[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: Closing on NIM requirements
Lakshmi,
> Walter
> Couple of poins. Unfortunately it will be difficult to get
> back with detailed
> comments
> before the end of the week as I am travelling for a ballot
> resolution meeting.
> These go very late. In general I agree with the high
> level requirements and I think it is a good starting point. I
> can get back to
> you
> by next Saturday.
I expect this discussion to continue for a while and for some additional
issues to be raised as well. Don't be too concerned about getting your
comments in before the week is out.
> As for the comment from Dave below, I agree with the general
> theme of being
> implementation idependent. A language like UML does include
> operations and
> attributes. This is definitely being used in ITU and other groups
> as a way to model without choosing a specific implementation.
> We have considered this to be implementation dependent.
>
I would like to hold off discussing language specific issues until after we
have closed on the requirements draft.
> I do not follow why in the mail below it is stated that
> methods are not part of
> an information model. Instead of calling it as method, would
> operations be
> a better term? Also I do not view operations as a grouping of
> attributes at all.
> The operations does have pre and post conditions that must be
> expressed.
> They have in and out parameters and they may or may not be attributes.
> These characteristics are not specific to CORBA IDL or a
> protocol model.
>
I am hopeful that Dave has clarified his position for you in the subsequent
messages.
> I would like to also propose that we should look at Rec
> M.3020 which does
> discuss
> information model without being specific to a protocol.
Please provide a reference (preferably a URL) for people like myself who are
not familiar with this work.
regards,
-Walter