On May 16, 5:01 pm, Nick Matthews <[email protected]> wrote: > In particular, you mention standing data; I think this is an area where > open source collaborative efforts can really be made to work. I'm sure > that anyone who has spent a few years on their family tree will have > become expert on some small areas of local history and geography, if > there was a simple way to contribute that expertise, without commercial > interests taking advantage, then I'm sure it would happen. FreeBMD and > friends are a good example of what can be done. If anyone is seriously thinking about doing this, I would suggest first taking a good look at both Open Street Map <http:// www.openstreetmap.org/> and the (British) Ordnance Survey Open Data <http://data.ordnancesurvey.co.uk/>, even though their relevance may not seem immediately obvious. Open Street Map (OSM) is much more than an open source version of Google Maps. It's real power lies in its underlying dataset which is currently licensed under the Creative Commons CC-BY-SA 2.0 licence. That provides a very powerful, general way of storing geographical vector data. Often these are rounds, but they can also be boundaries, and OSM includes a internal administrative boundaries for a lot of countries. Some of this information can be viewed here: http://tools.geofabrik.de/osmi/?view=boundaries The OSM model for representing the geographical shape of areas is pretty powerful, and copes well with enclaves and exclaves (and even enclaves in exclaves). Its terminology is a bit odd because it originated through the reuse existing components, but the ideas are sound. Effectively you have areas ("boundary relations") and boundaries ("boundary ways"), and a many-to-many mapping between them with a flag on the mapping to state whether it's an inner or outer boundary. This data allows a genealogy program to place small hamlets in the correct parish, and to answer questions such as "How far away is (the nearest bit of) Dorset?" -- useful if a candidate marriage registration has just been found in an unexpected county. Another advantage to looking at OSM is that there are lots of open source tools out there for drawing pretty maps from the OSM XML. That means that once you've got the data, you can easily generate maps showing all the parishes (or other units) in a given place. Ordnance Survey Open Data (OSOD) is also openly licensed under terms that are compatible with the Creative Commons CC-BY licence. This site only contains data on the present British administrative geography, including all civil parishes (or communities in Wales), wards, districts, boroughs, unitary authorities, counties and constituencies (of various sorts) in Great Britain, and vector data indicating their boundaries. It allows you to list all the parishes in a county, or find all of the neighbouring parishes, and without you (or presumably it) having to parse the vector boundary data. It's not all that easy to drive -- it's clearly designed with developers rather than end users in mind -- but there's a huge quantity of data available. It can be accessed as RDF, either in RDF/ XML or in Turtle (a subset of Notation3 that's easier to parse), and so should be easy enough to parse and import into a database. Even though the site only contains data on the present British administrative geography, but even that poses many problems that the OS have solved. For example, it deals with data in three languages: English, Welsh, and (rather less well) Gaelic. The dataset includes multiple hierarchies: constituencies, for example, are not coterminous with counties and districts. Further, the administrative geography of Britain is very heterogeneous, more so that of any other country I'm familiar with, though I'm certain there will be more extreme examples out there if I knew where to look. A system that can cope with Britain's administrative vagaries is necessarily going to be fairly flexible. The dataset includes tens of thousands of areas which is probably large enough to give a good test suite to play with. The big thing that neither of these provides is a vocabulary for talking about changes, but doing so in a way that doesn't lose the identify of the place. Just because Hampshire was officially called the County of Southampton until 1959 doesn't mean it was a different place, and in practice both names were used before and since 1959. Nor should a small boundary change affect that. But how major does a boundary change need to be before it ceases being small? Legal continuity isn't a good criterion: for example Cornwall legally ceased to exist and recreated on 1 Apr 1974, even though Cornwall was one of the few English counties not to be affected by boundary changes on that day. There's a lot to think about in doing a good job here, but I definitely think that something like this would be hugely beneficial to genealogists and historians. The fact that getting all the data is a huge amount of work shouldn't really matter. Once a data model and interface is established, even if it needs tweaking later, people can start entering data. And with a bit of data, you can fairly easily do some seriously cool things, like generating animated maps showing the territorial evolution of an area. Perhaps gimmicky in a way, but that sort of thing can also be a good way of visualising what has happened in an area, especially somewhere as complex as modern day Germany or Belgium. I think some unique selling points like that would be enough to convince more people that it's worth adding more data. Richard