[htdig-dev] Basic question - ranking algorithm

[xaxx <xa...@id...>]

I'm new to ht:/dig , but have seen the results of using it on one 
website.  The thing is, the results are uneven on the website and 
similar word-counts and other factors result in reproducible but hugely 
variant scores (like 6.9344338 vs 2.2333322 with no visually 
recognizable differences in the pages or keywords).

In the past, when I've worked on projects, one of the ways for me to see 
where changes produce effects was to manually calculate a few results 
and compare them to the output of the compiled program.

However, while there is a page that outlines the weighting factors 
[Ranking pages and the use of Meta tags with ht://Dig], its description 
of the actual program is that the "htsearch... uses a complex rule to 
rank the pages."

Is there some source for the "complex rule" in pseudocode or some other 
simplified presentation format, that explains...better yet, demonstrates 
the complex rule?

Thanks for any help.  If the only answer is to download the htsearch 
module, I'm about to do that...

WAH