[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: snmpconf BCP re: policy layering

Hi Bob...

At 03:41 PM 7/25/2001 -0400, Robert Moore wrote:

>  This isn't quite
>Jon's taxonomy, where it's levels of abstraction rather
>than scope of applicability that work their way down,
>in one dimension, to the instance level.  But I wonder
>if there's been some element of this case that has bled
>over into Jon's taxonomy?

That's very perceptive of you, because in a reduced way, that's exactly 
what I am consciously doing, partially.  At a high level, that's the key 
determinant for what a policy *does*.  I will argue that what this 
accommodates nicely is the non-SNMP information scope which we deal with 
all the time in commercial policy-configuration products.  In these, the 
'abstraction' we are dealing with may not be nice and hierarchical and 
lending itself towards a strict Saperian or even Waldbusserine layering 
(IOS config or TL/1, anybody?), but may be this nonlinear set of controls, 
upon which we are imposing hierarchical "scope" of application (which does 
lend itself to said layering).  The language of the BCP (I'm too tired to 
pull it up :) doesn't limit the applicability of the taxonomy/reference 
model to SNMPConf, or any SNMP-derived policy world, nor imho, should it.

I could elaborate at these points, but I don't know that it's incredibly 
productive at this moment.  Jon and I have been discussing a somewhat 
different model enhancement, which I want to refine with him over maybe one 
more private discussion before we advance it in the next day or 
so.  Suffice to say, however, that at least to my ears, a voice of "rough 
consensus" is being heard, and that the boundaries of *just what* to 
address are emerging.  Bear with, I think we're close to resolving this one 
way or another. :)