall into the category of OO programming mainly
concerned with action and behavior.

I have not concluded anything yet but those interested can e-mail me
to start reading my Scientific Report and help me to think.

I am trying to bombard the wrong reasoning in both multi-agent
reasoning and in class-based reasoning. Of course because computer
scientists do not know semiotics. It is impossible to model without
However Self is the sort of language
that grasps the semiotic reasoning
hence a candidate to be part of a truly
scientific project curiously based on the
hermeneutic reasoning. Or a reasoning
that grasps how the world really is when
Self extends to include entities smaller than
the object and bigger ones!!!

A Happy 2002 for all Selfers!


I just uploaded version 0.2.1 to http://www.gliebe.de/self .
This release fixes some bugs and allows to run Self on a display with 16 bit
color depth.

Best regards,

Harald Gliebe

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/