🚀 go-pugleaf

Peter Froehlich (p.froehlich@amc.cube.net) wrote:
:    Note  that  I  don't  propose to really change the language at all, just
: it's appearance.

Don't you think that the readability of the Oberon syntax is much better than
the readability of C (or C-like Java)? I think that a change of syntax would 
not at all result in an improvement.

:    As  the subject says, this is a weird idea (and I know that, don't point
: it  out  to  me)  just  good for producing the "marketing hype" that Oberon
: currently lacks.

I think, a big problem of today's software is that many ugly things are
introduced for the sake of "marketing hypes". Let's not do the same thing
with Oberon.

mfg, Roland

In article <4j0vqu$2m9@neptune.ethz.ch>,
rvoegeli@inf.ethz.ch (Roland Andre Voegeli) writes:
> Peter Froehlich (p.froehlich@amc.cube.net) wrote:
> 
> :    As  the subject says, this is a weird idea (and I know that, don't point
> : it  out  to  me)  just  good for producing the "marketing hype" that Oberon
> : currently lacks.
> 
> I think, a big problem of today's software is that many ugly things are
> introduced for the sake of "marketing hypes". Let's not do the same thing
> with Oberon.

 Do not you think that to some extent System-3 + Gadgets tried 
to follow the "marketing hype" of overlapped windows, as the original
System interface was considered too awkward? What happened then?
Did it get any appreciable market share among "end users" who were 
supposedly the target? 

IMHO Oberon needs marketing, not marketing hype. These can be two 
different things.

Just my 0.02,

Wojtek

skulski@fermi.nsrl.rochester.edu wrote:
: In article <4j0vqu$2m9@neptune.ethz.ch>,
: rvoegeli@inf.ethz.ch (Roland Andre Voegeli) writes:
: > Peter Froehlich (p.froehlich@amc.cube.net) wrote:
: > 
: > :    As  the subject says, this is a weird idea (and I know that, don't point
: > : it  out  to  me)  just  good for producing the "marketing hype" that Oberon
: > : currently lacks.
: > 
: > I think, a big problem of today's software is that many ugly things are
: > introduced for the sake of "marketing hypes". Let's not do the same thing
: > with Oberon.

:  Do not you think that to some extent System-3 + Gadgets tried 
: to follow the "marketing hype" of overlapped windows, as the original
: System interface was considered too awkward?

With Oberon System 3, J.Gutknecht tried a different approach to system-objects
than it was done in Oberon V4. V4 supports the "element paradigm". Unfortunately,
Elements are very strongly bound to texts. System 3 objects seem to me to be the 
more "regular" concept.
I don't know, what was the "real" driving force for the development of Gadgets.
Perhaps it may have been the "marketing hype", maybe it is just an other approach
to user interfaceing (more flexible user interface and so on...).
I'm tending to the second argument.
As you shurely know, user interfaceing is a complex thing without THE solution.

: Did it get any appreciable market share among "end users" who were 
: supposedly the target? 

No, but it gave some ideas to people such as the Oberon Microsystems staff. Just
look at Oberon/F.

: IMHO Oberon needs marketing, not marketing hype. These can be two 
: different things.

I agree with you.

mfg, Roland

In <224676F7H000D1F62H@p.froehlich.amc.cube.net> p.froehlich@amc.cube.net (Peter Froehlich) writes:


>   Note  that  I  don't  propose to really change the language at all, just
>it's appearance.

NO!  This is exactly one reason why I prefer Oberon to C/C++.  I can
read Oberon.  I can't read C/C++ (although I have worked with the
languages for more than 3 years).  I much prefer "PROCEDURE f" to 
"void f(void)".  This attachment to form is deeper than just appearance
since the pattern recognition process is directly mapped into the
thought processes of program design.  I'd say its like asking a Frenchman
to speak English (were not really changing the language only its appearance)!

-Doug 

-- 
-----------------------------------------------------------------------------
  Douglas G. Danforth                              danforth@csli.stanford.edu
  Visiting Scholar                                 (415)723-2487 x332 vox  
  Applied Speech Technology Lab                    (415)725-2166 fax
  Center for the Study of Language and Information
  Ventura Hall                    
  Stanford University, Stanford CA.  94305-4115             
-----------------------------------------------------------------------------

In article <224676F7H000D1F62H@p.froehlich.amc.cube.net>  
p.froehlich@amc.cube.net (Peter Froehlich) writes:
> Hi there!
> 
>    Seeing  all  the hype about JAVA around the net and in the printed media
> got me thinking, as usual on a programming language level.
> 
> ----- WEIRD IDEA -----
> 
> Two (seemingly unrelated) statements (both true I think):
> 
>    - JAVA is C/C++ like, and that sure helps a lot in advocating it's use.
>    - Modern Oberon(-2) compilers are based on abstract syntax.
> 
> One conclusion (and a request for comment):
> 
>    How  about  defining  a  concrete syntax for Oberon(-2) that attempts to
> "imitate"  the  _look_  of  the  C  language?   Integration with Oberon(-2)
> compilers would just require a new scanner and possibly a modified parser.
> 
> ----- WEIRD IDEA -----
> 
>    Note  that  I  don't  propose to really change the language at all, just
> it's appearance.
> 
>    As  the subject says, this is a weird idea (and I know that, don't point
> it  out  to  me)  just  good for producing the "marketing hype" that Oberon
> currently lacks.  :-)
>                   _

Consider yourself toast. :-)

Better: have an Oberon compiler target switch to make the compiler 
generate JAVA virtual machine code.

