Loyd Blankenship writes:

We've been kicking around the pros and cons of anonymous remailers here at io.com. One of the big problems is anonymous bombardment of a helpless newsgroup. This (and the problem of auto-screening anonymous mail) could be solved if there was a standard header keyword (or maybe even a new header field) that could be screened from a newsgroup. The group would have to be semi-moderated -- an automatic filter passes on all posts except those with the keyword in the appropriate header field.

Which reminds me of something I forgot to mention in my post yesterday about remailer policies and the properties of "ideal mixes." Remailer bombing (by volume, not the content) can of course be solved by *digital postage*, the fee charged by a remailer. As with ordinary postage, this reduces junk mail somewhat. However, digital postage (and digital money in more general forms) is not available now. A remailer could sell lists of numbers that act as postage, using reputation/trust to "honor" them. (To avoid using them for tracing, the lists could be bought from others, or traded with others, or possibly even "blinded" a la Chaum. Digital postage is sufficiently low-value (money-wise) that not as much attention to detail is needed, at least not for trial use of "Pretty Good Digital Postage.")

Words such as "anon" and "anonymous" might occur naturally in the headers. I'd propose something like "ANONYPOST" or "ANONPOST" that isn't likely to occur in nature.

Voluntary adoption of this type of standard by remailers would take away some of the ammo that the anti-anon frothers are shooting, and would go a long way toward improving the image of remailers in general.

Comments?

I think it's a good idea. Eric Messick has already proposed replacing the message names in mail with something to maker traffic analysis harder. For anonymous postings to newsgroups, a prefix system voluntarily adopted by users is another approach, e.g., "ANON: The Virtues of Anonymity." It's doubtful that all users, or all remailers for that matter, will ever adopt the same conventions for signalling an anonymous message, so the problems will persist, albeit on a different scale. Long range, a combination of pay-for-what-you-use digital postage and "positive reputation filters" will be what a. keeps newsgroups from being flooded with anonymous posts, and b. allows readers to find the messages they want to read out of a huge pool of messages. -Tim May -- .......................................................................... Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay@netcom.com | anonymous networks, digital pseudonyms, zero 408-688-5409 | knowledge, reputations, information markets, W.A.S.T.E.: Aptos, CA | black markets, collapse of governments. Higher Power: 2^756839 | Public Key: PGP and MailSafe available. Note: I put time and money into writing this posting. I hope you enjoy it.

mentor@indial1.io.com (Loyd Blankenship)

We've been kicking around the pros and cons of anonymous remailers here at io.com. One of the big problems is anonymous bombardment of a helpless newsgroup.

you are talking about a problem associated with a *mail to new gateway*. this is not the same as a *remailer*. In fact, the latter operators should not have to worry about the former.

This (and the problem of auto-screening anonymous mail) could be solved if there was a standard header keyword (or maybe even a new header field) that could be screened from a newsgroup.\

although I think the idea of anonymous identification tags in the header has `some' merit. but its an extremely problematic issue, because it could have the effect of censoring anonymous posting. the best goal is to allow the end user to make the decision, i.e. kill files, and *never* put any kind of a choke upstream that would prevent them from making that decision. hence, the solution is to have the mail-to-news gateway reject overly voluminous posting -- either posts that are too long, or too frequent posting from the same address or (the latter which of course can be thwarted to some degree) in overall frequency of accepting articles, such that some might get bounced back to the user if the site is being bombarded. of course, remailer operators have to guard against mail bombs too, but not in the overly sensitive, distributed way that NNTP servers do.