Re: (eternity) cost metrics (fwd)

[Jim Choate <ravage@EINSTEIN.ssz.com>]

Forwarded message:

> Date: Sun, 6 Dec 1998 00:26:19 GMT
> From: Adam Back <aba@dcs.ex.ac.uk>
> Subject: Re: (eternity) cost metrics

> Jim Choate writes:
> > Adam Back <aba@dcs.ex.ac.uk> wrote:
> > > I think a better deciding factor of which files remain and which don't
> > > is hard, anonymous ecash.  [...]
> > 
> > There is some window for abuse in this method. It allows a well endowed
> > entity to bias the information available. 
> 
> The alternatives, such as say one vote per person have problems too.

I'll agree that any system that relies on charging the originator of the
information, versus the user of the information, is flawed in this way.

Does your local grocery store charge the truck that delivers its tomato's?
Does a book publisher ask that the potential author pay for the book run and
advertising? Does the newspaper charge their reporters to put their articles
in the paper?

> Cash allows you to measure the scale.

No, it allows a mechanism in the short-term for the Eternity market to be
manipulated by parties with a political interest and large buckets of cash.

> So straight cash seems I think to be a good metric.

The point of the cash is to provide a motivation to keep Eternity up, NOT a
measure of the worth of any specfic piece of information from the
perspective of the server operator, it's the users of that data that define
it's worth. That worth can and probably will have many facets that can't be
reflected in simple dollars and by definition outside the keen of the
Eternity networks current worth estimation model.

In the majority of cases the Eternity operator won't have a clue as to the
actual market worth of some data.

There is also the issue of information worth collapse once the data is
available off the Eternity network that hasn't been touched. This works to
the negative of the Eternity networks long-term goal.

For example, let's say a whistle-blower comes in with some juicy info. The
Eternity server network, through some mechanism that I've never seen
sufficiently described, determine that the originator must pay some amount
of money to post it. The data gets posted and users grab it over the next
few days. Then once it's available in the alternate channels (eg printed on
a webpage of a newspaper) the future market potential for the Eternity
servers goes through the floor.

This motivates the Eternity network to over-value the worth of data which
further keeps folks out.

> Note also my earlier comment that I view an important eternity
> objective to be preventing negative votes on data availability.  You
> can only disseminate more, say contradictory, information not remove
> information.

Well, considering that Eternity doesn't do anything regarding archiving and
long-term storage there really isn't a worry here. All that's required is to
keep the price up until the window of relevance is gone. Then back out, the
price drops back to it's original cost but now that it's relevance is gone
it slowly (or not) fades away. Depending on the particular nature of the 

> You can attempt to disseminate more copies (make
> available with higher redundancy, and fund faster download) perhaps,
> but this does not drown out the other information.

Actualy this is a complete change of track and irrelevant.

The POINT is that if there is a piece of information that I want to keep out
of the hands of others I simply go to server after server and ask for a
copy. This activity raises the price. Since I've got lots of cash and I want
that document to be expensive I continue to buy copies further inflating the
price.

> If you worry that mega-corp M will buy all available space at a
> premium, there is an easy solution:

Straw man, if the Eternity servers have a clue as M Corp. buys more space
the server operators buy more drives to sell since their now in the data
space business and not the information brokering business. Hence M Corp. has
distracted the Eternity services from their primary operations.

So a workable strategy for M Corp. would be to inflate the price on the data
they want to hide as well as continue to dump monies into the Eternity
network buying space. Then at some point they abruptly stop their support
and the Eternity services collapse because they've now grown to such an
extent they can't survive without this artificial market manipulation. So
the Eternity servers go down, further obfuscating the original information
that M Corp wanted to hide.

The Eternity model also completely ignores the consequences of such a
manipulation in regards the re-building of a new service after the collapse.
The amount of trust that potential users have will be low further increasing
the difficulty of keeping the servers operating long-term.

Because the Eternity network relies only on the immediate short-term worth
of data it is brittle.

> > There is also the question of data degredation, in the sense of
> > worth, over time versus the archival/historical worth of the
> > data.
> 
> As long as people are interested to keep the data around in low
> priority low access speed storage they will pay for this to happen.
> 
> If they don't care -- well they don't care.  People who do care can
> fund it, and evangelize to others on the merits of doing this.  No one
> has any special right to force others to pay for keeping junk around
> for the sheer historical sake of completeness of it.

So, a primary ideal of the Eternity service is that as long as the data has
a worth below some level they'll drop it.

This is one of the reasons that a long-term service like this won't work. It
isn't reliable because I can't come back 50 or a 100 years from now to get a
piece of data unless there is sufficient interest by other parties to access
it as well.

This is short-sighted and limits the use of Eternity.


    ____________________________________________________________________

           If I can put in one word what has always infuriated me
           in any person, any group, any movement, or any nation,
           it is: bullying
 
                                                Howard Zinn


       The Armadillo Group       ,::////;::-.          James Choate
       Austin, Tx               /:'///// ``::>/|/      ravage@ssz.com
       www.ssz.com            .',  ||||    `/( e\      512-451-7087
                           -====~~mm-'`-```-mm --'-
    --------------------------------------------------------------------