ExitNodes for encrypted connects only are not possible. Why?

Karsten Loesing karsten.loesing at gmx.net
Sat May 9 17:36:56 UTC 2009


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 05/09/2009 01:38 PM, Gitano wrote:
>> It's unlikely that the criteria you pasted above will be changed. There
>> need to be some criteria, and if almost every node matches them, the
>> flag would be useless.
> 
> Ok, but adding one more 'secure' port beside 443 would be enough in this
> case.

I'm not sure what you are trying to achieve with that. The idea is not
to flag as many nodes that permit exiting as Exit nodes. The idea is to
relieve the exit nodes carrying most of the exit traffic from acting as
middle nodes, so that they can push more exit traffic. The same is done
for guard nodes, by the way. It's unlikely that your node would carry as
much exit traffic with the five ports you mentioned as compared to other
nodes that already meet the requirements for the Exit flag.

Of course the requirements could be lowered to assign the Exit flag to
more relays. But it defeats the purpose if too many nodes have that
flag. In the end, all nodes would see the same load as before, without
the Exit flag.

I'm not saying that the current definition for the Exit flag is perfect.
But right now we lack good data to come up with a better definition.

Best,
- --Karsten

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkoFvy8ACgkQ0M+WPffBEmXMawCgkzkbYdk1J4F6y7VSxdfxUKTm
LeoAoMNHbXYG6BqSIFu2dpq3VQ+He56t
=O2DW
-----END PGP SIGNATURE-----



More information about the tor-talk mailing list