Hello Tom,
I do not see incorporation of any of the observations made by my colleague
Simran Chadha in the new RFC 3525, as "promised" by you in your earlier
mail.
Please clarify. Thanks !
Suresh
----- Original Message -----
From: "Tom Taylor" <taylor@nortelnetworks.com>
To: "Simran Chadha" <simran_chadha1@rediffmail.com>
Cc: "Megaco List" <megaco@ietf.org>
Sent: Wednesday, February 05, 2003 12:29 AM
Subject: Re: [Megaco] Re: Usage of LocalControl,Local and Remote
Descriptors!
That "cannot not" is a typo -- I'll remove the "not" during the "authors'
48 hours"
interval prior to formal publication as a new RFC.
On your second point: the two paragraphs apply independently to
ReserveGroup and
ReserveValue. That is, it is possible to have ReserveGroup TRUE and
ReserveValue
FALSE -- in which case the MG reserves for multiple session descriptions
but is
restricted to one codec in each -- or vice versa. Or both could be TRUE,
or both FALSE.
On the third point, the intended meaning would be conveyed by replacing
"if it
cannot support any of the alternatives" with "if it can support none of
the
alternatives". Perhaps I can also do that during the "authors' 48 hours"
period,
though we would usually do it through the ITU-T Implementor's Guide.
The task of developing an unambiguous protocol specification is not
simple. Your
points have eluded others in the three years since first approval. Thank
you for
your contribution.
Simran Chadha wrote:
Hi List,
Can someone please help me at the earliest on the issues raised earlier
by me - particularly, usage of LocalControl,Local and Remote Descriptors
?
Thanks in advance.
Simran Chadha
On Thu, 30 Jan 2003 Simran Chadha wrote :
Hi List,
Can you please help me with clarifications in my understanding of
sections 7.1.7 and 7.1.8 on the use of LocalControl,Local and Remote
Descriptors.
As per draft-ietf-megaco-3015corr-03, in section 7.1.7,
If the value of a Reserve property is True, the MG SHALL reserve
resources for all alternatives specified in the Local and/or Remote
descriptors for which it currently has resources available. It SHALL
respond with the alternatives for which it reserves resources. If it
cannot not support any of the alternatives, it SHALL respond with a
reply to the MGC that contains empty Local and/or Remote descriptors.
The use of two negations - "cannot not" in the last statement will
imply that - If it can support any of the alternatives, it SHALL
respond with a reply to the MGC that contains empty Local and/or
Remote descriptors.
I don't think this is what is meant to be conveyed here.
Also in section 7.1.7 the first para starts - If the value of a
Reserve property is True, .....
while the second para starts - If the value of a Reserve property is
False, ....
Now as there are two Reserve properties - one Reserve Group and the
other Reserve Value, One could be False and the other could be True.
Then in this case what will be the inference - Should para 1 apply or
para 2 ? Both the paras would seem to be mutually exclusive in what is
needed to be done.
Also statements like "if it cannot support any of the alternatives ...
" - will have two meanings.
1. Any - Not even one - NONE of the alternatives is supported.
2. Any - Atleast one - While a few alternatives can be supported, a
few might not be.
The protocol needs to be more clear on such aspects in the documents
henceforth. The English language is understood in different parts of
the world. Variations to agreed usage of the language terms will make
it difficult for new users of Megaco protocol, like me.
Can someone please educate me on the exact usage and meaning of
LocalControl,Local and Remote Descriptors ?
Thanks and awaiting comments and clarifications
Simran Chadha
_______________________________________________
Megaco mailing list
Megaco@ietf.org
https://www1.ietf.org/mailman/listinfo/megaco
_______________________________________________
Megaco mailing list
Megaco@ietf.org
https://www1.ietf.org/mailman/listinfo/megaco