ietf
[Top] [All Lists]

Re: Call for Papers: IAB Workshop on Stack Evolution in a Middlebox Internet (SEMI)

2014-12-04 06:09:55
I wrote:

The problem, however, is that the review comment are so
surprising that the result of SEMI workshop is very simply
proven (see below) to be "incorrect and incomplete".
are there anyone who argue against the proof?

Yes, because it's not proof.

Why? Please argue, not just state.

Let me elaborate a little more.

It is not productive to declare something not a proof, because it
dose not make the problem address the proof disappear.

And the problem is that, with the following situation:

        a) there are a client behind a middlebox and a server
         communicating over the Internet

        b) the middlebox has ALG function with some protocol
        at some port

        c) the function somehow harms communication between the server
        and the client

        d) some part of the harmful ALG function is unknown to the
        client

the communication between the server and the client is harmed.

Then, the only possible approaches to completely remove the harm is:

        1) to avoid the port used by the ALG, which requires cooperation
        between the server and the client sides

        2) to replace/upgrade the middlebox and the client so that the
         harm by the middlebox is known and reacted against by the
        client

and another approach of:

        3) to make the client react against the harm of the function
        as much as possible even though the client is at a loss against
        the unknown part of the function

is, obviously, an incomplete solution.


As such, declaring the content of my previous mails "not proof" do not
make the problem disappear.

A possible productive counter argument is to show an approach 4) or
more which could completely remove the harm.

I do welcome it.

If you can't to do so, you should just accept that the problem is
real, even though it might deny the originally intended purpose of
SEMI2015.

Can you argue against the problem? Or?

                                                Masataka Ohta