On Sun, Dec 30, 2012 at 10:13:53AM +0100, Mikkel Eide Eriksen wrote: > On 30/12/2012, at 05.08, Ron Savage <r...@savage.net.au> wrote: > > On 30/12/12 13:46, Philip Durbin wrote: > >> I just noticed http://en.wikipedia.org/wiki/GEDCOM#GEDCOM_X which led me > >> to http://gedcomx.org and https://github.com/FamilySearch/gedcomx > > > > I've come across it but I thought it had long been abandoned. I see > > githubh says recent work on the code has taken place. > > I've been following it somewhat on & off for a few months. It's still > very much a work in progress, with drastic changes being made at > times.
I was following the effort until github changed their notifications a few months ago. I must admit to being not particularly impressed. GEDCOMX seemed to arrive full of bluster, Java and XML, and gave the impression that the only unsolved problems were things like deciding how to store the XML and then just finishing the programming. That was rather uninteresting to me. Far more interesting was the data model, and despite the programming going on, GEDCOMX didn't seem to have that nailed down. And as far as it was exposed, it didn't really seem to build on any of the other work that had taken place in this area over the last decade or so. Other people also noticed this and so there was some measure of working on specifications, but whilst the wiki at https://github.com/FamilySearch/gedcomx/wiki does describe the high-level aims of the project it's still unclear to me exactly how GEDCOMX is planning to solve those problems. As you can see from that page, the link to the data model points to a page which has yet to be written. Perfect is the enemy of good, and this has dogged many similar initiatives, but GEDCOMX seems to have swung in the opposite direction entirely. I would be pleased to see the project make real progress but I'm afraid that I find it unlikely. -- Paul Johnson - p...@pjcj.net http://www.pjcj.net
