Responding to Phil Karn's proposal that anonymous mail should be clearly marked as such: The only thing I object to in this is that it implicitly gives up one of the strongest arguments in favor of anonymity/pseudonymity, which is that there is already no way to verify identities on the net. In Phil's analogy with Caller ID, where the net is said to already be a "Caller ID" environment, the thing to realize is that the "ID's" are not necessarily accurate. To a large extent, identity on the net is an illusion. Pseudonymous remailers like Julf's should be seen as a natural extension of net culture. This situation is only going to become more extreme as the net continues to move beyond its original, relatively controlled, community of large universities and government research labs, to include the general public. As more and more "Free Net", Public Access, and BBS systems become part of the internet, there is either going to have to be a massive and universal crackdown on identity verification, which I think is unlikely, or else there is going to have to be acceptance that net identity doesn't necessarily correspond to real names. (I myself have had email-capable accounts in three names besides my own on various systems.) Granted, this argument did not persuade the facist forces which forced the shutdown of Julf's remailer, but that doesn't invalidate it. Julf's remailer was not shut down due to reasoned disagreement and a consensus that it was wrong; rather, its shutdown was (as far as is known so far) a demonstration of raw power by some person or small group. Responding to Marc Horowitz's point about problems with the idea of uncancellable messages: Marc raises the issue of volume abuse. I think it's important to note that, as far as I know, the Penet postings which people complained about where NOT examples of volume abuse. The objections to these messages were based on their contents. (In some cases, people objected to some messages not because of their contents or their volume, but simply because they were "anonymous"!) I realize that Marc was addressing the issue of uncancellable messages in general, not specifically with regard to anonymous messages. I am not an expert on news software but reading the debate on this issue in news.admin.policy it appears that the current system is far too lax in allowing cancel messages. It appears to be very easy to cancel postings made by someone else. This led to De Pew and his cancel daemon, which itself led to counter-threats for cancel daemons to be activated against De Pew and other posters from his site. All this points to design flaws in the cancel mechanism. I do think that it would be appropriate to put more restrictions on cancel messages, and digital signatures could play a part in this. Perhaps Marc's concern about payment for volume abuse could be dealt with by some limitations on large postings. I don't really know how Marc (or anyone) distinguishes between a 100K byte junk file in sci.crypt and a 100K byte file in alt.graphics.misc which he might find equally uninteresting and for which he has to pay equally. Maybe he's only reading sci.crypt? In that case perhaps a solution would be for the news transfer software to be enhanced to allow some filtering. Hal 74076.1041@compuserve.com P.S. Edgar asks about the mail-to-news gateway at ucbvax.berkeley.edu. Recently I saw postings indicating that this gateway had shut down. If anyone has information on mail-to-news gateways please post it here.