REMAIL: Virtual remailers

[anonymous@extropia.wimsey.com]

-----BEGIN PGP SIGNED MESSAGE-----

I run two remailers; one is hal@alumni.caltech.edu and the other is
hfinney@shell.portal.com.  The first one is on a system to which I have
free access, while the second is on a system for which I am paying about
$40 a month.

I feel like the second one is "stronger" politically, since I am paying
hard-earned money for it.  Also, I remember the summaries that were posted
of last year's Hacker's conference, in which it was stated that the owner
of the Portal system was eager to support remailers.  I don't know if he
realizes that he is indirectly supporting them now since I am using my
Portal account to run one, but I feel that if they get some complaints and
come to me, I might be able to get the boss to step in on my behalf.

Given this situation, it has occured to me that it would be easy to have
the caltech remailer forward all mail to the portal remailer, to be
remailed from there.  This way nobody would receive objectionable mail
from the caltech account, and no one would ask for it to be shut down.

In the past, we have had suggestions that this would be a desirable mode
of operation for our cypherpunks remailers: to have the "front ends" be
a different set of machines than the actual remailers.  It was said that
this would represent a sort of "distributed computing" environment, a
"virtual remailer" that would span the network, thus making it harder to
shut down.

I did not really agree with these arguments, since I felt that the targets
would simply be the final-remailing machines, since these are the ones
from which people would receive anonymous mail.  If they were shut down,
then other machines would have to come on-line to replace them.  Given
that such replacements actually existed, I felt that it would be better
simply to use them from the beginning as stand-alone remailers, so that
there would be more remailers out there for people to use.

Now I am faced with a concrete test of this principle, and I'm soliciting
suggestions.  Would it be better to keep my two remailers operating, even
though I might eventually have to shut one down due if complaints arise?
Or should I make one just a front-end for the other, thereby creating a
"virtual remailer" (a term I don't really agree with) which spans the
two machines and which makes it unlikely that the front-end remailer at
caltech will be terminated.

Are there any advantages to having the caltech remailer if it just feeds
into the one at portal?  I'm not sure I see much point in keeping it
operating if it performs no useful task.

On the other hand, if people do see advantages, we could create a set of
"second-tier" remailers which would be politically safe.  They would
always feed into one of the "first-tier" remailers which would be the
ones which would actually send mail.  No anonymous mail or postings
would ever come from these second-tier remailers, hence their operators,
owners, and sysops would receive no complaints.  Perhaps more people
would be willing to run remailers on this basis, knowing they were
relatively immune from political pressure.

Hal Finney
hfinney@shell.portal.com

-----BEGIN PGP SIGNATURE-----
iQCVAgUBLF2/d6gTA69YIUw3AQEHOQP/dU996sl0yQk8FlbSVG3LjUzLOIg7ktGs
57IRPU9zWJXOTGbxxhcA/p+kApXzU4hwnLV4ch9+DFst/hPFDMoHuuetmUMSLscL
EjaCz5ySzS532i/6TdfNbHMiDMgpWNIorQCysC+Ilpi5J9VCBXURbd0ZSlMPj19a
0crq5P/scvA=
-----END PGP SIGNATURE-----