This is a response to several people who have commented on my idea of
running a simulation of the world economy, and I should mention
especially Pete <[EMAIL PROTECTED]> who has some very useful ideas,
(as Tom Walker has also pointed out).
First a few general comments:
I want to do this myself, but I am willing to accept help, and I am
most interested in help collecting data. If I get enough help I am
quite willing to surrender all claims to this project to the
futurework mailing list as a whole and will gladly let someone else
take over and run the project if the list as a whole has no serious
objections at the time.
I would like someone more reliable than me to keep an archive of
whatever is produced, in case I get run over by a bus, and I'd prefer
it to be made available by FTP to anyone who wants it.
I'd like to use only public domain or freely redistributable math
libraries so the whole project can be freely redistributed under the
GNU public license.
I think the target language should be plain-vanilla ANSI C with as few
preprocessor directives as possible, and it should compile with the
GNU gcc compiler (djgpp on DOS or Windows systems) using -Wall to
enable all warnings -- but without any warnings produced. Having just
said that, I will probably do some rapid-prototyping using Pascal
which is easier and supports array bounds checking, then translate it
into C using the p2c translator at some point. I'd like to use a real
programming language, a good one, but unforunately none has ever been
written! (Though I could say a few nice things about ML or Python,
which most people have never heard of.)
In an earlier message (which I regret not responding to), Pete wrote:
> On the other issue, I have no fear of mathematics, nor engineering,
> in analysis of social issues. However, I will state categorically
> that algorithm-based analysis is inadequate to the task, and most
> likely actively deceptive. Nothing less than fullblown simulation
> is able to yield a valid analysis, but this is something easily within
> reach of current computing power. Systems engineering applied to
> the whole problem of economic srtucture is fully mature and powerful
> enough to handle the problem, and is long overdue to supplant the
> voodoo algorithms of orthodox economic theory.
I might have been more likely to respond to this comment if I
disagreed with it, such being human nature, but since it is so
obviously correct I just let it pass by unnoted -- sorry Pete.
In particular I agree with his endorsement of systems engineering, and
since I do, I think the first step must be requirements analysis,
followed by design, and only then can code be written -- except for a
small amount of rapid-prototyping as proof of concept.
Commenting on my plans to do a simulation Pete wrote:
> This is an approach I have advocated for a long time, so of course
> I'm all in favour. However there are some important points to make:
> a proper simulation is not a trivial project; I had envisioned it
> being the product of a team effort in the order of several man-years.
I'm still working on requirements analysis, but I have a few
preliminary design ideas in mind. I'd like to make it entirely
database driven, for several reasons, not least of which is my
agreement with Pete that this should really be a big project.
As I see it, the programs which actually run the simulation are in
some ways the least important part of it, and should take the least
work. The hard part is collecting good data and making it available
in some standard form. I envision the collection of data being
expandable almost endlessly, so that the simulation that might be up
and running in a few months could be expanded with several man-years
more work by simply collecting more data.
To carry on with Pete's comments:
> Most importantly, the simulation will be of no value if it is
> algorithm-driven. To reflect the true picture, it must be an FSA
> (Finite State Automata) model. Algorithms may be deduced from its
> results, but not ordained in its construction. ...
I more or less agree with this, though might have some quibbles with
using a finite-state machine, which does have some limitations. I am
quite in sympathy with Pete's views on algorithms, but I think the key
word is "ordained" -- there will be some room for algorithms loaded
dynamically as data for the purpose of approximating what is not
known. A model is just a model, after all, and cannot hope to do more
than estimate most things, but I do want to minimize any reliance on
"canonical" algorithms, to extend Pete's metaphor.
> ... The simulation should model the actions of individual players,
> and be iterated over cohorts over time. ...
I'd like to turn that around and say that the simulation will model
arbitrarily small units, which could eventually be individuals, but to
start out with the units will probably be quite large, such as
"third-world teenagers" or "literate women of childbearing age".
Exactly what units are used will be defined in the database.
> ... A well constructed simulation should be able to model
> any form of economy one can imagine, and not be limited by the
> constrictive assumptions built into an algorithm-driven simulation.
Of course. No doubt about that.
> By the way, algorithm-based simulations already exist, and are used
> regularly by economists, which should be a clear demonstration that
> they are no damn good. To see an example, check out the International
> Futures model which I believe can still be found at IFS.org.
I'd like to make it possible to run equivalent models, so we can take
apart the existing ones and see just why they are no damn good.
As I see it now, the database should include a number of named
simulations, each one of which could be loaded and run, modified and
saved under a new name, and printed out with graphs of results.
I would also like to try defining an algebra or space of models (both,
really, I think both algebraic and topological structure needs to be
represented), so that one could do various algebraic operations on
models -- combining two somewhat inadequate models that are
significantly different into a single model that is better than either
of them.
I hope this all sounds plausible to you, Pete. I need to formalize
this into a requirements document -- or the first draft of one, and as
soon as I get some of that written I will post it here.
Mark Measday also has some interesting comments which should be
reflected in formal requirements:
> Naively, should the simulation work well and teach lessons in management
> of the world economy to those looking for that data, how will the
> simulation recursively model itself within it own simulation? It must be
> an unknown quantity, no? As the results of its model produce unknown
> outputs which cannot be input to the model before it is run?
I would definitely like to model all real world processes to some
extent, and much as it may seem like ivory-tower poppycock, various
political and economic view are part of the real world -- they
influence decision makers and can be reflected in interest rates and
other factors.
I think we should incorporate a number of data records reflecting
political and economic views. The influence of a particular viewpoint
is questionable and will have to be guessed at, to start with, and
some models will simply have zeroes to indicate no influence
whatsoever. To choose an example at random, we might have a "Jay"
entry whose fields represent the influence of Jay Hanson's views on
the world economy as a whole. You are at liberty to provide your own
estimates as to the values of those fields!
And so, for completeness we ought to incorporate Mark Measday's
suggestion and include an entry for the effect of this simulation on
the world's economy as a whole. You are also at liberty to provide
your own estimates as to the values of the fields in that entry. I
like to think that more of the world's decision makers would trust the
results of our simulations than would trust the speculations of a Jay
Hanson, but then again, many of those people are politicians, not a
group known for their enlightenment.
It may seem that the simulation would have difficulty capturing its
own effects, but this is simply a matter of recursion (or iteration),
and not really a problem. One version of the simulation could be run
with the fields representing its own effects on the world economy set
to zero -- most certainly the right initial conditions. Another
version may have some higher values in those fields, and a more
sophisticated version may have second and higher order effects
represented so that the effects start out at zero and change with
time.
I wouldn't be doing this at all if I thought the simulations would
have no effect on anybody, ever. I like to think we could do it well
enough that people will take notice, at which point it may have some
small effect on decision makers. But who knows? It's worth a try,
and I want to try it. If anyone wants to help, please do!
I should mention that Jay posted a quite well written reply to my
earlier remarks on this simulation project, including a number of
numbers. I like numbers, myself, though I always view them with some
suspicion, particularly when they don't come with an error estimate.
If someone talks about 200 million people I always suspect they mean
plus or minus 100 million unless explicitly told otherwise. I'll make
a note of Jay's numbers and try to use them where appropriate. I'll
entertain suggestions about just what that appropriate use might be.
Regardless of anything said here or in my previous messages, I have
no personal animosity against Jay, and quite enjoy reading his
comments. He notes that "debating" on the internet is his idea of
fun, and who am I criticize that, though I think the word
"pontificating" might be more appropriate.
dpw
Douglas P. Wilson [EMAIL PROTECTED]
http://www.island.net/~dpwilson/index.html
