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

snmpconf BCP/Taxonomy: Plan of action

I referred yesterday in my mail to the notion of an emerging "consensus" as 
to what we are going to do regarding the policy-config-taxonomy discussion 
in the current BCP (aka section 7.2), with some of the confusion and 
concerns that have been expressed.

Jon sent out a note yesterday indicating some of our ultimate plan of 
attack, but we've been somewhat emboldened in proposing a quick multistage 
approach with the mail (and subsequent offlist followup) we've been making 
with Bob Moore.

Soooo...if it's acceptable to you, the authors of 
draft-ietf-snmpconf-bcp-05.txt are going to try the following things in 
preference order, to see which one "hits" the best with the group:

a) Jon and I have been forging a new set of per-layer attributes which we 
feel/hope addresses the need for instance-specific attribute notations to 
permeate the entire taxnomomy.  Frankly, we're trying to simplify it in the 
sense of not "multidimensionalizing" the entire reference model.  We'd like 
to, in ones sense, go beyond what Steve {W} has been advancing wherein he 
propses a single extra segmentation based on instance-specificity (without 
needing multdimensionalizing nevertheless).  Without getting into this too 
much, I'll be posting the idea of this later today (I hope), tomorrow if not.

b) This is not necessarily mutually exclusive of (a).  Bob Moore coined my 
favorite phrase of the day: "scope of applicability".  I sent a post to the 
list last night indicating that yes, in fact this does loosely describe 
some of the characteristics that have affected the usage model I have (and, 
I believe Jon as well) for the taxonomy.  In rereading the section in the 
BCP (and its preceding context in 7.1), the confusion and outrage :) on the 
part of Steve {W} and others here is increasingly understandable--that we 
are confining this to the information model involved, and its doesn't work 
as a taxonomy with that primary purpose in its current form.  HENCE, our 
action item if we get here is to *really* work on clarifying the text here 
to define WHAT we're trying to get at, what we AREN'T trying to get at 
primarily, and what we WON'T be effectively covering at all with the 
taxonomy.  In many ways, this may be the least contentious step to take 
next, but as Jon and I have been using a lot of intra-LATA phone minutes 
discussing what lies behind (a) above in the last week , I'd like to 
advance it just for review if nothing else.

c) as Jon indicated in a post to the group yesterday, if we can't find 
rough consensus in the preceding two steps, we're prepared to remove the 
discussion from the bcp altogether.  I'd hope the group would give us the 
opportunity for an honest attempts at (a) and (b) first however.



Wayne F. Tackabury              Internet: wayne@goldwiretech.com
Gold Wire Technology            Phone: (781) 398-8819
411 Waverley Oaks Rd., Ste 304
Waltham, MA  02452             Cell: (617) 699-6156