🚀 go-pugleaf

From article <15908@bellcore.bellcore.com>, by duncan@dduck.ctt.bellcore.com (Scott Duncan):
> In article <495@pai.UUCP> erc@pai.UUCP (Eric Johnson) writes:
>>Obviously the IEEE has an ACM membership list.  I only ask that
>>they please cross-reference this list against their own membership
>>lists, since:
> While I can't argue with Eric's other three points, this one does not seem to
> actually work out in fact, though it seems logical that it should, i.e., that
> membership list coordination would save money.

In addition to the points made by Duncan as to why the intuitively
attractive strategy of mailing list coordination is not cost effective,
it is also darn hard, even if the information is all in electronic form,
to coordinate names and addresses. I once worked with a group doing
analysis of census data, and this included matching up census and
voting data (did you know there are computer records of your voting?
Not _how_ you voted, of course, but which election you voted in and
what precinct you voted at?) and data from other sources, e.g.
proprietary databases of marketing information, etc. Variant spellings
of names, e.g.

	J. L. Smith
	John Smith
	J L Smith
	John L. Smith
	J. Smith
	John L Smith
	John Smith, Jr.
	John Smith Jr			(what, is his last name Jr?)
	John Smith Jr.		etc.,

and even worse, variants of addresses, e.g.

	3171 Research Park
	3171 Research Blvd
	3171 Research Blvd.
	3171 Research  Blvd.
	3171 Research Boulevard
	3171 Research		etc.,

required a substantial amount of code, with some clever heuristics,
to produce matches between the databases. Seems we aren't very
consistent about how we write our names and addresses, and neither are
we very accurate about how we type those into the computer terminal as
we read the handwritten information.

Of course, there are always some pathological cases where there
really are two people with similar names (John Smith and Jane Smith,
both listed as J Smith, one at 131 15th Dr. and the other at 131
15th St.) that caused no end of headaches.

Coordinating mailing lists is simply not cost effective, unless there
is some common unique identifier, e.g. the ACM keeps both your ACM and
your IEEE number. Of course, you can make this work for you: use a
different middle initial on all your magazine subscriptions, and then
trace who sold their subscriber list to whom.

John Sloan  +1 513 259 1384         jsloan%spots.wright.edu@relay.cs.net
Wright State University Research Center   ...!uunet!ncrlnk!wright!jsloan
3171 Research Blvd., Kettering, OH 45420       ...!osu-cis!wright!jsloan
Logical Disclaimer: belong(opinions,jsloan). belong(opinions,_):-!,fail.