From: Hal <hfinney@shell.portal.com>
Date: Tue, 20 Aug 1996 01:35:11 +0800
To: cypherpunks@toad.com
Subject: Re: Why BlackNet *IS* a Data Haven
In-Reply-To: <v02140b00ae3dd1760748@[204.179.131.64]>
Message-ID: <199608191403.HAA25315@jobe.shell.portal.com>
MIME-Version: 1.0
Content-Type: text/plain


mccoy@communities.com (Jim McCoy) writes:
>Your descriptions of BlackNet as a data haven seem to be completely
>based upon the presumption that an anonymous contact service and contract
>exchange is the functional equivalent to a data haven.  Here are a few
>reasons why I would disagree:

I think part of the confusion here is the name "BlackNet".  As I envision
the concept, BlackNet is not really an anonymous contact service, or in
fact a network of any sort.  Rather, it is a vendor.  It buys and sells
information.  The name, while provocative, is a bit misleading in this
regard.  (This is just my model, and may not actually correspond with
Tim's or anyone else's idea.  But I think it more closely matches the
data haven concept, and in fact is more consistent with the original
announcement.)

BlackNet has a public key, and a known virtual location in the form of
certain newsgroups that it monitors.  Anyone can initiate a
communication interchange with BlackNet by posting a message to those
groups, encrypted with BlackNet's key.  Presumably in that message will
be included return address information in the form of a key and a set of
locations that will be monitored for replies.  In this way ongoing
conversations can be maintained between BlackNet and customers who are
either buying or selling to it.

BlackNet would not be used (as I see it) for direct communication
between buyers and sellers of information.  How would the BlackNet
public key fit into this model?  The existence of a specific BlackNet
public key is part of what drives me to picture it as a vendor.
Rather, BlackNet will buy information (plus unrestricted rights to
disseminate that information), add it to its catalog, and then
advertise its availability and price.

>As a publisher of "naughty bits" I do not have the ability to just toss
>data up and assume that it will be there when someone wants it.  I am forced
>to continuously monitor the appropriate newsgroups to find messages from
>people asking me to post the blueprints to the orbital mind control lasers
>or kiddie porn.  I cannot put my data onto "the Net" with an expectation
>that any arbitrary user will be able to get the bits one month later.  To
>maintain persistence I need to constantly repost my data, making it easier
>for authorities to trace me through simple taffic analysis if nothing else.

This model pictures BlackNet differently than I do.  As I see it, once
you sell your data to BlackNet you don't have to take any more steps.
There may still be problems, in that you may feel that BlackNet is
setting too high a price on the data you want to distribute.  However of
course anyone is free to start up a competing service, if they want to
take the risks.  BlackNet fees will in the long run be determined by
competitive market conditions based on the costs of maintaining
anonymity.

>[Quoting Tim May:]
>> A person in the U.S. seeking the Necronomicon posts a message to BlackNet
>> (or any similar forum, using the same methods) asking for a copy of it, or
>> offering to pay for it. (Whether the information is free or for a fee is
>> not central to the idea.) This request is, of course, untraceable.
>>
>> Anyone, anywhere in the world, with a copy of this banned material on his
>> or her private machines may see this request and respond, either giving the
>> material away, or negotiating a fee. (As I said before, the absence of a
>> robust digital cash system, bidirectionally untraceable, is a known
>> limitation of all such systems.)

This is a little different from my picture of BlackNet, as I wrote above.
I would see BlackNet as being a particular seller of information, who
will respond to this message.  It could have competitors like SafeHaven,
StrongHold, InfoBase, etc., each of which will offer data for a price,
and each of which will have its own reputation for reliability.

>Now you reveal the objection I had to BlackNet being a data haven.  What if
>only one person has a copy of this banned material?  It may not be in this
>publishers interest to have the data available to anyone for posting in
>response to the query ("Information does not want to be free, it wants to
>be expensive and liberated...") and some data is not widespread enough or of
>interest to enough people to assume that multiple copies exist to those who
>read BlackNet postings. Therefore the only way for a publisher to maintain
>availability of their data is to constantly monitor the appropriate newsgroups
>and republish for each request, persistence is maintained only through
>eternal vigilence (much like liberty, only requiring a lot more effort :)

Here is where BlackNet as an information middleman makes the most sense.
Its business model includes the costs of this sort of vigilance, which
after all can be automated.

>> It's a data haven.

>No, it is an anonymous contact service.  To claim this is a data haven is
>like claiming that the classified ads in a newspaper are the equivalent to
>a mall; you could probably find the same goods if you looked long enough, but
>there is a reason that manufacturers sell goods through stores rather than
>just posting classified ads across the country.  When one does not have the
>time to check the classified ads, wants to goods from a reputable source, and
>wants the goods in a timely fashion they will go to a shopping center.

Actually we now have "virtual malls" online.  These are in their infancy
but eventually they could become as easy to use and reliable as regular
malls (for appropriate kinds of goods).  All that BlackNet (as I picture
it) lacks is a WWW interface, and even that could be provided if the
gateway server could be made immune to legal pressure and if various
technicalities about anonymous WWW connections could be dealt with.

As for reputations, if BlackNet is one of several vendors of
information, like its competitors, they can all develop reputations of
their own for reliability, honesty, availability, etc.  There may be
problems if the testimonials of customers are all anonymous, but in
some cases such methods as signed transcripts of information exchanges
can be used by one side or the other to justify claims that the other
side has cheated.

Hal