Even Better: Hook an Oberon compiler directly to the Web browser and
forget about JAVA completely. Actually I think that it was JIVA and a typo
stuck. :-)O<

richard

Richard Pitre wrote:

> Even Better: Hook an Oberon compiler directly to the Web browser and
> forget about JAVA completely. Actually I think that it was JIVA and 
> a typo stuck. :-)O<

Even more better: Extend OMI by Oberon-2 support and add the network 
functions of Java.

OMI has the most important feature of Java, multi-platform support, 
too, but doesn't share the disadvantage of low performance. OMI 
modules get _compiled_ during loading, in contrast to Java modules, 
which get _interpreted_ by the Java virtual machine during run time.
 
---peter.landmann@lrz.fh-muenchen.de----------------------------------
   'C is a whith syntax sugared assembler' (N. Wirth)

In article <4j41vs$31t@neptune.ethz.ch>, 
rvoegeli@inf.ethz.ch (Roland Andre Voegeli) writes:
> skulski@fermi.nsrl.rochester.edu wrote:
> 
> :  Do not you think that to some extent System-3 + Gadgets tried 
> : to follow the "marketing hype" of overlapped windows, as the original
> : System interface was considered too awkward?
> 
> With Oberon System 3, J.Gutknecht tried a different approach 
> to system-objects than it was done in Oberon V4. 

Thank you for the comment. I agree with you that System-3 innovation
is in its object management module. Such a module was missing in V4,
and consequently every application had to define its own local 
object management. Say, Dialogs have persistent objects of its 
own, so does Kepler. I am acutely aware of this fact since I also 
had to develop my own persistent object library, which I would 
not have done if there was such a V4 library agreed upon.

> V4 supports the "element paradigm". Unfortunately,
> Elements are very strongly bound to texts. 

Do not you think that V4 Texts plus Elements can be viewed simply 
as a lightweight user interface? This interface proved itself to be 
very powerful and flexible, but one should not attach too much 
weight to it. If one does not like Elements, one is always free 
to define some other user interface. Dialogs do exactly this.
Therefore, I do not quite agree with your "unfortunately". 
Elements are purely optional.

> System 3 objects seem to me to be the more "regular" concept.

The whole issue is then that V4 does not include Objects.Obj. 
Well, then why do not you or somebody else just release Objects.Mod, 
and V4 will become just as regular as System-3 ?

> : Did it get any appreciable market share among "end users" who were 
> : supposedly the target? 
> 
> No, but it gave some ideas to people such as the Oberon Microsystems 
> staff. Just look at Oberon/F.

This is of course good. But I still think System-3 failed to become 
popular even among us here on the net, what IMHO is a pity. It is a very 
nice System, but how can use it if it was not ported to my OS? 

> : IMHO Oberon needs marketing, not marketing hype. These can be two 
> : different things.
> 
> I agree with you.

OK, then what about this: the first step in marketing should be to stop 
making mistakes. Why is System-3 still not available for SGI, RS6000, etc?
Why are V4 and S3 still not merged into System-5 ? Why did you lose the 
best selling point, which was portability of Oberon programs across 
so many platforms: Unix, Amiga, Mac and PC ? These are obvious problems 
which were created within developers community. You guys are bright enough 
to see these problems, then why do not you just gather in one (not so large)
room and just make right decisions? This would be a very large step 
in the right direction. Marketing could follow later.

I hope all this is neither impossible nor outrageous. What do you think?

Wojtek

In article <3159C50B.4E9E@lrz.fh-muenchen.de>,
Peter Landmann  <peter.landmann@lrz.fh-muenchen.de> wrote:
> 
>---peter.landmann@lrz.fh-muenchen.de----------------------------------
>   'C is a whith syntax sugared assembler' (N. Wirth)
>

I think that lost a little in the translation.  Maybe, "C is
assembler sugared with syntax" ???
-- 
mailto:mgallo@slonet.org                http://www.slonet.org/~mgallo/

M. Gallo wrote:

> >   'C is a whith syntax sugared assembler' (N. Wirth)
> 
> I think that lost a little in the translation.  Maybe, "C is
> assembler sugared with syntax" ???

Yes, sorry for my bad english. The original quote was:

'C ist ein mit Syntax verzuckerter Assembler'

(from an interview with Niklaus Wirth in the german 'Amiga-Magazin' 
10/89, p. 14)

---peter.landmann@lrz.fh-muenchen.de----------------------------------

skulski@fermi.nsrl.rochester.edu wrote:

>
> OK, then what about this: the first step in marketing should be to stop
> making mistakes. Why is System-3 still not available for SGI, RS6000, etc?
> Why are V4 and S3 still not merged into System-5 ? Why did you lose the
> best selling point, which was portability of Oberon programs across
> so many platforms: Unix, Amiga, Mac and PC ? 

Three questions, all have the same answers:

First because S3 is not released in source code! This makes V4 interesting for
people who shouldn't use it ( implementors of systems). V4 and S3 are so
close, that it even wouldn't be worth renaming System 3 when including and
porting V4 stuff.  ( not necessarily to Gadgets, which can sometimes be
difficult to program). This, of course, will not happen until S3's source will
be made avaiable...

And second, it is ETH politics to release V4 with the Oberon-2 compiler and to
NOT release S3 with the Oberon-2 compiler. So they can't merge them. 

But consider that even today S3 has four more ore less incompatible editors (
Edit, ET, Script and TextGadgets). Together with Write, you would have five!
Personally i hope that System-5 will someday become a more object oriented,
stripped down and cleaned up Gadgets system that runs on top of both, V4 and
S3.

Peter