[extropy-chat] Inside Vs. Outside Forecasts

[Adrian Tymes]

--- Robin Hanson <rhanson at gmu.edu> wrote:
> Do we overestimate technical change because we tend to take an inside
> view, imagining the particular process that produces some innovation,
> instead of an outside view, looking at how long similar innovations
> have
> taken in the past?

That can be turned on its head.  It has frequently been observed that
predictions overpromise in the nearterm, and underpromise in the long
term.  Perhaps the problem is not so much that outside techniques
should be emphasized in all circumstances, as that people tend to
focus too much on the current case's details (inside techniques) in the
near term, and ignore or overlook relevant details (outside techniques)
in the long term.

The solution, then, would be two-fold.  When predicting things in the
near term, look to similar situations for likely outcomes (where you
can not find solid reason why the situation at hand will be different -
although, you can often also look to similar situations with poor
results, find out what caused them, and look for said causes in your
current situation to make sure that your situation really will be
different).  When predicting things in the long term, do look for
historical analogues but keep in mind any relevant changes you now face
(example: when trying to imagine future government systems for
long-lived societies, one may see the wealth of monarchies and other
dictatorships in the past, but one should not overlook the general
empowerment of common citizens over the past two centuries - which has
greatly aided the spread of functional republics, and kept many though
not all of them from falling to corrupt dictators, and eventually
helped restore most of those that have fallen).

It is not often the case that one technique can totally dominate
another in all circumstances.  Whenever I see such a claim, in a field
with which I have some familiarity, I usually look to see where the
"dominated" technique can be superior in certain circumstances - or if
the new technique is actually a refinement of the old, acknowledging
the old technique's (limited) place (which is most likely to truly
dominate).