December 2003 - cypherpunks-legacy

Premail and transparent email
by raph＠netcom.com 17 Dec '03

17 Dec '03

It's quite gratifying to see that people are actually using premail and like it. I see premail as a prototype for _real_ transparent email encryption. A lot of people are intimidated by the need to get premail off the ftp site, unzip/untar it, and set all the configuration variables to get it running right. This "intimidation factor," of course, only applies to *x people. Everybody else is completely out of luck. I think the same problems hold with most of the scripts that are out there. Every time I've gotten something to play with, I've had to diddle with pathnames, or the makefile, or whatever. The real solution, I think, is to get all the needed components for transparent email encryption into the standard releases of the tools. I'm currently working on exactly this project. In rough outline, PGP will run as a "server" process. Mailers would connect to the server, and pass all incoming and outgoing mail through it. One advantage is that clients would contain _no_ crypto content, so there would be no problems with exportability. The server would contain much of the functionality of premail. I showed an early prototype at the last cpunks meeting. Initially, I am doing all the work in *x, just because that's what I have tools for, but ultimately it should work for Windows and Mac as well. My intent is to get large numbers of people to use PGP to encrypt all of their email, including casual stuff. This won't happen until encryption and decryption are _totally_ transparent. -- Raph Levien -- raph(a)netcom.com

2 1

premail 0.30
by Michael K. Sanders 17 Dec '03

17 Dec '03

I've got premail v0.30 installed here as /usr/lib/sendmail. I added the $config{"sendmail"}="real_sendmail" option, but this means that ~/.premailrc is not used. But that users can't specify their own options... Specifically, it looks as if doing it this way I can't automatically PGP sign my mail without adding it to the global script, but that would mean _all_ mail out of this site would be signed with my id. Any suggestions how to go about this? It makes it so convenient to have it replace sendmail, but I'd like to sign my mail as well. -- Michael K. Sanders -- msanders(a)ataxia.res.wpi.edu ataxia: NetBSD/Amiga 1.0 - Creating Chaos out of Anarchy for a Better Tomorrow Ataxia Home Page

2 2

Re: The Market for Crypto--A Curmudgeon's View (fwd)
by Jonathon Fletcher 17 Dec '03

17 Dec '03

On Tue, 29 Nov 1994, Sandy Sandfort wrote: [ ... stuff r'moved ... ] > > Personally, I'm against mandatory digitally signatures as a > rerequisite for posting to Cypherpunks. On the other hand, I > like the idea of having the list software automatic verify > digital signatures. This is a valuable service I'm usually too > lazy to perform for myself. > > Here's my suggestion. Eric should unilaterally impose his first > step, i.e., all unsigned messages and messages with spoofed > signatures will henceforth be flagged as such. Let's see what > effect, if any, that has on the way people post their messages. > After the protocol has been in effect for some time, we can > re-open the topic for further discussion. > This is a good idea - certainly a nice way to emphasize (sp) signing posts. Having majordomo verify signatures automatically and add message content if either (as sandy suggests) signature is missing or bad. One question would be whether majordomo should add content upon verification of a signature, or upon failure to verify a signature (missing or bad). Depends on which would have most 'positive' appearance, and (more importantly) would generate least extra volume in the long term. Personally I think that flagging the messages with bad or missing signatures is a better idea - as more people sign articles on the list there will be less flagging volume sent out by md. Nice idea Sandy. ... only snag is that md has to be modified again. Has Eric the time and desire to fit this in. I can't help with the perl - still not grokked it properly. -Jon -- Jonathon Fletcher j.fletcher(a)stirling.ac.uk " .. all opinions expressed or implied are my own and not necessarily those of my employer or any other party ... "

