Re: [tlug] [OT] Say _no_ to the Microsoft Office format as an ISO standard

["Stephen J. Turnbull" <stephen@example.com>]

Jean-Christophe Helary writes:
 > 
 > On 8 juil. 07, at 03:43, Stephen J. Turnbull wrote:
 > 
 > > The problem with behavioral economics is that you look at people doing
 > 
 > Why don't you simply call that sociology ?

Because it isn't.  I don't have a quarrel with socioloty.  Sociology
is about a broader class of social relations than an (honest)
economist will claim to be able to handle.  Sociology starts from
behavior in groups, and claims the power to explain it regardless of
the content of that behavior.  Economics looks at behavior with
specific content, namely resource flows, and decisions about those
resources, and uses specific constraints of material balance and
goal-orientedness to explain behavior.  If a social context lacks
resource constraints or predictable goals, economics has no way to get
a handle on it.

The distinction between behavioral economics and conventional
economics is that behavioral economics listens to what people say
they're thinking as they perform economic behaviors, while
conventional economics assume that they have some goals and works
backward to figure out what goals are consistent with the observed
behavior.  This allows us to *characterize* behavior in terms of
certain calculations, but real people rarely do such calculations.
Nevertheless, these calculations generate predicted behavior that
qualitatively resembles actual behavior.  In my experience, behavioral
economics simply is not more successful at prediction, but on the
other hand the models become much more fuzzy and are harder to use as
modules to build bigger models.

Of course the overlap of the fields of economics and sociology is
large, but they really are different fields.  Sociology has a lot to
say about the economic realm.  For example, social dynamics can
explain the phenomenon of fashion in terms of why people follow
fashion leaders, or on the contrary wish to become leaders.  Economics
can then help explain the dynamics of fashionable markets.  Or take
potlatch economies, whether the fish-based ones of Athabaskan cultures
or the code-based one of GNU (as explained by ESR).  Economics can
work with the goal "people like to throw parties", but sociology can
*explain* it in more basic terms.