REM said:
Well, it is a group project. Most programs had links and we simply had
to check the info to see if anything changed. This is simply way too
much for one person to do. I think that it will go smoother next year,
now that we've passed the learning curves.
The FAQ describes the list as "freeware that the members of
this newsgroup consider to be the best available in
different categories".
Of course, the reality may be slightly different. Focusing
on the phrase "best available", note that over 90% of the
programs with descriptions were nominated and seconded and
over 90% of those had 10 or more votes and will presumably
be on the list possibly along with a selection of the rest.
Thus its really what descriptions exist that mostly
determines the makeup of the final list. The nominations
and seconding has almost no effect (in fact, I suspect if
they were eliminated the list would be exactly the same) and
the voting removes the bottom few as opposed to taking the
top few. This seems reasonable since its work to create a
description so no one is going to bother to do it if they
don't think there is a very good chance that the program in
question will get selected but it nevertheless remains a
potential quirk in selecting the best.
Consider AutoHotKey and AutoIt as examples. These are
actually derived from the same source code but the first
had a description submitted and the second did not.
Regardless of their relative merits (e.g. AutoHotKey is
open source whereas AutoIt is closed source but AutoIt
includes AutoItX for which there is no counterpart in
AutoHotKey) it would seem that if one of them is acceptable
for voting so is the other. Thus the vagaries of which programs
happen to have list participants motivated enough to promote
those programs by writing a description and possibly getting
other users to vote for it seems to be a key driving factor,
not just which is best in some ideal sense.
Of course, even if the result does not entirely reflect its
FAQ description I still find the list useful.