CDR: Re: Re: -C-P- Re: would it be so much to ask..

Asymmetric all at biosys.net
Wed Sep 20 06:36:47 PDT 2000


At 02:21 09/20/2000 -0400, Riad S. Wahby wrote:


>You should read up on Type 1 and Type 2 remailers.  Both involve
>encryption.  In the case of Type 2 remailers, you only need to trust
>one in the chain that you use in order to be sure that your identity
>is securely hidden.

I do understand how both types work, however, the opportunity for 
subterfuge is always present. I was making a point that the assumed 
security of a remailer should not factor in if you intend to put yourself 
at risk.  Assumptions are dangerous all over the place, and if your 
assumption could get you into trouble, it's better to verify it or not 
instead of just proceeding blindly, if at all possible.


>Wrong again.  By default in versions of sendmail since 4.9, all sent
>mails are logged right along with the failures--and this includes the
>IP address from which the connection was made to the SMTP server.
>Simply setting your SMTP server is not nearly enough.  If 'they' have
>the IP address from which the mail was sent, 'they' have you.  As I
>said above, please read up on Type 1 and Type 2 remailers before
>making such outrageous claims.

What is the outrageous claim?  That someone could purposely set up an 
insecure remailer, claim that it's secure, and that people could then 
unwittingly use it to incriminate themselves?

As for Sendmail, you are correct; Sendmail defaults to a log level of nine, 
but it's utterly admin configurable, and with successful messages logs the 
delivery as successful.  This setting is also only present in the default 
configuration if you get the tarball from sendmail.  Options vary by OS 
vendor, and loglevels may be different (by default) on a newly installed OS 
that ships with sendmail (most unix variants, if not all, do) or if 
sendmail was installed via a package distribution method.

This could go on and on, but it's twisting the point I was making a great 
deal and turning the argument away from the topic that started it.  I'm not 
debating the relative merits of remailers, I know they serve a need that 
cannot be duplicated with great ease, and honestly my comment about faking 
the return addresses was in no way to say that this method could replace 
remailers; I was just pointing out that it is a far stretch of the 
imagination to call it "port 25 hacking" or whatever Tim said.

I'm going to stop myself here.. The original email I sent was simply a 
question about what could be done to possibly quell the flow of spam 
generated by this list, followed by a few suggestions.  It's gotten utterly 
out of hand because somebody apparently took it as a personal attack, and 
responded with a series of attacks of his own.

As I said before, if [you] (the reader on the list) don't want to hear 
these questions, then what's good for the goose is good for the 
gander.  Filter my emails if you like; It'll be a lot easier for you to 
filter them automatically than it will be to filter the spam messages in 
any event.


>Finding open relays that don't do logging is difficult at best.

I agree they are not easier to find than just trusting the word of someone 
that a particular remailer is secure, but in my opinion that is ass 
backwards.  Trust should be a lot harder to earn than simply doing a little 
legwork.  You suggested finding a remailer in a country unfriendly with the 
one that is likely to come after you for posting whatever you wish to post, 
so that it will be harder for law enforcement to pry any information out of 
the hands of the remailer operator; I find the same logic applies to doing 
the header forging.  If you're in the US as in your example, using an smtp 
server in Iraq (for example) to send your email through is a pretty safe 
bet that even if it logs every line of the file including the message body, 
that the chances of them cooperating one iota with the authorities is 
pretty small.


>As I said above, in the case of the Type 2 remailer, you only have to
>trust one server in the chain, and presumably you can find one that
>you're likely to trust not to disclose information to the people from
>whom you want to hide your identity.  In the case of a US national,
>for example, post through a remailer in a country that the US doesn't
>like much--there are plenty of those--and you're fine.  That, or trust
>that, for example, the MIT LCS remailer is reasonably secure (and it
>is--I know the person who runs it), and make sure it's in your chain.

Just as a preamble here, I'll say flat-out that I totally understand the 
need some people have for anonymous remailers.

