Once again, we will be holding a PGP Key signing party at the IETF-63
meeting in Paris. We have been scheduled to meet at 1930 on the evening
of Wednesday, August 3 in room 322M.  As usual, if the plenary runs over,
we will start approximately five minutes *after* the plenary session ends.

Please note the unusual time, due to the change in the meeting schedule
for this week.  Note that even though the printed agendas say 2200, we
will meet at approximately 1930, as indicated here and in the online
agenda.  In the unlikely event the agenda should conclude early, we will
begin the key-signing approximately five minutes after the plenary, in
order to allow people to take advantage of the extra time for dinner or
evening social events.

The procedure we will use is the following:

o People who wish to participate may do so in one of two ways. You may
  bring slips of paper with your name, e-mail address, key-id, and key
  fingerprint. (One way of generating this if using gpg is "gpg
  --list-keys --fingerprint my_username@hostname") You should bring
  enough for everyone who may attend; given recent attendance patterns,
  around 50 should be more than enough. (You can generally fit 10-12
  strips containing your key fingerprint on a single sheet of paper, and
  then cut out strips to hand out.)

o Alternatively, you may email an ASCII extract of their PGP public key
  to <jhutz@cmu.edu> by noon on Wednesday, August 3(*). Please include
  a subject line of "IETF PGP KEY", and please DO NOT MIME-ENCRYPT your
  e-mail. Send it to me as plain text, and do NOT base-64 encode things.
  (My process is not quite as automated as Ted's, so I'll probably be able
  to notice and fix any problems, but it's better not to take chances).

  The method of generating the ASCII extract under Unix is:

        pgp -kxa my_email_address mykey.asc (pgp 2.6.2)
        pgpk -xa my_email_address > mykey.asc (pgp 5.x)
        gpg --export -a my_email_address > mykey.asc (gpg)

  If you're using Windows or Macintosh, hopefully it will be Intuitively
  Obvious (tm) using the GUI interface how to generate an ASCII armored
  key that begins "-----BEGIN PGP PUBLIC KEY BLOCK-----".

o By 1700 on Wednesday, you will be able to fetch complete key ring
  from any of the following locations with all of the keys that were
  submitted:

        /afs/grand.central.org/project/ietf-pgp/ietf63/ietf63.pgp
        http://grand.central.org/dl/ietf-pgp/ietf63/ietf63.pgp
        ftp://grand.central.org/pub/ietf-pgp/ietf63/ietf63.pgp

o At 1930, come prepared with the PGP Key fingerprint of your PGP
  public key; we will have handouts with all of the key fingerprints of
  the keys that people have mailed in.

o In turn, readers at the front of the room will recite people's keys;
  as your key fingerprint is read, stand up, and at the end of reading
  of your PGP key fingerprint, acknowledge that the fingerprint as read
  was correct.

o Later that evening, or perhaps when you get home, you can sign the
  keys corresponding to the fingerprints which you were able to verify
  on the handout; note that it is advisable that you only sign keys of
  people when you have personal knowledge that the person who stood up
  during the reading of his/her fingerprint really is the person which
  he/she claimed to be.

o Send the signed keys to the owners, and, optionally, to the PGP key
  servers. Some poeple opt to NOT send the signed keys to the
  keyservers, but rather choose to send them only to the e-mail address
  on the key's userid, encrypted for that particular key. This tends to
  ensures the validity of the e-mail address.

Note that you don't have to have a laptop with you; if you don't have
any locally trusted computing resources during the key signing party,
you can make notes on the handout, and on the strips of papers, and then
take these and sign the keys later.

Acknowledgement: The bulk of the text of this message was taken from the
messages usually sent by Ted Ts'o to announce IETF key signing parties.

(*) Normally I'm pretty lax about the "noon" deadline, accepting things
as late as during the plenary.  However, in order to keep things running
on time with the unusual schedule of this week's meeting, I intend to
prepare the photocopied fingerprint lists before the plenary session.
So, I'll be applying the deadline more strictly than usual.

-- Jeffrey T. Hutzelman (N3NHS) <jhutz+@cmu.edu>
   Sr. Research Systems Programmer
   School of Computer Science - Research Computing Facility
   Carnegie Mellon University - Pittsburgh, PA