1 0

Re: Hazards of encouraging forged dig sigs
by Mark Grant 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- On Wed, 30 Nov 1994, L. McCarthy wrote: >Encouraging cryptographically >valid signatures was the first suggestion I'd seen in this entire debate >which seemed to promise tangible benefits; encouraging cryptographically >invalid signatures is the first notion which appears to offer tangible >detriment. As one of the few people who verify all PGP signatures for which I have keys, I have to say that encouraging people to put garbage in PGP signature blocks would be extremely annoying. Either I'll have to go back to using a non-PGP-aware mail reader, or fix it to do something sensible with such messages. While I'm going to have to do the latter at some point anyway, there are much more useful things that I could be doing... Mark -----BEGIN PGP SIGNATURE----- Version: 2.6 iQCVAgUBLt3Kw6FlWzerDvH1AQEyTgP/bPz1NLXeicLD81Ijj8S6/m+D3Tmb363Q +Xj5MDsW+xBNC+p8ViKzZ+zvjZg5Cr6MsQ24TWT3n7DId7l2pAWyAH5JWf7OuW9u xvrfMfFas04RuOE8mjO7Gcp9E3eFNKz9MNQGlG7YwkPLX/twbF7+gX1RWkPGV2JE +b+KOjVkQbs= =lLbP -----END PGP SIGNATURE-----

1 0

where to get secure edit a0.3.3 for macintosh
by Thomas J. Bryce 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- I've received a number of inquiries about where to obtain secure edit a0.3.3 for the macintosh. Please finger me at tjbryce(a)amherst.edu if you want information on this. Thank you Tom tjb(a)acpub.duke.edu -----BEGIN PGP SIGNATURE----- Version: 2.6 iQCVAwUBLtuTJk8YjrUhOUC5AQEoAAP/W+kl5cOkuohw5QtafawUFA5kKSrhhv/o rE8mzGVsntNJ52NcCS7ImONTKxH+rHmcrbhYi8A0rVAaf0byYDSZ50PHCqoM4WPc ccNQ4zMO75N38uZ8/pLO9w3nYw5Y386737IND8QOjHmLc/jUQlJqdFuGzxtYq2r1 RzQgKu9xjS4= =m0X4 -----END PGP SIGNATURE-----

1 0

Re: premail 0.30
by Alex Strasheim 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- I've also installed premail as /usr/lib/sendmail, and it works fine. I haven't had any problems with it reading my ~/.premailrc file. I was a little queasy about installing it as sendmail, but I'm very glad I did. It hasn't caused any problems at all, adds a lot of functionality, and doesn't seem to extract too high of a penalty in terms of overhead. I'm not sure how sendmail forks for incoming mail, but my impression is that the additional instance of sendmail is created without calling premail. If that's true, then premail installed as sendmail doesn't add any overhead at all for incoming mail. This was a big concern for me, because I don't have very much memory or processor power, but so far everything's been working great. == Alex Strasheim | finger astrashe(a)nyx.cs.du.edu alex(a)omaha.com | for my PGP 2.6.1. public key -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAwUBLtvvxREpP7+baaPtAQFq5wQAzT295CbJIQXJU8chI4Cwm9rHGa4mbqNV +geDhS02z/ttYjMIoXRh1066YmnSEB3/3uqQPL4K1w8SP3z7uYfQwDwKwbmRdoPa p03ksQ8PEK/9dbRc05azjz4PFpMTqk4HDhyQnM3xRo34DPNbnfl4HUSMbccvdCEw vS4Vip00zX0= =+du+ -----END PGP SIGNATURE-----

