RE: ::scr #include "meta/identifying types of post thread"

[Dan Argent <dan@xxxxxxxxxxxxxxxxxxxx>]

> -----Original Message-----
> From: Greg McCarroll [mailto:greg@xxxxxxxxxxxxxxxxxxxxx]
> Sent: 17 April 2002 20:32
> To: scr@xxxxxxxxxxxxxx
> Subject: Re: ::scr #include "meta/identifying types of post thread"
> 
> 
> * Richard Clamp (richardc@xxxxxxxxxxxxx) wrote:
> > http://mirth.unixbeard.net/~richardc/scr.stats
> 
> I've always been a great fan of these S/N measurements and yet I know
> they have their critics.

In my defense, as top-of-the-list, I don't think it's applicable to
"conversational concepts".
It's like paying somebody for the number of lines of code they write per
day. 

I know ::scr is a waffly list, but if I think that I can make a point with
one line of text,
I assume that everyone won't need the next step, or related concepts spelt
out for them.

> Basically its very hard to determine am measurement of quality, as its
> very subjective, Schwern suggests we instead use an artificial
> measurement, and accept it is not really ideal.
> 
> In other words, If a problem is sufficiently hard, don't solve it,
> just ignore it or cheat.

Or rather, quality cannot be measured because it is subjective. (there, I
did it again)

:-)

dan