Prev: Best practice for password hashing (proposal)
Next: Shamir Secret Sharing Implementation Query
From: The Dougster on 16 Jun 2010 14:32
The thread I started June 20, 1999 here in sci.crypt is available at:

http://groups.google.com/group/sci.crypt/browse_thread/thread/7f87e056cef19481/add64476e5345deb?hl=en&ie=UTF-8&q=sci.crypt+dgoncz+sexual#

Highlights are:

From David A Molnar:

-----Doug Goncz <dgo...(a)aol.com> wrote:
> It is for the good of the public that the government or a health agency might
> wish to keep records of sexual contacts between people. On the other hand, the


Please be careful with rhetoric like that. You might set some people
off.
-----

From Michael J. Fromberger:

-----
Doug Goncz ) writes:

>It is for the good of the public that the government or a health
>agency might wish to keep records of sexual contacts between people.



With all due respect, I think this is the biggest load of hoo-ha
since
the advent of deconstructionism. What possible "public good" could
be
served by having the government or health officials keep track of
such
contacts?
-----

and
From David P. Jablon:

-----
Surprisingly, in very special cases, cryptography can help.

In a paper on "comparing information without revealing it",
one of the authors was motivated by a sexual harrassment privacy
problem: Two people wanted to compare notes to see if they
were talking about the same third person (a victim), but neither
wanted to reveal the name of that person in case it wasn't the same.
-----

In reply to David P. Jablon's point today, about ten years later.
("Few, but ripe")

The status of the zero-knowledge proof of knowledge is reviewed in
Wikipedia.

http://en.wikipedia.org/wiki/Zero-knowledge

"In cryptography, a zero-knowledge proof or zero-knowledge protocol is
an interactive method for one party to prove to another that a
(usually mathematical) statement is true, without revealing anything
other than the veracity of the statement."

This was tried in this context (sort of) by CheckTonight.com, who
(which) provided a repository for *negative-only* STD/STI results
uploaded by fax from validated doctor/lab offices to the services, and
a protocol for CheckTonight.com users to meet, greet, validate, and
fornicate without undue delay or risk. Whois CheckTonight.com reports:

( http://reports.internic.net/cgi/whois?whois_nic=checktonight.com&type=domain
)

Domain Name: CHECKTONIGHT.COM
...
Status: clientDeleteProhibited
Status: clientRenewProhibited
Status: clientTransferProhibited
Status: clientUpdateProhibited
....

I'm not a whois expert, but if renew is prohibited, the other stuff
I've read means they tanked.

Too bad. I was one of their first users.

I'd like to resume this thread. Please feel free to participate if
you've read into that first link a bit.

*Main Point*

I think the research that needs to be done is to determine a fifth-
normal form database record format for "a sexual contact", establish a
minimal encoding (a compact and portable notation) readable by both
humans and computers, and associate a cryptographic standard with the
encoding (as a standard for encrypting the encoding) such that the
people who need to keep track of their own and their partner's sexual
contacts (and their partner's partners, to any degree of association)
have the tools they need to do so, in human or computer format, as
they choose, and have available a method of encryption for storage or
transmission that is appropriate to the task.

That thing I wrote in the first post should probably have been written
"It would presumably be for the public good..." rather than "It is for
the public good". I have *no idea* what motivates large groups of
people, or governments, or bureaucracies.

Let me speak again to David P. Jablon's point. A zero-knowledge proof
of knowledge would allow Alice, a potential partner for Bob, to query
Cathy, and determine (to an asymptotic 0 or 1 probability), say, that
Bob has HIV, without Cathy knowing that Alice is asking about Bob. (I
think. I do not do proofs very well.)


Cheers,

Douglas (Dana) Goncz, CPS
Replikon Research
Seven Corners, VA 22044-0394





From: Joseph Ashwood on 21 Jun 2010 05:29
"Mr. B" <not(a)supplied.com> wrote in message
news:hvil69$i9$1(a)speranza.aioe.org...
> Joseph Ashwood wrote:

> Well, this sounds like the basis of an interactive proof system -- Bob has
> an incentive to produce some specific results, and Alice demands proof
> that
> Bob is not lying.

Actually it is different. Alice has incentive to prove Bob has specific
results. Bob has incentive to produce the same results. and the verifier
which we somehow ended up labelling Charlie has an incentive to produce the
same result, otherwise he will lose clients. Add in that it is still
attempting to prove a negative, and the result is very clear, it won't work,
ever.

> So Alice should only accept results from a certain maximum amount of time
> in
> the past.

And already Alice should insist on Bob being tested, she doesn't. Alice
should already assume Bob has tested positive for something (1/3 chance),
she doesn't. The time frame for validity is less than 0 for the test, Bob
can test negative and be infectious.

>> That ranks right up there with "I have no idea why she gave me her
>> number"
>> we'll see whether or not she chases your Cadillac with a 9 iron.
>
> Well that is why I proposed the alternative, placing the entire system in
> a
> portable device that people carry on their person (much in the same way
> that
> people carry condoms with them). Thus, the marital strife is created when
> a
> spouse is caught with such a device -- but that is no different than
> catching a spouse carrying condoms around. We already encourage condom
> use,
> so perhaps we would also encourage people to use such devices (perhaps a
> catchy slogan).

There's a big difference though. Condoms are cheap enough to throw away, and
common enough to buy on the way. Such an electronic device requires a
stateful system, so the device is inherently around longer, costlier to
acquire, and takes longer to establish. Even with all that, condoms are
already a reason for a 9 iron wakeup call.

> Really though, the requirements of this system seem to guarantee that it
> will never be deployed.

That's the smartest thing any of us has said about this. We may disagree on
the reasons, but it simply will not be put into use.
Joe

 | 
Pages: 1
Prev: Best practice for password hashing (proposal)
Next: Shamir Secret Sharing Implementation Query