1 0

Elm and premail (long)
by John A. Perry 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- I have just finished integrating premail 0.30 with Elm on jpunix.com. Since there have been some questions about how to do this type of thing, I thought I'd give a blow-by-blow description of what I did and the pro's, con's and alternatives to what I did. 1) I installed premail as a wedge to the sendmail daemon. In other words, per Raph Levin's instructions in his README file, I renamed sendmail to something else and then renamed premail to sendmail and have it pointing to the renamed sendmail. I realize this requires system admin privs and is a big step to take, but I have testing premail working as the sendmail program with great success and at this point I have no qualms about premail/sendmail being one in the same. 2) I created the ~/.premailrc file in my home directory per Raph's README file. There is an option at the bottom to create aliases and that is what I did at first. Then, I discovered that the alias database created by Elm is compatible with premail!! 3) I edited ~/.elm/aliases.text to make changes similar to the following: The alias raph = Levin; Raph = raph(a)kiwi.cs.berkeley.edu to raph = Levin; Raph = raph@kiwi.cs.berkeley.edu^key=raph@cs.berkeley.edu Elm doesn't seem to mind this new alias structure and premail parses it just fine! I could have added a ^chain=3 also, but I decided that I may not want to email Raph through a set of remailers every time. I decided I would add the extra header during message creation. More on this later. What this does is that from now on, whenever I send email to Raph, it will automatically be encrypted with his public key with no further intervention on my part! When I want to chain the message through some remailers, I just take the H)eaders option from the Elm pre-send menu and add a header line: Chain: 3 This causes the message to be chaine through three random remailers before it gets to Raph. I also added two addition Perl scripts to Elm to help in the creation of encrypted messages for people NOT in my alias file, signing messages, and reading PGP encrypted messages that are not MIME compliant. These scripts are mailpgp and morepgp. These Perl scripts were published with version 2.3 of PGP but work fine with 2.4 and up. In Elm, I went to the options menu and replaced the default editor with mailpgp as well as the visual editor. I also replaced the default builtin+ display parameter with morepgp. These two Perl scripts are listed at the end of this message. The mailpgp script will allow you to encrypt and sign messages right before they are sent. The morepgp script will decrypt/verify incoming messages on the fly. The really nice benefit of these scripts is that you can reply to cleartext version of PGP encrypted messages. With the combination of these Perl scripts, Elm (or Pine for that matter) can be turned into an encrypted mail handler with a minimum of fuss. morepgp: #!