Viewer Deluxe: Hi <biglo>
I’m quite impressed by your latest study of Game Collection: PGN English - Problematic games. My personal favorite is “Too High Rating”. Many people/parsers would’ve missed it.Did you indeed run your tests through the whole database of 517000 games?
Can you share a bit more about the method and the PGN parser you used?
I’m looking at all this from CVD’s point of view. As far as I can see the few cases where the parser could’ve done a little better aren’t really worth the effort. In the Web context, there will always be some situations where the “garbage in-garbage out” principle applies. My main concern is, of course, to make sure that each and every valid PGNs is treated as such.
Can we draw any conclusions about that from your tests?
I’m especially curious about a group of problems you called “Annotation Problem - 3rd Party Software”. I agree that some of them have minor issues but, after a very close inspection, I found several of them with perfectly valid PGN – e.g. R Kempinski vs Epishin, 2001 (PGN imposes no restriction on the number of annotations). Again, I’d highly appreciate your opinion.
Thanks for the time you spent on this project.