Re: iSCSI: Text/Operational keys: booleans and ranges

[Paul Koning <ni1d@arrl.net>]

>>>>> "Mallikarjun" == Mallikarjun C <cbm@rose.hp.com> writes:

 Mallikarjun> Julian, I agree with your proposal to denote ranges with
 Mallikarjun> ":".

 Mallikarjun> Also, as somewhat of a digression, I think we need to
 Mallikarjun> say something about the legality (or lack there of) of a
 Mallikarjun> negotiation involving multiple ranges as in:

 Mallikarjun> Key=a:b, c:d, e:f

 Mallikarjun> My not-too-strong opinion is that it could be useful to
 Mallikarjun> allow this...

We don't have an identified requirement for this.  We don't have a
reason for adding this to a spec that is supposed to be as close to
frozen as we can possibly get.  

This is NOT the time to keep adding features just "because it could be
useful".

The spec as currently written describes the use of exactly one range.
Leave it that way.  I don't have a problem with changing the delimiter
to ':' but I DO strongly object to any further extensions.

     paul