/usr/local/bin/perl # -*- Perl -*- # written by jason steiner, jsteiner(a)anwsun.phya.utoledo.edu, Jan 1993 # # if you use or make improvements to this program i would appreciate # copies of your modifications & your PGP public key. # # Modified by Greg Spencer, greg(a)graphics.cornell.edu, May 1994 # Mostly just cleaned up things and added stuff like automatic # addition and detection (and ignoring) of keys to be added to keyring, # and signal catching, as well as environment variable control of # most site-specific stuff. # # Must set the following environment variable: # # PGPCOMMAND set to the pgp decryption command # # PAGER set to the desired pager command # # NOTE that this program NEVER writes sensitive data to a disk file. # it will only slurp it into memory, so if you have a HUGE file, you might # have problems. # setup some variables ($pgpcommand = $ENV{'PGPCOMMAND'}) || ($pgpcommand = "/usr/local/bin/pgp"); # just used for tmpfile names... ($logname = $ENV{'LOGNAME'}) || ($logname = "nobody"); # ($pager = $ENV{'PAGER'}) || ($pager="/usr/local/bin/less -i -n -s -S -c -M"); ($pager = $ENV{'PAGER'}) || ($pager="/usr/bin/more -c"); $|=1; $topgp = 0; $tokey = 0; $pgpused = 0; ($tmpdir = $ENV{'TMPDIR'}) || ($tmpdir = "/tmp"); #temporary file name $tmpfile = "${tmpdir}/.pgp1.$logname.$$"; $tmpfile2 = "${tmpdir}/.pgp2.$logname.$$"; # trap signals so we do not leave # garbage around sub catcher { local ($sig) = @_; print "Caught a SIG$sig -- exiting\n"; close (TMPFILE); close (OUTPUT); close (PAGER); unlink ($tmpfile); unlink ($tmpfile2); } $SIG{'INT'} = 'catcher'; $SIG{'QUIT'} = 'catcher'; $SIG{'HUP'} = 'catcher'; $SIG{'KILL'} = 'catcher'; # make sure nobody can read stuff umask 077; # prepare a data area @tmpdata = (); @newkeys = (); while (<>) { if (!$topgp && m/^-----BEGIN PGP .*-----/ && !m/^-----BEGIN PGP PUBLIC KEY BLOCK-----/) { $topgp = 1; $pgpused = 1; unlink ($tmpfile); open (TMPFILE, ">$tmpfile") || (unlink ($tmpfile) && die "Cannot open $tmpfile for output.\n"); } if (!$topgp) { push(@tmpdata, $_); } if ((!$tokey) && (m/^-----BEGIN PGP PUBLIC KEY BLOCK-----/)) { $contains_keys = 1; $tokey = 1; } if ($tokey) { push (@newkeys, $_); if (m/^-----END PGP PUBLIC KEY BLOCK-----/) { $tokey = 0; } } if ($topgp) { print TMPFILE $_; # OK to write this to a file -- it is encrypted! if (m/^-----END PGP .*-----/ && !m/^-----END PGP PUBLIC KEY BLOCK-----/) { $topgp = 0; close TMPFILE; open (CLEAR, "$pgpcommand -f < $tmpfile |") || (unlink($tmpfile) && die "Cannot open pipe to PGP.\n"); $blocktype = $_; $blocktype =~s/^-----END (PGP .*)-----/$1/; $blocktype =~s/PGP MESSAGE/DECRYPTED MESSAGE/; $blocktype =~s/PGP SIGNATURE/SIGNED MESSAGE/; chop ($blocktype); push (@tmpdata, "-----BEGIN $blocktype-----\n"); while (<CLEAR>) { push (@tmpdata, $_); if ((!$tokey) && (m/^-----BEGIN PGP PUBLIC KEY BLOCK-----/)) { $contains_keys = 1; $tokey = 1; } if ($tokey) { push (@newkeys, $_); if (m/^-----END PGP PUBLIC KEY BLOCK-----/) { $tokey = 0; } } } close CLEAR; print STDERR "\n"; unlink ($tmpfile); push (@tmpdata, "-----END $blocktype-----\n"); } } } select (STDIN); $|=1; select (STDERR); $|=1; select (STDOUT); $|=1; # This handles things if we found keys that need # adding to our keyring # note that we are only writing the KEYS to the file. if ($contains_keys) { print STDERR "PGP Keys found, attempting to add...