Re: DRAFT Montreal minutes - X#NodeArchitecture comments

William Studenmund wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> On Jul 31, 2006, at 2:36 PM, David Wysochanski wrote:
> 
>  > Ping on the 2 items below - any comments?
>  >
>  > Also adding a 3rd item:
>  > 3) Change key value format from list-of-values to
>  > "iSCSI-name-value" (see 7/13/06 post).
>  >
>  > For this one, I'd rather leave it as list-of-values, since
>  > this is simple, underscores are just about as readable as
>  > spaces, and iSCSI-name-value seems a little strange
>  > to me in this context.
> 
> Actually, how about "iSCSI-local-name-value".
> 
> The thing I don't like about list-of-values is that it implies, to me 
> at least, list negotiations. Which we don't want here. :-) Also, list-
> of-values just vies us "text-value" plus comma. Among other things, 
> we don't have spaces, nor do we have CR.
> 


Don't you think this would look a little strange though:

X#NodeArchitecture

    Use: LO, Declarative
    Senders: Initiator and Target
    Scope: SW

    X#NodeArchitecture=<iSCSI-local-name-value>



The reason I think it looks strange is that what is being defined
is not an iSCSI name.  The "list-of-values" sounds more generic,
especially if you compare the definitions in 3720.
Granted, "list-of-values" and "iSCSI-local-name-value" are just
tags to describe string rules, but it looks odd.

However, it sounds like what you really want is just to expand
the character set beyond what text-value defines, to something
more broad such as UTF-8.  I guess I don't see that much value
in expanding it (are underscores that more unreadable than spaces,
and do we really need CR's?), but maybe you have some particular
use cases in mind.

Anyone else on the list have an opinion on the character set?

_______________________________________________
Ips mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/ips