Re: ANN: SmartEiffel 1.2 (transitional) project

Lothar Scholz wrote:
> way to implement things - this means: to do what ISE is doing even if
> we don't like a feature/syntax.
It is out of question. As we all know, the ECMA proposal is not a
normalization document. For me one can normalize something
which is implemented and tested. Many points of ECMA are not.
So, when I say that ECMA is not a mormalization work, this is a fact
and not a personal opinion.
The paradox is that many proposals of ECMA are pure research work!
Take for example this idea of attached type: not implemented, not
tested. A pure research work.
Conversely, we, the SmartEiffel project, will promote stability
and a kind of language conservatism.
For us, 99% of the Eiffel syntax and mechanisms we want is already
there in the current working SmartEiffel. Our guess is to make
progresses only with better tools. As an example, we do want to
eradicate Void calls by using dataflow analysis and not with a
new type-sytem redefinition.
We do not want to modify the language again and again, we just want
to improve our compiler and tools.

So the amazing conclusion is that if you want stable code, just follow
the compiler written by the pure research team, and, if you want to take
the risk to go for a new non-tested research-like language, just go for 
the ECMA standard.

> So we need a branch where a remerge with some later SE 2.x (or 3.x)
> version will have a very very low priority just because the 2.x
> implementation is terrible bad (compiler speed) and the language
> design decisions are very questionable.
For compiler speed, I do not agree. If my memory is good, Philippe
Ribet has posted here a message about the complexity of the new
algorithms of 2.x. And if you were using our working version,
you would know about the new performances we get. But I presume
you are probably still using SmallEiffel ;-)
For language decision, our point is really amazing. Just compare the
number of modifications we made and the number of proposed modifications 
in the ECMA standard. Who is forking?
Anyway, Lothar, we all know the kind of guy you are ;-)
As we say in French, Cordialement,
-- 
--------------------------------------------------------------
[email protected] -- IUT (Nancy 2) -- INRIA Lorraine
http://SmartEiffel.loria.fr  --  The GNU Eiffel Compiler
POST: Loria, B.P. 239,54506 Vandoeuvre les Nancy Cedex, FRANCE
Voice:+33 0383913140 Mobile: +33 0665362381 Fax:+33 0383581701