Re: Closing on NIM requirements

[Harald Tveit Alvestrand <Harald@Alvestrand.no>]

At 17:42 16.04.2000 -0400, Jon Saperia wrote:
>on 04/16/2000 4:58 PM, Harald Tveit Alvestrand at Harald@Alvestrand.no
>wrote:
>
> > UML, from my brief acquaintance, has about half a dozen different modelling
> > methods, all better at showing some properties of the system than others.
> > One OO credo (or at least part of one) is that data and operations need to
> > be closely bound together, because neither is understandable without the
> > other; I subscribe to that.
>
>I have a stated a preference for UML in the past. The interesting question
>that this discussion begs is: if people believe LDAP and SNMP (and others)
>are not able to effectively represent what is in the 'higher-level' models,
>what should be done?

there's a distinction between "effectively represent" and "represent" that 
I think is important.
We may have to accept that when you try to manipulate a business inventory 
system through SNMP, there are things you just can't get at easily - or at 
all; see the recent discussions on managing configurations through SNMP, 
for instance.

>I have mentioned a number of times that the technology
>specific details, whether they be LDAP, SNMP, or anything else will always
>tend to impinge on the higher layer modeling. Even in what are generally
>considered to be OO languages, there are important differences. For example,
>how do I do multiple inheritance in Java?  The point is not to pick on Java
>or any other technology. My point is that either the modeling language be
>reduced to the least common denominator - which probably nobody wants, or a
>plan be put in place to bring up the infrastructure elements to the point
>where they have what people feel is needed.

or that we accept that some of the tools can't do everything, which can be 
a Good Thing - it allows people to choose appropriate tools for a job.

I'd prefer the modelling language to be the superset - something which is 
able to describe any system that can be manipulated easily with SNMP and 
LDAP, and which is also able to model some systems that can't.

>  In that case a fairly protracted
>but appropriate discussion of tradeoffs would probably have to take place.

the question of subset/superset should probably be part of the requirements 
document. My reading of the requirements doc was that it required the superset.

See for instance requirement 15, which explicitly says that the infomodel 
should be easily able to describe something that isn't easy to describe in 
LDAP. (actually the specific example is representable, but requres a new 
attribute syntax rather than a new attribute type - that's something that's 
not beautiful in LDAP)

I might have had my odd-colored glasses on while reading.

                       Harald



--
Harald Tveit Alvestrand, EDB Maxware, Norway
Harald.Alvestrand@edb.maxware.no