\n"; open (TMPFILE2, ">$tmpfile2"); foreach $_ (@newkeys) { print TMPFILE2; } close (TMPFILE2); # strange things happen if we do not # read/write directly from /dev/tty (perl bug??) system ("$pgpcommand -ka $tmpfile2 >/dev/tty </dev/tty 2>&1"); unlink ($tmpfile2); # get rid of it asap $pgpused = 1; } # copy the contents of @tmpdata to the pager we want to use. open (PAGER, "|$pager") || (unlink ($tmpfile1) && die "Cannot open pipe to $pager.\n"); # do "press any key to continue" # only if we had some output from PGP # (like a verified signature) # again with the /dev/tty thing (weird!) if ($pgpused) { $q=''; open (TTY, "</dev/tty") || (unlink ($tmpfile1) && die "Couldn't open /dev/tty\n"); select(TTY); $| = 1; print STDERR "Press any key to continue...\n"; $q = getc (TTY); select (STDIN); } # output the decoded thing to the pager foreach $_ (@tmpdata) { print PAGER; } @tmpdata = (); close OUTPUT; close PAGER; # cleanup unlink ($tmpfile); mailpgp: #!/usr/local/bin/perl # -*- Perl -*- # written by jason steiner, jsteiner(a)anwsun.phya.utoledo.edu, Jan 1993 # # if you use or make improvements to this program i would appreciate # copies of your modifications & your PGP public key. # # Modified by Greg Spencer, greg(a)graphics.cornell.edu, May 1994 # Mostly just cleaned up things and added stuff like automatic # addition and detection (and ignoring) of keys to be added to keyring, # and signal catching, as well as environment variable control of # most site-specific stuff. # # Must set the following environment variables: # # EDITOR or VISUAL set to editor of choice # # PGPCOMMAND set to the pgp decryption command # # PGPID or SIGNATURE set to the id you wish to # have used for your pgp signatures, etc. # # PREFIX set to the forwarded message prefix that you use # # setup some variables ($visual = $ENV{'EDITOR'}) || ($visual = $ENV{'VISUAL'}) || ($visual = 'pico'); ($pgpcommand = $ENV{'PGPCOMMAND'}) || ($pgpcommand = "/usr/local/bin/pgp"); #($myname = $ENV{'PGPID'}) || ($myname = $ENV{'SIGNATURE'}); ($prefix = $ENV{'PREFIX'}) || ($prefix = "> "); $topgp = 0; $blanks = 0; $paragraphs = 1; $blankcompress = 1; $name=@ARGV[$#ARGV]; #temporary file names $rplyfile = "$name.rply"; $pgpfile = "$name.pgp"; $ascfile = "$name.asc"; $clrfile = "$name.clr"; # trap signals so we don't leave # (possibly sensitive!) garbage around sub catcher { local($sig) = @_; print "Caught a SIG$sig -- exiting\n"; close (OUTPUT); close (PGPFILE); close (CLEAR); close (INPUT); unlink ("$pgpfile"); unlink ("$rplyfile"); unlink ("$clrfile"); unlink ("$ascfile"); } $SIG{'INT'} = 'catcher'; $SIG{'QUIT'} = 'catcher'; $SIG{'HUP'} = 'catcher'; $SIG{'KILL'} = 'catcher'; # parse the input file to see if we're replying to an encrypted message # user may need to type in pass phrase to decode umask (077); open (INPUT, "<$name"); open (OUTPUT, ">$rplyfile") || die "Cannot open $rplyfile for output.\n"; while (<INPUT>) { # make sure to allow printing of key blocks if (!$topgp && (!m/^$prefix-----BEGIN PGP .