Re: Methods, Inheritance, Exceptions, etc. (was: Re: Methods in SMIng ?)

[Juergen Schoenwaelder <schoenw@ibr.cs.tu-bs.de>]

>>>>> Joel M Halpern writes:

Joel> I believe that the relevant counter-argument is that people are
Joel> already doing those ugly things by hand, without the clarity
Joel> that comes from explicit method descriptions.  Continuing to use
Joel> a kludge because that is what people have done does not seem a
Joel> good answer when we can do significantly better.

And all I am saying is that "doing significantly better" requires
method support at the SMI level _and_ at the protocol level. (The
alternative to map generic methods invocations on the current protocol
operations with lots of machine generated objects is really really
ugly and so far, I have not seen a substantial proposal on how to
actually do that.)

In other words, we need cooperation between WGs to get methods right
and EOS is currently not chartered to do this. That is why I propose
to ignore methods right now in SMIng and to worry about methods when
it is time to do so - as an extension to SMIng.

/js

-- 
Juergen Schoenwaelder      Technical University Braunschweig
<schoenw@ibr.cs.tu-bs.de>  Dept. Operating Systems & Computer Networks
Phone: +49 531 391 3289    Bueltenweg 74/75, 38106 Braunschweig, Germany
Fax:   +49 531 391 5936    <URL:http://www.ibr.cs.tu-bs.de/~schoenw/>