Am 11.05.2011 17:52, schrieb Richard Smith: > I've spent years looking for decent genealogy software that suits my > needs, and I'm almost at the stage of giving up and writing my own. > However, before I do that, I thought I'd ask on this newsgroup whether > anyone has any suggestions of suitable software. > [Detailed specifications] Hello Richard! When reading your specs, I was reminded of two IT-terms: "knowledge representation" and "fuzzy logic". When looking for existing software I would consider these to be more important than genealogy as mere application domain. Whatever you may find, I am interested in the results :-) regards, Kay

On May 11, 8:00 pm, Kay Wischkony <[email protected]> wrote: > When reading your specs, I was reminded of two IT-terms: > "knowledge representation" and "fuzzy logic". When looking > for existing software I would consider these to be more > important than genealogy as mere application domain. I completely agree. Unfortunately, there doesn't seem to be much in the way of existing technology (or even research papers that I've noticed) that put the two together. There are certainly lots of technologies that tick the "knowledge representation" box, and I have in past considered trying to build a genealogy framework on top of the W3 semantic web stack (RDF, OWL, SPARQL, RIF), but each time I look at it, I find myself put off by the lack of a decent reasoning engine, the fact that RIF is still evolving, and the difficulty in talking about a statement without asserting that it's true. For example, "John Smith was born on 1 Jan 1900" is a statement, and easily representable in RDF, but in genealogy you frequently want to make statements about statements without saying that the statement under consideration is true. For example neither "The Jones family bible says 'John Smith was born on 1 Jan 1900'" nor "The statement 'John Smith was born on 1 Jan 1900' is probably false" imply "John Smith was born on 1 Jan 1900". The process of making statements about statements is called quotation, and is not well supported in most of the semantic web technologies, but it's essential in any fuzzy logic application where we need to deal with contradictory sources. The same black-and-white logic applies to the rules formats (such as RIF) used in the semantic web. RIF allows us to say "If a person was baptised on some date, then he or she was born within the previous year", but as genealogists we don't want rules like that. Our rules are much more fuzzy. We want to say "If a person was baptised on some date, then, in the absence of evidence to the contrary, he or she was probably born within the previous year". Again, that's very hard, and maybe impossible, in the current generation semantic web technologies. That said, there are a lot clear parallels between the Gentech data model (GDM) and the semantic web technologies. The core part of the GDM is the assertion, and this is very similar indeed to an RDF statement, and it might be productive to think about expressing the GDM in terms of a suitable semantic web technology (probably OWL). I'll give some thought to whether I want to spend time doing that. Richard