Re: [Pan-users] GNKSA

[Kurt Schilling]

Duncan wrote:
> Kurt Schilling posted on Sun, 10 Jul 2011 04:46:02 -0400 as excerpted:
>
> > Yes, NSR is still alive. The newsgroup tends to be relatively active on
> > a day to day basis. However, the regulars tend to be pretty much
> > "anti-PAN" since the re-write to C++ primarily because of the lack of
> > granular scoring.
>
> ??
>
> Granularity indicates grains, division into discreet quantities, as 
> opposed to unitary or continuously divisible.  (A more modern description 
> might be digital, as opposed to analog.)
>
> I don't really see how that can be accurately said of pan's scoring, 
> since it indeed has effective scoring granularity in units of one from 
> -9999 to +9999.  Pan certainly has scoring related deficiencies, 
> deficiencies, including as has been discussed here many times, the 
> ability to do anything practical with scores (automatic delete/mark-read/
> download based on score, presumably ignore/negative/watch, respectively, 
> by default), and the ability to score at all based on the body or whole 
> message (scoring based on non-overview headers was added by khaley fairly 
> recently, but there's no GUI for it; it's only possible vie direct 
> scorefile editing), but those deficiencies wouldn't appear to relate to 
> scoring granularity in any way that I can figure.  There's /way/ more 
> flexibility in that regard than people are ever likely to use, in 
> practice.

In this case the argument as been the change from the scoring format and 
rules used in the earlier version of PAN versus what we have today in 
the C++ versions. As many people who are used to being able to do as you 
 remark, the lack is seen as a serious flaw by those who are frequent 
contributors to NSR. In fact there have been many instances where a 
respondent advocated not using the C++ re-write version at all, but to 
use the older version only.

Warm regards,


Kurt Schilling