*-----/ || m/^$prefix-----BEGIN PGP PUBLIC KEY BLOCK-----/)) { if (m/^$prefix*$/) { if ($paragraphs) { if ($blankcompress) { if ($blanks == 0) { print OUTPUT "\n"; $blanks = 1; } } else { print OUTPUT "\n"; } } else { print OUTPUT; } } elsif (m/^[ \t\r]*$/) { if ($blankcompress) { if ($blanks == 0) { print OUTPUT "\n"; $blanks = 1; } } else { print OUTPUT; } } else { print OUTPUT; if ($. == 1 && !m/^$prefix/) { print OUTPUT "\n"; $blanks = 1; } else { $blanks = 0; } } } # make sure to skip key blocks because we already did 'em in morepgp if (!$topgp && m/^$prefix-----BEGIN PGP .*-----/ && !m/^$prefix-----BEGIN PGP PUBLIC KEY BLOCK-----/ ) { $topgp = 1; unlink ($pgpfile); open (PGPFILE, ">$pgpfile") || die "Cannot open $pgpfile for output.\n"; } if ($topgp) { $_ =~ s/^$prefix//; print PGPFILE $_; # make sure to skip key blocks because we already did 'em in morepgp if (m/^-----END PGP .*-----/ && !m/^-----END PGP PUBLIC KEY BLOCK-----/) { $blocktype = $_; $blocktype =~ s/^-----END (PGP .*)-----/$1/; $blocktype =~ s/PGP MESSAGE/DECRYPTED MESSAGE/; $blocktype =~ s/PGP SIGNATURE/SIGNED MESSAGE/; chop ($blocktype); $topgp = 0; close (PGPFILE); system ("$pgpcommand $pgpfile -o $clrfile > /dev/tty 2>&1"); open (CLEAR, "<$clrfile") || die "Cannot open $clrfile for input.\n"; print OUTPUT "$prefix-----BEGIN $blocktype-----\n> \n"; $blanks = 0; while (<CLEAR>) { if (m/^[ \t\r]*$/) { if ($paragraphs) { if ($blankcompress) { if ($blanks == 0) { print OUTPUT "\n"; $blanks = 1; } } else { print OUTPUT "\n"; } } else { print OUTPUT "$prefix\n"; } } else { print OUTPUT "$prefix"; print OUTPUT; $blanks = 0; } } close (CLEAR); unlink ($clrfile); unlink ($pgpfile); print OUTPUT "$prefix-----END $blocktype-----\n\n"; } } } close OUTPUT; close INPUT; unlink ($name); rename ("$rplyfile", "$name"); system ($visual, @ARGV); while (!$q) { print "\nSign this message? [Y]: "; $q = <STDIN>; $q =~ s/[ \t\n]//g; $q = substr ($q, 0, 1); if (($q eq 'Y') || ($q eq 'y') || ($q eq '')) { push (@opts, '-st', '+clearsig=on'); $q = "y"; } elsif (($q ne 'N') && ($q ne 'n')) { $q = ''; } } # note that it is the default to NOT encrypt, # simply because not everyone has PGP (unfortunately :-) $q=''; while (!$q) { print "Encrypt this message? [N]: "; $q = <STDIN>; $q =~ s/[ \t\n]//g; $q = substr ($q, 0, 1); if (($q eq 'Y') || ($q eq 'y')) { push (@opts, '-e'); $q = "y"; } elsif (($q eq 'N') || ($q eq 'n') || ($q eq '')) { $q = "n"; } else { $q = ''; } } if (@opts) { if ($q eq 'y') { print "Enter receipients, each on a separate line, terminate with EOF or a single `.':\n"; { print "> "; if ($_ = <STDIN>) { chop; last if ("$_" eq '.'); push (@receipients, "$_"); redo; } last; } } system ($pgpcommand, '-a', @opts, "$name", @receipients); if ($? == 0) { unlink ($name); rename ("$ascfile", "$name"); } } - -- John A. Perry - KG5RG - perry(a)jpunix.com WWW - http://jpunix.com PGP 2.62 key for perry(a)jpunix.com is on the keyservers. PGP-encrypted e-mail welcome! Finger kserver(a)jpunix.com for PGP keyserver help. Finger remailer(a)jpunix.com for remailer help. -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAwUBLtvwZ1OTpEThrthvAQFwlAP+MxX6olbMempfh6UYdTDGruTngH+WgRsa BacTB86oNIjlllDfZB55KJyuUs5dpP+gRRDW4BZTK6zyNuzy3tv5iErQnvDiV/Tn PjSKmJJFs7HnC88aC830eQ+ojGaXzZCE2IbaTm/a7R6SU9nLc/KnJYY5pMHjdVx+ uN0xwmRrBYw= =SUyb -----END PGP SIGNATURE-----

