"Black, David" <david.bl...@emc.com> writes:
>> I'm definitely in favor of polling the working group about TCP-SO, but
>> I'd like to suggest that we make the question much more precise. As it
>> stands I don't really know what "support" means in this context.
>
> Really? The above summary up to the point of plug-pulling behavior
> is an accurate restatement of the question ...
But in your previous email, you talked about two different ways of
pulling the plug:
| - Should tcpinc support TCP simultaneous open? Yes or No
| -- If "No," should TCP simultaneous open connection attempts
| be abandoned or proceed unencrypted?
How do I vote if I'm okay with either "proceed unencrypted" or minimal
support, but I'm strongly opposed to "abandon the connection" or more
invasive TCP-SO support? As I stated in my previous email, the options
seem to be laid out like this:
no |
support | support
|
A B | C2 C3 C4 C5
As currently phrased, the question before the working group is whether
to select the left or right set of options. However, my preference is
to be as close to the line as possible: C2 (current ENO) is my
favorite, but I can also easily live with B (proceed unenecrypted). I
find both A (abandon connection) and C4 (always work) to be horrible
options (C4 even worse than A), and believe C5 (leave it to specs) is
not forward looking enough and would have to be combined with C2 to
avoid ruling out efficient use of post-quantum crypto down the line.
So I don't know where this leaves me with respect to the current
question. If we can rule out A immediately, this would make me happier,
but even so the left vs. right choice would not clear to me, given that
I prefer B to C4 and C5.
David
_______________________________________________
Tcpinc mailing list
Tcpinc@ietf.org
https://www.ietf.org/mailman/listinfo/tcpinc
