[Freecats-Dev] Database structure & indexing - NOW we MAY have something

[Henri Chorand]

Hi all,

I've thought over the fuzzy matching issue [how do we do it?] during the 
last 24 hours (yes, I did also sleep a few hours and had a normal day's 
work) and I now have a first solution.

All I can say about it is that it should work (return reliable results), 
even if a lot probably remains to be done and if I can't tell if 
performance will be OK or miserable. Still, in a true free software 
spirit, better publish early, so here you are now.

The previous document I sent yesterday contained a number of mistakes 
which are now corrected, and did not provide anything near finished 
concerning fuzzy search. Please drop it.

This time, I'm sending it as an RTF document, because I could not make 
indents in a HTML document with Open Office - OO native format is 20 Kb, 
OO HTML is 40 Kb and OO RTF is 80 Kb.

I could ask all the project team to get and install a copy of Open 
Office, but the Mac version is still at the Final Beta stage :-(

I also sent this document to my former maths teacher (after that, he 
obtained a PhD in computing and now works in a software house), and to a 
French expert on text indexing - I worked one year in his R&D team, 
designing and writing software documentation, and he's a very nice man. 
So, let's wake up math wizards all around, offer them some coffee and 
beg for improvements ;-)


Regards,

Henri

DB_indexing.rtf
Description: RTF file