1 0

Security Services
by Ian Farquhar 17 Dec '03

17 Dec '03

>From The Australian, Tue 29th November: Secret Service Files Lost in Blaze, by Cameron Steward (Foreign Affairs Writer) "Hundreds of top-secret documents relevant to the Federal Government's inquiry into the Australian Secret Intelligence Service were destroyed by a fire that swept through the ASIS headquarters in Canberra at the weekend." "The fire has erased highly sensitive ASIS files and archives that detailed the activities and operations of the troubled spy agency over the past decade." "The blaze is a big blow to the Government's inquiry into accountability and management of ASIS because it destroyed many of the records needed to adequately assess the organisation's performance." "But despite the suspicious timing of the accident, the Government does not believe it was arson and says that the blaze appears to have been sparked by an electrical fault." "'I can confirm that the fire broke out at 2am on Saturday morning on the fourth floor of the Department of Foreign Affairs and Trade Building,' a DFAT spokeswoman said." "'The damage appears to have been fairly extensive and is believed to have been caused by an electrical fault.'" "Sources yesterday said the area damaged by the blzae was far greater than was originally revealed by the authorities, with the initial damage estimate of about $200,000 ballooning to at least $1 million." "The fourth floor of the DFAT building has housed ASIS for years." "One intelligence source said last night that it was 'highly coincidental' for ASIS to experience such a fire in the midst of a Government inquiry into its operations." "It was rumored yesterday that the fire also destroyed thousands of files that the service was alleged to have kept on Australian citizens." "Two former ASIS officers alleged in March that the organisation illegally held files on thousands of Australian citizens, despite the fact that it is not part of its role involving gaining foreign intelligence." "The former agents also claimed ASIS has bribed politicans in Malaysia and had helped British Intelligence obtain confidential information that harmed Australia's interests." (Small note here which was not in the article: the ASIS jargon for MI6 is 'head office'. Many of the officers were trained by MI6.) "The allegations by the former officers prompted the Government to conduct an inquiry into the control and accountability of the organisation." "The inquiry, headed by former NSW Supreme Court judge Mr Gordon Samuels, is also examining the 'protection of ASIS intelligence sources and methods' as well as the 'resolution of grievances and complaints relating to ASIS." "The inquiry has been in progress for several months. It is still taking evidence and is due to report to the Government early in the New Year." This article was transcribed without permission from the paper, for the purposes of research and study as defined in the "fair use" provisions of the Copyright act. ------- Yet another depressing but compelling reminder that our privacy has as many enemies within the Government as without. BTW, this was the same TLA which I mentioned some weeks ago in connection with the bungled raid on the Sheraton Wentworth Hotel. Ian. #include <std.disclaimer>

2 2

Mandatory sigs
by Mark Rogaski 17 Dec '03

17 Dec '03

-----BEGIN PGP SIGNED MESSAGE----- I would have to agree w/ not REQUIRING digsigs, but flagging posts w/out them. I guess that's my write in. doc -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQB1AwUBLtv1T3sYwuXlZ+UZAQGXvgL/c21Yws0KUdzea12xExOszs1ZUuVUOKJX O7QtKbX96mZsh6lDUH+QS4uoNetqYdwwmbvW9PWcCkyksbzmjJpQOWzXwaDTw87S duLvkPNd+obUHUrawzUU+d0TNl/kXQIr =5DO0 -----END PGP SIGNATURE-----

1 0

Re: Mandatory message signing
by jrochkin＠cs.oberlin.edu 17 Dec '03

17 Dec '03

At 9:59 PM 11/29/94, L. McCarthy wrote: >Hold the phone ! As I understand the law, only one party to a telephone >call has to be aware of the recording for it to be perfectly legal. Someone >not party to the call can't do it, but any one of the people talking can do >it. I believe it varies from state to state, but this is indeed the norm. >handle authentication like this. Ideally, some people would get together >with Brent Chapman and incorporate authentication of signed messages in a >future release of Majordomo. I'd love to volunteer for a project like this >but I don't believe I can spare the time. Why stop at authenticating signed messages? I'd like to see some mailing list software that would send mail out to you encrypted if you want (sure, the list is probably public, but you might not want people knowing you subscribe. If list traffic was encrypted, and you routed it through Ghio's remailer-alias doohickey, the fact that you subscribe could be kept completely hidden), and would require authentication on a message to change your address, or to change your public key (which would also be possible). And would have config parameters such that the list operator could make it only accept signed submissions, or even only accept submissions encrypted to the list. [not that I'm advocating this on the cypherpunks list, but it might be good on other lists, especially private ones]. And Ideally, I'd even like the list to be able to mail to you through the remailer-net, by prepending your encrypted address block, and sending to a remailer. Although the existence of Ghio's forementioned doohickey (what is that service being called?) makes that somewhat superflous. I was planning on writing such a beast myself, and might still get around to it eventually. The problem is that I'm unlikely to write something from scratch as robust as the current list server software (I'm not interested in writing good list software, just in implementing the crypto), and I'm not really skilled enough to understand the majordomo code enough to modify it. But I might try, one of these days.

1 0