From my perspective, reading C is a lot easier than reading Stata code.
I agree with the global point (it's hard enough to replicate what people
do already), but increasingly we will have to adopt the idea that making
data available isn't enough.  One will also need to provide documented
code, including seemingly minor choices like random number seeds and the
rest.  I think it mistaken to assume a priori that one language is easier
than another (though I'd make a case for Python or C); rather, it is how
many of us use a given language that increases the chances that a
community can evaluate a given piece of work.

That said, I'd like to recommend an empirical approach.  Some languages /
packages clearly work better than others for the tasks that most of us do,
and a little benchmarking of the sort Sekhon has done (and we are in
process of doing at Duke -- results to follow) would reveal a lot.
Stata, for example, has released a version that runs on multiple cores,
that for some tasks (but not all you'd think of) makes things faster
without requiring people to learn parallel programming.  R, as another
example, does poorly for large data sets, though it does have some very
cool libraries (e.g., http://jackman.stanford.edu/papers/download.php?i=1
or http://cran.r-project.org/doc/packages/wnominate.pdf).

Last, if I were just starting out and wanted a combination of speed, ready
access to quality libraries, and a clean format (e.g., for those of you
who hate `var' style stata notation), learning C + matlab is very
powerful.  Social scientists don't tend to use matlab (thus violating my
point above), but lots of other scientists do.

Best
Scott

On Fri, 6 Oct 2006, Jake Bowers wrote:

> Hi All,
>
> Thanks to all for the insightful and interesting comments about R,
> Stata, SPSS as well as about my off topic question!
>
> I had to post one more time, just to articulate one important
> criticism of encouraging wide adoption of lower-level, high-level
> languages (like C and Fortran): quality control. As Neal Beck put it:
>
> " "do you really want to now trust our colleagues to not only do good
> work but also to program correctly?" - how many gauss/r complicated
> maxlik/simulation resuls do you really believe?"
>
> Although as Jas and Phil note, the durability and power of these
> languages are not in doubt, Neal's point deserves mention. Heck, I
> know I have plenty of worry in my heart about and endlessly test my
> own programs.
>
> Best
>
> Jake
>
>
> Jake Bowers
> currently:
> Robert Wood Johnson Health Policy Scholar, '05-'07
> Institute for Quantitative Social Science
> Harvard University
>
> on leave:
> Assistant Professor of Political Science
> Faculty Associate in the Center for Political Studies, ISR
> University of Michigan
>
> http://www.umich.edu/~jwbowers
>
> **********************************************************
>              Political Methodology E-Mail List
>         Editor: Karen Long Jusko <[log in to unmask]>
> **********************************************************
>         Send messages to [log in to unmask]
>   To join the list, cancel your subscription, or modify
>            your subscription settings visit:
>
>           http://polmeth.wustl.edu/polmeth.php
>
> **********************************************************
>

**********************************************************
             Political Methodology E-Mail List
        Editor: Karen Long Jusko <[log in to unmask]>
**********************************************************
        Send messages to [log in to unmask]
  To join the list, cancel your subscription, or modify
           your subscription settings visit:

          http://polmeth.wustl.edu/polmeth.php

**********************************************************