Newsgroups: comp.lang.basic.visual,comp.lang.c++,comp.lang.c,comp.lang.pascal,comp.lang.tcl,comp.lang.ada,comp.lang.smalltalk,comp.lang.perl,comp.lang.asm.x86,comp.lang.fortran,comp.lang.postscript,comp.lang.java,comp.lang.clipper,comp.lang.forth,comp.lang.cobol,comp.lang.rexx,comp.lang.eiffel,comp.lang.python,comp.lang.lisp,comp.lang.scheme
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.mathworks.com!newsfeed.internetmci.com!howland.reston.ans.net!ix.netcom.com!netcom.com!gperkins
From: gperkins@netcom.com (Glen C. Perkins)
Subject: Re: Language "ranking" based on posts to users groups
Message-ID: <gperkinsDJwqtH.JBI@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
References: <4at4t0$j5j@garden.csc.calpoly.edu> <4b0vt6$24k@warp.ipacific.net.au> <4b32ii$3c@raffles.technet.sg> <4b9ugn$ksg@bert.traverse.com>
Distribution: inet
Date: Wed, 20 Dec 1995 23:05:41 GMT
Lines: 45
Sender: gperkins@netcom.netcom.com
Xref: glinda.oz.cs.cmu.edu comp.lang.c++:165902 comp.lang.c:168022 comp.lang.tcl:40108 comp.lang.ada:38920 comp.lang.smalltalk:32457 comp.lang.asm.x86:15099 comp.lang.fortran:35979 comp.lang.postscript:38163 comp.lang.java:11495 comp.lang.clipper:9062 comp.lang.forth:25376 comp.lang.cobol:7015 comp.lang.rexx:13679 comp.lang.eiffel:12210 comp.lang.python:7542 comp.lang.lisp:20265 comp.lang.scheme:14624

Dan, the original poster, was kind enough to drop me a note
explaining the source of the numbers. It appears as though
the numbers are being misinterpreted by most of us.

The numbers used were simply the message numbers of the most
recent message posted in each group. I have no idea whether
those message numbers are really an accurate count of all
messages ever posted to a group, but if they are, then that's
what was being ranked: the total number of messages ever posted
to a particular group (with some groups being combined, as he
said.)

This, then, is far more a measure of the age of the groups than
it is a measure of their *current* popularity. Of course, popularity
is the Y axis of the curve whose "area under the curve" was
being measured but in this case, since the X range varies so
dramatically from group to group, it's difficult to draw
almost any conclusion.

It seems to me that the "total number of postings last month"
for each group would be a bit more interesting, and might not
be too hard for some enterprising young programmer looking for
a quick project to manage. (Dan, by the way, has no interest
in going any farther with it. He didn't intend it to be any
big deal to begin with.)

While it's true that the nature and mix of postings will vary
a great deal from group to group, there is still a sense in
which number of postings per unit time acts as a *rough*
proxy for interest level. No matter what the content of the message,
in general, the fact that you will take the time to write
something at all (rather than just doing something else)
means you have some interest in the subject. Because of this,
I think it would be interesting to see the current monthly
tallies. The monthly tallies graphed over a long time
period would make a very interesting web page, if someone
feels like doing it. It would be fun to see the leap
of interest in Pascal that came with Delphi, the jump in
Perl brought by the web, the recent Java wave, the steadiness
of Cobol, etc.

__Glen Perkins__
-- 
+---------------------------------------------------------------------------+

