Re: comments on draft-presuhn-referent-00.txt

[Margaret Wasserman <mrw@windriver.com>]

> References between data structures are (a) unavoidable and (b) always
> a pain to deal with. Any serious simplification comes from either
> restricting what you can do or having some support on the boxes. But
> if implementing reference counting or similar things is already too
> high of a cost, what can we do?
If you were implementing the MIB and the datastructure at the same
time, implementing reference counts (or automatic clean-up of
dependent rows) wouldn't be a problem.

The problem comes in when you try to map a MIB onto an existing
datastructure...  If the MIB requires one model, and the current
implementation uses another (that can't be changed because other
modules and/or customer code may depend upon it), things can get
ugly.

So, IMO, it is best for the MIB not to place too many constraints
on the behaviour of the underlying datastructures.

Margaret