[rc5] Some thoughts on finding the key

David McNett nugget at slacker.com
Wed Oct 22 23:29:55 EDT 1997


Phil Reed was heard saying:
>According to this announcement, the key was found on October 19, but
>the announcement was not made by Bovine people until October 22. There
>are good reasons for a lag (perhaps the client that found the key
didn't
>report in right away), but three days?

I anticipated this question.  For the curious, here's a timeline of
events as they occurred:

19-Oct-1997 13:25 - Key 0x0x532B744CC20999 is submitted to the
distributed
                    net servers as a possible solution.  This packet is
                    automatically loged to a SUCCESSES log file on the
                    stats server as a part of the stats processing

20-Oct-1997 13:00 - I notice the success report in the logfile and
track
                    down Tim Charron to test it fully.  Note: at this
point,
                    we still do not know if the solution will test out,
nor
                    is this the first possible solution to appear.

20-Oct-1997 14:15 - I email the key address to Tim and cross my
fingers.

20-Oct-1997 18:14 - Tim is the first person in the world to read the
                    unencrypted plaintext.  He emails me the solution
                    text.

20-Oct-1997 18:15 - Panic ensues. Desperate attempt to reach Bovine and
                    Duncan fail.

20-Oct-1997 19:25 - With shaking hands, the solution is sent via email
to
                    RSA Labs.

20-Oct-1997 19:26 - RSA's autoresponder emails back: "You've won.
Don't
                    do anything until we contact you."

20-Oct-1997 19:27 - Key distributed.net coordinators are informed.  We
                    wait.  Eventually Duncan and Bovine turn up.

20-Oct-1997 22:45 - RSA Labs calls.  They want us to speak with their
                    press agency for drafting the RSA Labs Press
Release.

21-Oct-1997 01:10 - RSA Labs press agency calls.  Initial interview is
                    conducted.  They want to know things like "How many
                    people, how many keys did you test, how long did it
                    take, etc..."  They take a statement.  They ask us
                    to not release that the key has been found until
they
                    have submitted their press release to the newwires.
                    
22-Oct-1997 18:00 - After two major revisions, RSA Labs has finally
drafted
                    a press release that we all agree is satisfactory
and
                    accurate.  We wait for approval on press release
from
                    decisionmaker at RSA Labs.

22-Oct-1997 16:30 - Project Gutenberg is contacted.  We speak briefly
                    explaining that they may be about to receive a
deluge
                    of press inquiries and we discuss the whens and
wheres
                    of our donation.

22-Oct-1997 20:38 - RSA Labs calls.  They have just submitted the press
                    release to the newswires.  It should "hit" at about
                    21:00.  We agree that distributed.net will be safe
                    to announce the news at 21:15.

22-Oct-1997 21:15 - distributed.net publically announces that the
contest
                    is over.


>Overall, even though I have not been active on the mailing list, I've
been
>closely monitoring it since I joined the effort in late July. During
that
>time, I noticed one very strong attribute of the Bovine crew: they
have
>been very reticent about saying ANYTHING.

I'll be the first to admit that the mailing list has traditionally been
the least-effective way of communicating with the distributed.net
organizers.  This has certainly been a weakness of ours.  The sheer
volume of mail that most of us receive means that mailing list traffic
is typically the first thing to be ignored when the work to time ratio
goes critical.  Perhaps we need to find someone to be official list
moderator (much as we now have an official "help line" volunteer) to
maintain a constant presence on the mailing lists.  (volunteers
anyone?)

For good or for bad, IRC is by far the best way to contact any of us.

>And now, when we have grabbed the brass ring and won what we set out
>to do, we find a strange three day lag between the victory and the
>announcement of the victory. And, that would not be so bad by itself,
>except that rumors started appearing shortly after the stated time of
the
>discovery of the key.

Although the lag was at the request of RSA Labs, the "false alarm" was
the
result of an RSA Labs employee jumping the gun and posting the
announcment to their website.  As I understand it, the announcement was
only there for a matter of minutes and yet at least two observant
distributed.net participants spotted it and alerted the list (I, for
one, am quite impressed!  We certainly were NOT going to miss the key
with people like
this on our team!)

As for whether or not distributed.net should get your future cycles, I
am probably not the best one to speak on that.  Suffice to say that if
your future plans involve either SETI or GIMPS (Mersenne Primes) then
there is no conflict of loyalties.  Both of those are slated to be
official distributed.net projects in the very near future.  All of
which, of course, depends on Duncan and the V3 protocol spec, which he
will be explaining very soon (posibly this evening)

Hopefully this will dispell any suspicions you may have about the
sequence of events surrounding our incredible victory.  There was most
certainly no under-handedness involved.

-David McNett
 nugget at slacker.com

----
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.



More information about the rc5 mailing list