🚀 go-pugleaf

Hi!

On Di 19.03.1996 around 08:13:32 ludwig@inf.ethz.ch (Stefan Ludwig) wrote
in message "Language "purity" and the ETH (was: bitwise and & or in Oberon?)":

> >   There  are  no technical problems in implementing an Oberon compiler for
> >another  system,  eg.   the Amiga.  But you have to change the language for
> >this, sometimes even outside Oakwood-style EXTERNAL interfaces, which makes
> [...]
> You can always define interface modules in pure Oberon,
> which encapsulate the underlying system interface by accessing it via SYSTEM.
> It's possible, safe and works very well.

   It's  just  not  very  efficient to pass my arguments on the stack first
just  to  have  the  interface routine stash them into registers and make a
SYSTEM call.

   Oakwood  should have defined that "EXTERNAL" means "_really_ far removed
from  Oberon".  The situation now is that for efficient interfacing we have
to extend the language but can't properly "mark" the module as unportable.

> >   IMHO the ETH position still contributes to the lack of such a wave.  But
> >   I guess we could argue this point endlessly.
>
> What position is that? What has been true three years ago must not be true
> today.

   I  think the position I spoke of is pretty clear to most of us:  The ETH
gives us a language and that's it.  No further discussion needed.

   I was going to argue about the "what was true" thing when I noticed that
you  _might_  indicate  that the ETH position is changing.  Is that so?  Is
there  hope for a ETH workgroup getting together with "us out here" to work
on Oberon-3?

> Who wants what? I don't see much initiative in the industry working towards
> a standard (with or without ETH's consent). If you out there want to change
> something, get together, organize yourself, propose something, talk to the
> creators.

   I  work  on  a proposal/evaluation of language features (and you know it
:-)  and  would  happily  contribute the results to a "Standard for Oberon"
project.  But I feel that someone else would have to take the lead, since I
fear noone takes "a mere student" seriously.

> It shouldn't be too difficult to standardize the language once
> and for all, incorporating the experience gained over the past 9 (nine!) years.

   Who's  talking  about "once and for all"?  A standard is needed, call it
Oberon-96 or whatever.  This would by the reference for some years.  But at
least I want to see it updated every 5 years or so.

> We're *using* the language and the
> system to develop applications.

   I still hope that _some_ ETH involvement in such a process would happen.
It somehow makes me sad to "take away" their language.
                  _
By(T)e...   /| _ //3000 /\ People who use Pascal for serious programming |\
   Peter... \| \X/\miga \/ fall into a fatal trap.  (Brian W. Kernighan) |/