Re: [Ecrit] Content-ID: Way forward without SIPCORE impact

[Randall Gellens <[e-mail]>]

At 12:58 PM +0000 10/6/16, Christer Holmberg wrote:

>  Hi,
>
>  One way to solve the Content-ID issue would be to ALWAYS use a 
> multipart MIME, even if we only carry one body part. See example in 
> section 12.2.2.3 of RFC 6086.
>
>  Another advantage would be that we would always the same 
> Content-Disposition header field value for the MSD/control body 
> parts. Currently the value is dependent on whether there are other 
> body parts in the message or not, which is a little strange. So, 
> the multipart/MIME would have C-D:info-package and the MSD/control 
> body part would have C-D:by-reference.
>
>  Comments?
>
>  Regards,
>
>  Christer
>
>  (It may still be a good idea to define usage of Content-ID for SIP, 
> but then the ECRIT specs would not be held up due to that work)


Hi Christer,

In view of the slow progress of your draft, I think the proposal to 
always use a multipart is the best option.  As you say, it simplifies 
the implementation because the C-D will always be "By-Reference".

(As I recall, the same idea was floated some time back, during the 
discussion about INFO.)

I agree that your draft should still progress, as it seems an 
oversight that C-ID is not included.

--Randy

-- 
Randall Gellens
Opinions are personal;    facts are suspect;    I speak for myself only
-------------- Randomly selected tag: ---------------
Red Tape Holds Up New Bridge
--Newspaper headline