Subject: Re: near/medium future digital media economics
From: "Stephen J. Turnbull" <>
Date: Wed, 24 May 2006 18:11:40 +0900

>>>>> "Ben" == Ben Tilly <> writes:
    >> Didn't I write that you *did* make connections to those nodes?
    >> You just didn't "talk" to them.  You talked *through* one, and
    >> received option value aka network externality from the other.

    Ben> That's technically true.  It is also true that my telephone
    Ben> calls go through a variety of switching stations.  But
    Ben> ideally I do not know or care about those stations, and under
    Ben> ordinary circumstances I do not care which stations I go
    Ben> through or how many there are.

My point is that connectivity depends on lots of things you don't know
about.  But the virtual connection to a securities trader is a whole
different matter; you do care how many people you are connected to on
the other side of the market, that's simply the law of supply and
demand.  The more potential partners, the better the price you will get.

    Ben> This is all true.  And were I to go to Japan (not in my plans
    Ben> any time soon, but just supposing) then I'd check to see if
    Ben> we could physically meet.  Conversely if you were passing
    Ben> through the Los Angeles area, I'd hope that you'd return the
    Ben> favour.


    >> We *are* connected, and that connection has implicit value
    >> (option value, network externality) whether or not we actually
    >> use it.

    Ben> True.  However it is the fact that it has been used that
    Ben> established that value.  For instance I'm also potentially
    Ben> connected to your boss,

... and you don't care.  Of course not; this is a well-known fact
about personal connections between one person and another.  But that's
not what started this thread.  My understanding of Tom's thesis is
that the Internet may make some kinds of anonymous networks (which
were heretofore impossible) as profitable as markets.  Any analogies
to interpersonal connections are suspect.

    Ben> The common result pops up in many different ways.

    >> Mostly asymptotic.  As Edward Lasker said, "Before the endgame,
    >> the gods have seen fit to place the middle game."  For the
    >> forseeable future we need to talk about the middle game.

    Ben> I believe that the asymptotic behaviour generally dominates
    Ben> by the time you have hundreds of thousands of participants.
    Ben> (Possibly tens of thousands.)

Sure, but maybe half of those now alive will finish their lives with
no real opportunity for personal contact beyond a few hundred people
in their immediate geographical neighborhood.  *We* are rich beyond
belief, but not enough to make a huge difference to the *poor*, not
even if we share like crazy.

    Ben> Since I discovered the Internet, I have formed many
    Ben> connections with former strangers who share some common
    Ben> interest.

You're still talking about interpersonal connections.

    Ben> However in the absence of the Internet, I would have formed
    Ben> stronger connections to people that I knew through
    Ben> happenstances of geography and social circles.

    Ben> How can one quantify how much better off I am because I've
    Ben> formed connections with people that I share more interests
    Ben> with?  I can't even quantify how much better off I think that
    Ben> *I* am, let alone anyone else!

We can construct plausible theoretical models that allow us to show
existence of a utility function, and we will almost certainly be able
to use that utility function to show that given any plausible
"attention budget", your utility from internet-mediated personal
connections is bounded above as some finite factor times your base
utility from "happenstance".

If we're going to get superlinear benefits from networks, we need to
try harder than that. :-)

    >> No, that's not what I mean.  I'm distinguishing among whether
    >> the searches are explicitly initiated by a human, or whether
    >> they are implicit in a higher-level request.  Eg, suppose I
    >> asked a human secretary who was handling my FSB correspondence
    >> to "get me Ben Tilly's phone number."  He might search my
    >> emails; even if your public FSB persona doesn't provide it,
    >> your personal email might.  Google.  Call the phone company.
    >> Check the paper files.

    Ben> OK.  So you're thinking of searches that are currently done
    Ben> by people but which *could* be done by computers.

No.  It should be obvious why not---that falls into the personal
connection class, perhaps with a more or less unbelieveably large,
*but constant*, enhancement factor, and then your power laws will
follow anyway.

    Ben> Are you just guessing that my number is unlisted?

Heh.  I just needed that assumption to get to the ObscureRef to a
Heinlein novel. :-)

    Ben> Incidentally Google is trying to be that Sufficiently Smart
    Ben> Search Engine.  And the route that they are taking is to try
    Ben> to aggregate everything into one search database, rather than
    Ben> trying to integrate different databases.

Ah, but IMSEO they have to.  See below.

    >> I'll take $100 of that bet, please.  Small, unmarked
    >> bills---you can just send it now.  If it's an "emergent
    >> effect", then no single company *can* deliver by definition.

