RE: [ietf-provreg] EPP Document Updates

[Edward Lewis <Ed.Lewis@neustar.biz>]

At 16:43 -0400 8/3/05, Hollenbeck, Scott wrote:
>>  From: Klaus Malorny [mailto:Klaus.Malorny@knipp.de]

>>  One thing that really puzzles me in this context is that a
>>  registrar cannot
>>  allow the disclosure of one component (that is normally not
>>  disclosed) and to
>>  disallow the disclosure of another component (that his
>>  normally disclosed) at
>>  the same time. But maybe I am just misunderstanding the
>>  wording. I think I have
>>  to reread this at a later point in time.
>
>Remember that the whole "disclose" thing got crammed into the spec at
>the last minute at the demand of the IESG.  There were parts that didn't
>make sense then (to me anyway) and I'm sure they still don't make sense
>now.  This is another one of those things that we should really consider
>removing (if no one is using it) or revising (if it's not quite right).

Our implementers support dumping the discloser elements as currently 
specified.  The issue is that the communication is "binary" (ON/OFF, 
YES/NO).  There is legacy data that is not tagged (it wasn't on 
collection), i.e., it should be NULL, UNSPEC, or DON'T CARE.

One bug is that on "create" the submission might be left unspecified 
because the client "doesn't care".  But on "info" the submission, if 
un"tagged" is assumed to follow default behavior.  Doesn't seem very 
symmetric in a protocol sense.

A suggestion is made to also make this a per time tag (but I have a 
sneaking submission that we nixed that in the discussions in 
early/January 2003).
-- 
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Edward Lewis                                                +1-571-434-5468
NeuStar

If you knew what I was thinking, you'd understand what I was saying.