Re: [Masque] MASQUE and Tor "pluggable transports"

[David Schinazi <[e-mail]>]

Thanks for the link, Philipp. That was news to me, and really interesting!

The first thought that comes to mind is that we could use QUIC encryption
between relays (instead of DTLS) and even for the layered onion encryption.
That would require changing how QUIC authenticates, but IETF QUIC was
designed to allow that by having quic-transport
<https://quicwg.org/base-drafts/draft-ietf-quic-transport.html> and quic-tls
<https://quicwg.org/base-drafts/draft-ietf-quic-tls.html> be separate
documents.

Now it's just a matter of writing (a lot of) code :-)

David

On Wed, Aug 7, 2019 at 3:05 PM Philipp Winter <[e-mail]> wrote:

> On Tue, Aug 06, 2019 at 12:29:37PM -0700, David Schinazi wrote:
> > We're still at the early stages for MASQUE - the document doesn't yet
> > define a wire-format that can be implemented - but once we get a first
> > draft of that written up and start implementing it, I'd love to
> > experiment with pluggable transports to see how well the obfuscation
> > bit of MASQUE plays with them.
>
> Thanks for your interest!  I'd be happy to help out once MASQUE reaches
> this stage.
>
> > Additionally, once we've accomplished the goal of using MASQUE to help
> > obfuscate Tor traffic, there is interest in the community to go for
> > something more ambitious, which is a tighter integration between
> > MASQUE (as a way to proxy QUIC over QUIC) and onion routing. We
> > suspect that this integration could help with overall performance, but
> > a lot of code needs to be written to see if the theory hold water.
>
> FYI, here's a recent summary of this effort, which you may already be
> familiar with:
> <https://lists.torproject.org/pipermail/tor-dev/2018-March/013026.html>
>
> Cheers,
> Philipp
>