[Masque] Feedback requested on paper draft that mentions MASQUE

[David Fifield <[e-mail]>]

I'm writing a paper for the upcoming Free and Open Communications on the
Internet (FOCI) workshop. The topic is somewhat related to MASQUE, in
that I advocate for the use of QUIC or a similar session/reliability in
the interior of circumvention protocols. It's not quite the same as
MASQUE, because I'm not suggesting to use QUIC as the outermost layer
exposed to the censor, but as an inner layer providing greater
flexibility in program design.

This is the current draft. The final revision is due 28 July.
https://www.bamsoftware.com/papers/turbotunnel/turbotunnel-20200721.pdf

My primary concern is to ensure that I am representing MASQUE fairly and
accurately. Of course I welcome comments in general.

	MASQUE is a proposal to colocate covert proxy servers with
	existing web servers, over HTTP/2 (TLS/TCP) or HTTP/3
	(QUIC/UDP). Despite the use of QUIC, MASQUE is not really an
	example of the Turbo Tunnel idea. The key difference is that
	MASQUE puts QUIC on the outside of the protocol stack, not the
	inside, and makes QUIC itself act as the obfuscation layer, with
	the goal of blending in with normal web traffic. Alternatively,
	MASQUE could be regarded as an optimization in which the same
	protocol (QUIC) works across three layers: session/reliability,
	obfuscation, and network transport (refer to Figure 1). The
	observation is that not only is QUIC a convenient session
	protocol, it also makes a good cover protocol, because of its
	encrypted-by-default nature and its increasing use on the
	Internet. Collapsing three layers into one avoids the overhead
	of encapsulating QUIC packets into some other protocol. The
	efficiency comes at the loss of some flexibility: the
	obfuscation of MASQUE is not ``pluggable''; the only option for
	obfuscation is as web traffic.

There's some additional discussion on MASQUE here:
https://github.com/net4people/bbs/issues/9

Thanks for your work on MASQUE.