[Coin-standards] An argument for public domain modeling infrastructure

[Leonardo B. Lopes]

> 
> At 12:45 PM 4/11/02 -0500, Leonardo B. Lopes wrote:
> >On Thu, 11 Apr 2002, Alan King wrote:
> >
> >It is interesting that this idea of a modeling standard keeps coming up
> >over and over again. That is not what I see as the outcome of this work.
> 
> Maybe it *should* be?????
> 

I don't think so. I could be wrong, but I don't think that a modeling
standard will happen. In addition to the technical aspects, which I've
talked about earlier, there is a strategic aspect. Modeling environments
add real, immediate value, which low level standards don't, by themselves.
In addition, they are hard to design and implement. So they fit the
proprietary business model very well, and are not good candidates for open
source or similar development models like this one.

> Java works this way.  The vendors are free to write their own JVM's, and 
> there is a reference implementation that Sun will license to other vendors.

Maybe I'm interpreting the Java case incorrectly, but I think it
strengthens the argument *against* an open source language. As I
understand it, the Java license allows licensees to reimplement the JVM,
which is basically an OS-abstraction layer, and not the language itself.
Without that, Java would make no sense at all. In fact, that was in rough
terms what their fight with MS was all about. MS was adding all sorts of
windows-specific extensions to the language, and Sun didn't like it. In
addition, Sun has thrown a nice chunk of change down the drain with java.
The only reason they did it was that it suited their strategic interest.
Good luck finding someone with that much to gain, and that much cash to
shell out, in our industry.

> You 
> need to design a modeling language that has extensibility along with an 
> implementation that supports extensions.  And from what I hear from people 
> who have worked on this, it is really hard to design and implement such 
> languages (modeling or otherwise).
> 

Yes, it is. I'd actually argue it can't be done. Or let me weaken that a
little: If it were done, the benefit from the standard would not dominate
the tradeoffs in the design.
 

========================================================================
Leonardo B. Lopes                                       leo at iems.nwu.edu 
Ph.D. Student                                              (847)491-8470
IEMS - Northwestern University              http://www.iems.nwu.edu/~leo