[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Junk Phone Calls, Metered Usage, and Cellphones


>Relevance to Crypto? The "junk e-mai" issue, calls for regulations (which
>I'm against), technological solutions (Caller ID lets users decided to
>accept a call or not....same idea could be used with e-mail, a la Hal
>Finney's "You have a message of size X from size Y entitled Z" proposal for
>positve acceptance of remailed messages), and the value of True Names (and
>True Numbers).

more and more I think the problem of "junk info" which we keep running
into on this list is of key importance
to future information technologies. loosely translated, I would call
this the same problem that rears its head in many different information
transfer formats. spam in newsgroups, junk mail in the email box, 
sales calls on the phone. can it be solved? caller ID is in one sense a
solution to this problem. I suspect we will be seeing increasingly
sophisticated solutions.

in short, imagine a communication system with various entities.
this could be a newsgroup, a phone system, an email system, etc.

the fundamental problem is allowing the transfer of information that
is "approved" by consenting parties, and rejecting the transfer
of information (and preventing denial-of-service attack) for 
anything otherwise.  a very tricky problem, because the value
of information is subjective in the eyes of sender and receiver,
and often people want to receive information and cannot tell whether
they want it or not until they see it.

for cpunks, there are additional goals. ideally pseudonymous 
communication or anonymous communication would be allowed.

for the phone problem, one possible way of solving this is to
have passwords. in addition to giving out your cell phone number,
you give the person a digital password. you are free to vary this
password yourself for multiple entities. they have to enter the
right password for the call to go through (or for any charges
to accrue). if you get junk calls, you at least know the individual
password that was "compromised". you can reject that password
in the future as being "dirty" and hand out a new password via
your trusted channels to anyone who tells you they can't get
through to you any more (and you don't necessarily want that
to stay that way <g>).

this system is remarkably similar to the way that intelligence
agencies work, with their agents and spies. they deal with the 
problem of knowing which channels have been "compromised" and
working with countermeasures. in fact I suspect
that the intelligence community has developed very sophisticated
ways of dealing with information transfer and "spoofing" (bad data so
to speak) that might have major applications to the design of future 

such a system could be applied to email. I send out email passwords
to my trusted associates (they might even be included in the email
address itself). email that doesn't have a proper password I could
either delete or put into the "low priority" bin. unfortunately
this restricts email whereas one of the great aspects of email
is its lack of controls and preventions in contacting people.

but notice that one could still have a lot of relative freedom
in this system. suppose that I gave a lecture to a large audience
of people. I could then create a new password for that audience,
and release it to them. if I get email under that password, I know
it was somehow from someone in that audience, and it would be
worth more to my attention, so to speak, than junk email. it has
slightly more value than being totally "out-of-the-blue anonymous". 
if the address became too popular, or got into the hands
of a marketer, I could hook it up to a form
letter or disconnect it.

in a sense this would be like something like having the ability 
to create or disconnect multiple phone numbers whenever you want.
I suspect such systems will become more prevalent in the future.
whenever you interface with other people, you will be given
the opportunity to put it under your own personal "information
channel" of choice. "inquiries on this subject should be addressed
to [x]".

a similar mechanism is used by advertisers to gauge the efficacy
of their advertising. they say in the ad, "mention this ad to
get [x]". this is setting up an independent information channel
for identification. they also might set up a separate phone 
number for a given advertisement, and see how much traffic they
get through that phone number independently. all this is
invisible to the customer.

the "junk info" problem becomes much more difficult to solve
with public forums such as newsgroups, and the above approach
would not seem to apply.

others have proposed solutions that are related to packaging
money in messages. "I will pay you 5 cents to read this message"
which can be collected upon opening it. a very interesting
proposal I think we will see actually put into practice.

I suspect we will continue to see interesting innovations that
focus on the problems of information dissemination. it's amazing
how far we are into the information age without some key
problems being solved yet. plenty of room for some innovative
thoughts. I continue to believe there are some elegant solutions
waiting to be found.