> OK I see your point. It seems to me just scidb is getting a bit > overloaded with this.
You don't see the final result, this will be exposed with the first final version. Please keep in mind that Scidb is definitively not ChessBase, nor is it Scid, although it's a chess database, of course. > Anyway my opinion is semi-structured data (xml) is the best way > to do it. Yes, that's a good formulation, a "semi-strcutured data (xml)", that's what I like to do. A full-structured XML is not of interest for this purpose. A full structured XML is appropriate for Web browser content, for example. > Would be nice opening a new open project defining the new format > and provide API and implementation for parsing in a lot of languages. I've planned to provide a C/C++ library, would be nice if some volunteers will do the "translation" into other languages. I cannot do all.
