> Robert,
>
> The discussions earlier this week about the nature of explanation
> yielded 2 notions about the necessity of historical contingency in
> modeling. One referred to 'real historical data', that is, the
> elements of the model reflect a sampling of some actual situation, and
> can be explained as some abstract transformation between 2 historical
> data points. The other referred to the idea of understanding the
> historical situatedness of the modeling methodologies employed, such
> that one can explain what one is doing and why. It's probably useful
> to consider these as different kinds of explanation.
>
> I have not been reading JASSS lately, except when specific papers get
> recommended, so can't comment on whether their reviewers are pushing
> for greater validation against historical data. There was a
> (humbling) article in the 14 June edition of Nature about mentoring
> better reviewers .
>
> As to the methodology notion about explanation, there could be an idea
> about the responsibility of authors to employ the history of their
> methodologies when explaining their results across disciplinary or
> research group silos. I've been reading Thurston (
>
http://arxiv.org/PS_cache/math/pdf/9404/9404236v1.pdf ) about just how
> difficult this can be (at least for mathematicians) and Corfield (
>
http://www.dcorfield.pwp.blueyonder.co.uk/HowMathematicians.pdf )
> about how research groups might formulate their methodologies and
> programs so they can be effectively communicated.
>
> We might take JASSS to task for setting the bar too low, but to be
> fair the problem may simply be that this kind of modeling is not far
> enough along as a discipline for its practitioners to have the
> training and expertise to do the latter kind of explanation. It may
> also be that thus far it takes most of a given career to get any good
> at it. In either case these are early days, and it seems to me there
> is at least the hint of a path, if there is the will to build towards it.
>
> Carl
>
> Robert Holmes wrote:
>> The latest issue of Journal of Artificial Societies and Simulation is
>> available at
http://jasss.soc.surrey.ac.uk/JASSS.html>>
>> I dunno, after our discussions about the nature of explanation,
>> reading JASSS left me thoroughly depressed. Want to guess how many
>> papers compared their simulation results with real historic data?
>>
>> Robert
>> ------------------------------------------------------------------------
>>
>> ============================================================
>> FRIAM Applied Complexity Group listserv
>> Meets Fridays 9a-11:30 at cafe at St. John's College
>> lectures, archives, unsubscribe, maps at
http://www.friam.org>