I concur with the uselessness of the bits.
However, I've heard some implementers who where talking about
implementing extra code/logic to try to match what they thought the
semantic of those bits was/should be/should have been...
This is adding complexity, brittleness and opening doors to all kind
Yeah. I have to agree with James and Brian: in retrospect, the M/O bits are
useless and further discussion at this point is even more useless.
- Ralph
On 11/29/06 4:58 AM, "Brian E Carpenter" <[EMAIL PROTECTED]> wrote:
>> The M&O bits were
>> defined long before we had DHCPv6 in place.
>
> An
As per Magnus request, the draft has been updated. See inline what has been
changed.
BR,
/L-E
-First comment is an exception, it is technical: 4.4 page 11 doesn't
metion that ROHC has the capability to work without a feedback channel,
i.e., on simplex/one-way links. As it is both an unique an
The M&O bits were
defined long before we had DHCPv6 in place.
And they were discussed to death in the WG before the draft reached
WG consensus. I'm not inclined to reopen that discussion.
(Somebody added ietf@ietf.org to this thread. I have removed it.)
Brian
__