    Ben> Huh?  If there is an emergent effect in how people work that
    Ben> is mediated by searching, I agree that Google will not
    Ben> capture the value of that emergent effect.  But unless the
    Ben> current search landscape changes radically, it will be Google
    Ben> that will deliver the searches which that effect is based on.

I think I disagree.  An emergent effect is distributed, almost by
definition.  This emergent effect will have to be part of the search
process itself; the other aspects of people's work can't change very
much.  Google is not going to provide the emergent effect.

On the other hand, by providing a portal to the distributed network,
ie, the searches that people do directly, Google can capture much of
the value, just as stockbrokers can capture some part of the value of
the stock market because you have to be a member of the exchange to
trade on it.  (This is why Google is trying to assemble a single
database, IMO.)

    Ben> A good test case might be Jabber networks.

No, because they are interpersonal.  You don't Jabber people you don't

    Ben> Wikis make another good comparison case.

Wikis are more like it.

    >> But note, you may not be participating "zero".  Somebody may be
    >> republishing your words posted here, over there.  Isn't that
    >> precisely what free licenses are intended to enable?

    Ben> This has certainly happened to me before.  However I still
    Ben> don't consider myself to be a participant on that board.

Ah, but Tom (for the purposes of his essay) would, I believe.

    Ben> This is particularly important for this question - we're
    Ben> trying to discuss how the value of a network changes
    Ben> depending on how many members it has, and the discussion of
    Ben> network effects focus on the value to one person of having
    Ben> other people pay attention to that network.  Those arguments
    Ben> don't extend to the situation where there is an effectively
    Ben> infinite number of "members" who pay absolutely no attention
    Ben> to said network.

The financial markets are a counterexample.

    Ben> Incidentally with some glaring exceptions (the bubble era
    Ben> comes to mind), markets have evaluated valuations more in
    Ben> line with n log(n) than with n*n.  For instance when two
    Ben> competitors interconnect, their combined valuation doesn't
    Ben> sharply rise like Metcalfe's Law would project.

In fact, as far as I can remember, it doesn't rise as much as n log n
would predict.

But note this depends as much on the firm's ability to *capture* the
value as much as on the *total* value.  If the price per customer is
limited by competitive factors to the cost of a connection, then the
market valuation of the combined firms is based on cost-plus pricing,
not on value-based pricing.

    Ben> In other words this is one of the cases where the value
    Ben> placed on you by the second-best connection matters.

Which is an order statistic ....  :^)

I'll concede you have a point, that the problem with bargaining is
different from the problem where the seeker just chooses the best from
a menu of ultimatums.  However, although use of rival offers as
bargaining tools does change the balance between seeker and sought, I
don't think it changes the nature of the problem---you still end up
with the order statistic.

    Ben> I'm saying that I think the effect of Google increasing how
    Ben> much time people spend on the Internet is less important than
    Ben> its effect of enabling people to find what they're looking
    Ben> for on the internet.  I admit that I don't have quantitative
    Ben> results backing that up, but I know that for myself, Google
    Ben> hasn't changed how much I use the internet.

Who was talking about Google changing how much people use the
Internet?  I thought we were talking about what happens as more people
use the Internet for what ever reason they do that.

    >> In this case, yes; I just mistook the estimate badly.  But
    >> several fundamental ideas of economics depend on inherently
    >> biased estimates of that kind.  Eg, we know for sure that the
    >> average value of purchased goods is greater than the market
    >> price, but we (at least "we economists", but not necessarily
    >> limited to them) go ahead and call market price "value",
    >> anyway.

    Ben> Which then makes it hard to figure out how to properly value
    Ben> something like open source software. :-)

Yes.  That is *precisely* why sensible economists are so reluctant to
*completely* abandon intellectual property.

In open source projects, just like any service business, we talk about
how we want to provide "users" with the best software we can.  But at
the end of the day, the only user we've satisfied is the one we met in
the mirror while shaving or putting on lipstick.  We haven't a clue as
to what the rest of them want, and our software mostly shows it.

That's why you see relatively few new "killer apps" out of the free
software community per se (as opposed to universities, beltway
bandits, and other grant-mongering research institutions), and so much
taillight-chasing in the large commercially supported projects.  It's
not a lack of ability or creativity; it's a lack of any need to keep
any customer but oneself satisfied.  At best you end up serving your
existing user base, but you have no criteria by which to compare the
importance of their needs to those of potential users.

Graduate School of Systems and Information Engineering   University of Tsukuba        Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
        Economics of Information Communication and Computation Systems
          Experimental Economics, Microeconomic Theory, Game Theory