OT: Crazy Programming

[Huaiyu Zhu]

Patrick <postmisc at yahoo.com.au> wrote:
>
>"James J. Besemer" <jb at cascade-sys.com> wrote in message
>news:mailman.1021636709.31088.python-list at python.org...
>> [...]
>> Laura's choice of a particularly subjective example actually underscores
>> the point.  Although there is a lot of inescapable subjectivity regarding
>> ranking wines, a general consensus nevertheless emerged regarding a great
>> number of "measurements".
>
>It's possible that this consensus emerges as a result of training, rather
>than as a result of any quality inherent in the wine. Or, on second
>thoughts, both: An aspiring wine taster learns to recognise and appreciate
>the very qualities that an expert has determined to be "good". To some
>extent, the choice of these qualities is arbitrary, but once they've been
>adequately described and considered authoritative, they can be recognised
>and, to some extent, measured. Of course that says very little about whether
>the tasters are recognising "quality", rather than "a quality". The latter
>seems far more likely to me (which in no way undermines the value of refined
>taste).

It is also possible that all these are true, namely, rankings are influenced
by inherent properties of the object being ranked, by circumstantial
conditions of the person doing the ranking and by social conditions that
have positive feedback on collective behavior.  Generally we say it is
subjective if the result varies a great deal depending on the personal
circumstances; we say it is objective if it is quite uniform under most
circumstances.

Those rankings that are really objective usually turn out to be quantifiable
and could be performed by machines.  Those that cannot be measured often
turn out to be not as objective as once believed.  This is not a statement
on the a priori dependency between objectivity and measurability, but rather
a reflection on the unreliability of human beings as ranking machines.

Huaiyu