signing messages

-----BEGIN PGP SIGNED MESSAGE-----


Seems like one way to encourage the use of digital signatures is to
start forging messages from people who don't ordinarily sign their
messages. Necessity is the mother of invention, and all of that.

I finally started signing my messages on a regular basis as a result
of Detweiler forging a message which purported to be from me. On the
other hand, I think Tim has been the most frequent target of 
Detweiler's forgeries, and I don't detect much of a creep towards
signing messages on his part. 

Eric, would you mind clarifying the purpose of the "sign-or-delay"
rule? Last time this came up I assumed that it was to encourage 
folks who had 95% of the tools/initiative to start using crypto 
techniques on a day-to-day basis to get off their asses and do so;
but other people seem to have different ideas about the purpose(s)
of such a practice. 

I think it might be interesting to try the "sign-or-delay" rule on
a part-time basis - perhaps weekends only, or never on weekends, or
only during December, or whatever. To me, it seems useful as sort of
a "Great American Smoke-Out Day" for crypto; to get folks to go just
one day where they use crypto in a practical, applied way, to prove
that they can do it. What they do after that is their own business.


-----BEGIN PGP SIGNATURE-----
Version: 2.6.2

iQCVAwUBLtvFC33YhjZY3fMNAQGc+QP/R2kBRlCNVLDVJNOLOGOWv6URYmnj+qTt
poo1LKtz31Mzj+rBAiXPZSYY5xPtTXKD/7X8dU3JYyJbH12kwvH/RS1GS4mEV++V
QDJD6L84EekrdFy0piP7jsGDMq2SQsHnI6G3GG0koGoMN/3u/UbDiCG3+yJ1b5u1
iMCS8dZQTfA=
=eCnh
-----END PGP SIGNATURE-----

(I haven't been getting list mail all day...just a few message getting
through Netcom's mail bouncer, so....)

Greg Broiles wrote:

> Seems like one way to encourage the use of digital signatures is to
> start forging messages from people who don't ordinarily sign their
> messages. Necessity is the mother of invention, and all of that.
> 
> I finally started signing my messages on a regular basis as a result
> of Detweiler forging a message which purported to be from me. On the
> other hand, I think Tim has been the most frequent target of 
> Detweiler's forgeries, and I don't detect much of a creep towards
> signing messages on his part. 

Several points, and I'll try not to repeat points I made in my long
essay of early this morning:

1. Only one person has reported to me that they were unable to verify
my PGP sig (Lance Cottrell reported this...if others did, maybe their
messages haven't gotten through to me)). From this I conclude that few
people check PGP sigs.

(The "PGP 2.7" and the ASCII message in the sig might've provided some
clues.) 

2. This does not make such sigs useless of course, as the main value
is in "critical" situations. (Legal cases, forgeries, diplomacy,
contracts, etc.)

3. Again, crypto is about economics. In the military, crypto is a big
part of operations (maybe 5% of staff on ships is connected with
crypto, communications, etc.). But the military has real needs, and
can afford (via our tax dollars) to have such efforts. Most of us are
not dealing with such critical uses.

4. Speaking for myself, I have not generated or transmitted a file I
felt *needed* to be signed, encrypted, etc. This is not to say such
situations don't exist for others, won't someday exist for me, etc.
Just things as they now stand.

(When contracts are handled electronically, when payments are made
electronically, etc., then such uses will be more apparent. But I am
fairly open about my politics--indeed, I fly the flag of crypto
anarchy in visible places--and have few files I transmit that I need
to encrypt. Your mileage may vary.)

5. The Detweiler thing was amusing. No such thing as bad publicity
(unless it's the Pinto-um RISK chip). Detweiler's forgeries had no
legal effect on me, no lasting effect. Also, those who were  "taken
in" by his forgeries would hardly be in a position to verify my sig
(to know who I was, to look up my PK on a keyserver, to jump through
the hoops needed, and to ensure that the "Tim May" they checked was
not in fact a phony keyserver entry...the several "BlackNet" public
keys, only one of which I generated, are instructive).

I don't discourage anyone from using crypto, from signing messages,
from routinely encrypting, etc. I just reject arguments that crypto is
"essential," today, when in fact it clearly isn't. Crying wolf and all
that.

In 2-4 years, a lot of the current incompatibilities and lack of
usability will have been worked out. About the time I expect to
actually _need_ to use more crypto.

--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: 2^859433 | Public Key: PGP and MailSafe available.
Cypherpunks list: majordomo@toad.com with body message of only: 
subscribe cypherpunks. FAQ available at ftp.netcom.com in pub/tcmay

From: "Timothy C. May" <tcmay@netcom.com>
    Date: Tue, 29 Nov 1994 22:23:09 -0800 (PST)

    1. Only one person has reported to me that they were unable to verify
    my PGP sig (Lance Cottrell reported this...if others did, maybe their
    messages haven't gotten through to me)). From this I conclude that few
    people check PGP sigs.

A safer conclusion would be that few people report signature failures,
although I suspect that your conclusion is also correct.

I noticed that your message's signature failed, but chose not to
report it.  As I recall, it failed because I didn't have the
appropriate key, although I do have your 0x54E7483F key and the key
that it appeared to be signed with wasn't available from the MIT key
server.

I also noticed Bill Stewart's signature failure on Message-Id:
<9411300425.AA21554@anchor.ho.att.com> -- ASCII armor stripping
failed.

In both cases, I assumed that the sender was trying to spoof the act
of signing and I further assumed that you were more careful to match
the form of a signed message than Bill was.  The fact that you've been
doing some spoofing lately only strengthened by sense that this was
another gag.

Often, but not always, when I see a Bad Signature message I let the
sender know about it.

			Rick

From: Greg Broiles <greg@ideath.goldenbear.com>

   Seems like one way to encourage the use of digital signatures is to
   start forging messages from people who don't ordinarily sign their
   messages. Necessity is the mother of invention, and all of that.

How about a vacation-like program that automatically finds .sig
blocks, stores them in a database and appends them at random to other
posts?

   Eric, would you mind clarifying the purpose of the "sign-or-delay"
   rule? Last time this came up I assumed that it was to encourage 
   folks who had 95% of the tools/initiative to start using crypto 
   techniques on a day-to-day basis to get off their asses and do so;
   but other people seem to have different ideas about the purpose(s)
   of such a practice. 

Some of the reasons I've explained just recently.

You are correct in the reason you state, also.  Providing an incentive
for those who are mostly there already will push many to act.  I think
that is a good thing.

One benefit I did not anticipate is an outcome of the large number of
people actually having gone through the process of setting up their
own signing mechanisms.  There are many more people now who have
hands-on experience setting these crypto mechanisms for themselves and
who consequently have a much better understanding of the
implementation issues involved.  For some problems action is ten times
more effective than theorizing.

   I think it might be interesting to try the "sign-or-delay" rule on
   a part-time basis - perhaps weekends only, or never on weekends, or
   only during December, or whatever.

This is a good suggestion.  It makes the transition even more gradual.

Eric