Re: Wrap up and last call: sin6_scope_id semantics

[JINMEI Tatuya / 神明達哉]

>>>>> On Thu, 16 Aug 2001 23:28:03 +0900 (JST), 
>>>>> Atsushi Onoe <[EMAIL PROTECTED]> said:

> Some isssues:
>       - need to define the mapping between scope and top 4 bit?
>               my guess is yes.  MIB peaple may want to map a value
>               into alias (e.g. "link") from a remote system.

Yes.  As for the mapping, using the 4-bit "scop" value in multicast
addresses is the easiest (and most intuitive) way, although we should
consider the backward compatibility issue below.

>       - should care existing implementations?
>               yes.  But I guess most of such applications use a interface
>               index for scope_id, so it can be saved either:
>           - assign scope value (4bit) 0 to link-local.
>               or
>           - implementation MAY accept the value 0 in implementation-
>             specific manner, but SHOULD NOT return 0 for scope (4bit).

If we take the latter way, even older applications will see non-zero
zone IDs for the link-local scope.  Doesn't this annoy the
applications?

>       - need to introduce new functions to map scope-name and the value?
>               I hope not to define new functions, so want to _define_
>               the textual representation (e.g. "link", "site", ...)
>               instead.

I agree.

                                        JINMEI, Tatuya
                                        Communication Platform Lab.
                                        Corporate R&D Center, Toshiba Corp.
                                        [EMAIL PROTECTED]
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page:                      http://playground.sun.com/ipng
FTP archive:                      ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------