That said, I have not personally had occasion to use one yet, or need 
to.  I have been in situations where anonymity would have possibly been 
desirable to some, but more often then not I have chosen to simply waive 
any kind of real anonymity, and just get out there with what I was doing.

I used to run a very large site critical of the CoS, and I kept 
-everything- online.  My domain name records were forged, and I didn't go 
out of my way to attach my actual name to anything, but I didn't go to any 
great lengths to hide it either.  When the CoS found out about it, they 
sent Ms. Kobrin after me.  She claimed she wanted to send me a hardcopy of 
their copyrights on the material that I had posted, because I told her if 
she could prove ownership of any of it, that I would take it down.  I told 
her that an email copy of the information would be sufficient, and that she 
should send it straight away.  I even took the material down and gave them 
two weeks to produce.  When they didn't, I emailed them and put the 
material back online.  This part repeated, and I repeated my request.

Instead, I received another nasty letter, and then a few days later a call 
from my upstream provider.  They denied her request to deliver up my name 
and address, but told me that if I didn't take the information down that it 
would be a violation of the service agreement, and that they would 
disconnect the frame relay.  I explained to their (the ISPs) lawyer the 
situation, and that they did not actually own the copyrights in 
question.  He responded with "I know, but considering their history, they 
are very willing to take us to court over this, and honestly we don't want 
to deal with that.  Take it down or we shut you down, we don't care who's 
right, we just don't want a lawsuit."

Needless to say, faced with the entire site being removed, I removed the 
materials.


> > very good chance at hitting something
>
>Again, I ask you to produce an example of an open relay that you are
>reasonably sure does not do logging.

After my diatribe above you ask me to find a server that I *trust* is not 
doing logging?  In that case I'll trust only those that I admin, so that 
even in the case that they are doing logging, I can remove the logs myself 
afterwards.

Finding an open relay first off is easy though.. www.orbs.org.  Finding one 
that doesn't log, difficult to verify logging or not, so you just look for 
one run by an entity unlikely to cooperate, as we covered before.


>So please filter, and don't complain.  Or unsubscribe.  It's the
>responsibility of new readers to peruse the archives.  If you had done
>so, you would not have angered those who have heard this argument 10^9
>times.

THAT is exactly what I'm talking about.  I wasn't complaining, at least not 
as loudly as some of the rest.  I was trying to get something -done- and 
there is a difference.

As far as I'm concerned, if people don't want to hear this again, then THEY 
can filter or unsubscribe.  I personally like to believe any amount of 
discussion on this list is more meaningful than the spam, even if it's all 
been said and done before.

It is an open list after all right?


>No.  The people of the list expect that you have gone over the
>archives so that what you say is not repetitive and a waste of time
>and bandwidth.  If a bit of time and bandwidth spent now can reinforce
>the practice of archive reading before you post, then it is well
>spent, and is, in the long run, a net savings of both bandwidth and
>time.

I see.  But trying to find a way to save even more time and bandwidth by 
even attempting to figure out a solution to this problem is not as valuable?

It comes down to a simple bit of confusion on my part.  I cannot understand 
the mentality of someone who has the time and resources to effectively 
combat the spam on this list, and yet who does not have the time or 
resource to either respond in a somewhat civil fashion, or to just delete 
the message along with the rest of the refuse.

You seemed to be a bit more level headed, so while I still totally disagree 
that it's a waste of time to try and figure a way around this problem, I 
haven't utterly lost respect for you as I have with Tim.  "Pillar of the 
community" or not, the guy is an utter asshole.


-------signature file-------
PGP Key Fingerprint:
446B 7718 B219 9F1E 43DD  8E4A 6BE9 D739 CCC5 7FD7

"I don't think [Linux] will be very successful in the long run."
"My experience and some of my friends' experience is that Linux is quite 
unreliable. Microsoft is really unreliable but Linux is worse."
-Ken Thompson, Interview May 1999.

http://www.freebsd.org
FreeBSD - The Power to Serve

http://www.rfnj.org
Radio Free New Jersey - 395 streams - 96kbps @ 44.1khz





More information about the cypherpunks-legacy mailing list