I-D Action:draft-york-sipping-p-charge-info-10.txt

[Paul Kyzivat]

I guess this draft has been going on for a long time.
I took a quick look at this version.

Because it is intended to document existing practice I won't take issue
with the good and the bad of the proposed form.

I do have some issue with the formal syntax and the accompanying formal
semantics. There are two header parameters defined: npi and noa. I
suppose those names have some mnemonic signficance, but it is nowhere
described in the draft. They might as well be "foo" and "bar".

Similarly the value syntax for those is gen-value, which is more or less
"anything". Appendix A lists values from ANSI T1.113 in relation to npi,
but there is no normative statement that the values should/must conform
to that, or what should happen if they do not. And there isn't even that
kind of hint for noa.

If, as is hinted at in appendix A, the values for npi are to be 3 digit
numbers (or numbers in some range), then it would be helpful to document
the syntax that way.

If the intent is that these values really can be "anything", which must
be agreed to out of band by the parties using the header, then that
should be stated.

Similarly there really is no documented significance to the URI in this
header. There should be some statement about the intended purpose of
this URI.

Thanks,
Paul

BTW, we don't really have a sipping group any longer. I presume this is
to be treated as an individual submission or AD-sponsored. I've replied
to the sipping mailing list, but that won't get a lot of readers. It
might be best to announce it on the dispatch list.

A New Internet-Draft is available from the on-line Internet-Drafts directories.
Title : P-Charge-Info - A Private Header (P-Header) Extension to the Session Initiation Protocol (SIP)
Author(s) : D. York, T. Asveren
Filename : draft-york-sipping-p-charge-info-10.txt
Pages : 13
Date : 2010-10-25
This document describes 'P-Charge-Info', a private Session Initiation
Protocol (SIP) header (P-header) used to convey billing information
about the party to be charged. This P-Header is currently in
production usage by a number of equipment vendors and carriers and
this document is submitted to request the registration of this header
with IANA. This P-Header may also be used in some situations to
carry the ISUP Charge Number parameter for PSTN interconnection.
http://www.ietf.org/internet-drafts/draft-york-sipping-p-charge-info-10.txt
ftp://ftp.ietf.org/internet-drafts/
Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.
_______________________________________________
I-D-Announce mailing list
https://www.ietf.org/mailman/listinfo/i-d-announce
Internet-Draft directories: http://www.ietf.org/shadow.html
or ftp://ftp.ietf.org/ietf/1shadow-sites.txt

_______________________________________________
Sipping mailing list https://www.ietf.org/mailman/listinfo/sipping
This list is for NEW development of the application of SIP
Use sip-***@cs.columbia.edu for questions on current sip
Use ***@ietf.org for new developments of core SIP

[Dan York]

Paul,

Yes, this draft has been around for a long time. I was going to submit it as part of the RFC 3427 process but it was then suggested that I wait because of the RFC 3427bis process that became RFC 5727. So it was kind of stuck in limbo for a bit and now we'd like to move it forward.

Post by Paul Kyzivat
I guess this draft has been going on for a long time.
I took a quick look at this version.

DY> Thank you. I've not received much feedback outside the folks who care about this header because they currently use it.

Post by Paul Kyzivat
Because it is intended to document existing practice I won't take issue with the good and the bad of the proposed form.

DY> Thank you. That *is* the intent... P-Charge-Info has now been in usage for probably 5 years and so this is intended to document that usage.

Post by Paul Kyzivat
I do have some issue with the formal syntax and the accompanying formal semantics. There are two header parameters defined: npi and noa. I suppose those names have some mnemonic signficance, but it is nowhere described in the draft. They might as well be "foo" and "bar".

DY> Good point. They come from the world of PSTN and should be spelled out here.

Post by Paul Kyzivat
Similarly the value syntax for those is gen-value, which is more or less "anything". Appendix A lists values from ANSI T1.113 in relation to npi, but there is no normative statement that the values should/must conform to that, or what should happen if they do not. And there isn't even that kind of hint for noa.

DY> Good feedback.

Post by Paul Kyzivat
If, as is hinted at in appendix A, the values for npi are to be 3 digit numbers (or numbers in some range), then it would be helpful to document the syntax that way.

DY> How do we specify digits in ABNF? For NPI we want to restrict it to 3 digit numbers, but I was not clear on how to do that in ABNF and so I wound up doing "gen-value".

Post by Paul Kyzivat
If the intent is that these values really can be "anything", which must be agreed to out of band by the parties using the header, then that should be stated.
Similarly there really is no documented significance to the URI in this header. There should be some statement about the intended purpose of this URI.

DY> The URI is the billing identifier that is being transmitted from one system to the other. I thought this was explained in the use-cases in section 4, but when I look at sections 4.x I can see that I am not being explicit there about the info being inserted into the URI.

Post by Paul Kyzivat
BTW, we don't really have a sipping group any longer. I presume this is to be treated as an individual submission or AD-sponsored. I've replied to the sipping mailing list, but that won't get a lot of readers. It might be best to announce it on the dispatch list.

DY> Correct, I am assuming at this point it will be an individual submission. I have kept the name intact purely because it did originate back in the SIPPING group in 2008 and at this point I want to finalize the draft and submit it for expert review. Good suggestion to send it over to DISPATCH, though.

Thanks again for the feedback,
Dan

[Tom Taylor]

Hi.

On 26/10/2010 9:10 PM, Dan York wrote:
...

DY> How do we specify digits in ABNF? For NPI we want to restrict
it to 3 digit numbers, but I was not clear on how to do that in ABNF
and so I wound up doing "gen-value".

[PTT] That's easy: NPI = 3DIGIT

See section 3.7 of RFC 5234.

_______________________________________________ Sipping mailing list
https://www.ietf.org/mailman/listinfo/sipping This list is for NEW
development of the application of SIP Use

_______________________________________________
Sipping mailing list https://www.ietf.org/mailman/listinfo/sipping
This list is for NEW development of the application of SIP
Use sip-***@cs.columbia.edu for questions on current sip
Use ***@ietf.org for new developments of core SIP

[Paul Kyzivat]

DY> How do we specify digits in ABNF? For NPI we want to restrict it to
3 digit numbers, but I was not clear on how to do that in ABNF and so I
wound up doing "gen-value".

You can find examples in 3261.
If you want exactly three digits, you can simply use: 3DIGIT
or if you want variable up to three you can use: 1*3DIGIT
(3261 gets its definition of DIGIT from 2234. You can inherit it from 3261.)

Thanks,
Paul
_______________________________________________
Sipping mailing list https://www.ietf.org/mailman/listinfo/sipping
This list is for NEW development of the application of SIP
Use sip-***@cs.columbia.edu for questions on current sip
Use ***@ietf.org for new developments of core SIP