[tcpm] Updated draft-tcpm-moncaster-tcp-rcv-cheat-01

Hi,

We have now submitted a new version of our draft for testing TCP
receiver's responses to packet re-ordering and loss. We have
incorporated all the comments that were received in response to the
initial draft. Hopefully this new version will prove more acceptable to
the people who had concerns about the initial version.

Hopefully it will be uploaded onto the ID shadow directory shortly. In
the meantime, HTML and txt versions of the draft can be found here:
http://www.jungle.bt.co.uk/people/rbriscoe/ext/pubs.html#tcp-rcv-cheat

The main changes that we have made are:

Clarified that this is testing whether receivers are compliant with TCP
regarding responding to out-of-order segments
Highlighted that we don't expect all TCP implementations to incorporate
this test
Added text on possible interactions with other TCP variants

Our overall aim with this draft is to set out a safe method to test
receiver compliance with TCP over packet reordering so that if
optimistic acknowledgements do materialise as a threat, the IETF already
have a safe solution that can be used to counter the threat.

We would welcome any comments, especially if people still feel uneasy
about the overall aim of the draft and how it sets out to achieve it.

We are keen that this document should be adopted as a WG item. If it
were to be adopted we would suggest that the filename be altered to
reflect the changed emphasis of the draft by altering "-rcv-cheat" to
"-rcv-compliance".

Toby

________________________________________________________________________
____
Toby Moncaster, <toby.moncaster@bt.com> Networks Research Centre, BT
Research
B54/70 Adastral Park, Martlesham Heath, Ipswich, IP53RE, UK.  +44 1473
648734 