David J. Looney wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On Sat, 4 Jun 2005 23:48:53 -0700
Todd Walton <[EMAIL PROTECTED]> wrote:
"Data" is not the plural of "anecdote".
Snarfed. But this ususally seems to be quoted in the opposite order:
allan franko <allan.franko at cancerboard.ab.ca> - "The pleural of
anecdote is not data."
The plural of anecdote is statistics.
Do not outright dismiss anecdotal evidence. It allows you to form a
hypothesis, make a prediction, and watch the results. Science requires
*testability*--data is only required in the service of that goal.
Mountains of data are as useless as no data if no testable hypothesis
can be made.
While I prefer "hard" data, sociological research often cannot get at
that data. Things like drug usage, sexuality surveys, etc. are all
based around aggregations of "soft" data. That makes them less certain,
but not necessarily less useful.
In addition, business data is *hard* to get at. This list should feel a
specifal affinity for how hard it is to produce decent data and
predictions about software development.
Finally, while you attack his anecdotes, you provide no refutation of
your own. While extraordinary claims require extraordinary evidence,
his claims are far from extraordinary. Most are backed up by multiple
sources as well as news reports and 10K filings.
I don't believe a priori that WalMart is a problem, but the mounting
evidence certainly suggests that they need to be watched.
-a
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
