armin van harten

Dear fellow Msters,

since it seems that all M-activities have come to an end and the usual job
for a M-professional is to port old, muddled M-software to an oracle based
RDB without too much loss in performance and features, I would like to ask
the community, if there is still blood enough in this body to start a M-
based project with some exciting aspects in theory and practical use.

In short terms:
M encounters the same final problems like assembler language did 30 years
ago.
The solution was not to use assembler directly as a programming language
any more but use it as an implementation language.

M is a tree oriented, associative, typeless language. Strong features but
too strong for most of the application programmers..... Anyway! These
basics lead to abstractions and solutions very different from the 2
dimensional flat models, which reached the end of there possibilities and
even companies like oracle and MS are looking for a way out of it. Many
applications perform better with a document object model (DOM) than with an
entity relationship model (ERM).

Many M pro's use DOM's, but the theory and basics still not well founded,
if not still unrevealed and so those models usually look strange, heuristic
and isolated.

Just like doing OOP directly with assembler, doing DOM programming with
plain M would be just academic but nothing for practical use.

What we need is:

A language concept to preserve the few valuable characteristics of M but
add the loads of modern concepts.

A theory of tree based data objects, like we can find the theory of
relational structures in linear algebra. This would be necessary to find
the right abstractions for basic systems like a treestructured - database
or something like that.

Define a core system
Define communication between systems
Define how to move objects through different systemlocations

implement the particular parts and test them and all that jazz.

After this, there will be no M programming in application layer anymore,
but we will have the core idea moved into the presence.

There will be a lot of thinking and writing in the project. All M-sters
know that programmin g M will be the minor part.

Needed is knowledge in compilerwriting, systemdesign, mathematics,
networking, treestructures and at least in M

This is the first call, just to look if there is still somebody out there
and if there are folks who like to do something completely weired because
everybody will call them crazy at least for the first steps of the project.
You will hear that you are inventing the wheel again and until the system
will present its first applications most of the audition will not see what
it is about.

Are there enough crazy people out there?

What can I do? IfmD-consulting has to work for its money, so we can not
supply manpower unlimited, but we can provide basic materials, webspace,
our own theories and developments about persisting objects of the last 10
years and maybe little bit more.

I, personally, would love to find a core group here in Germany, so we would
have the chance to meet personally, but if the development shows a
different focus, no problem. Explicitly: everybody is invited to attend,
from every nation, color or religion (We invite those Javabeans too,
sometimes they definitely have usefull ideas ;-))

OK, thats it for now. I am very curios to see if there is a lifesign and
what it will be.

Regards
Armin van Harten

Ifmd - consulting GmbH
Sophienstrasse 36a
35576 Wetzlar
+49 6441 445408
@@vanharten(at)ifmd-@@consulting.@@de