"Harrison Genealogy" <bill@harrisongenealogy.co.uk> wrote in message news:mailman.2.1349001728.21514.gencmp@rootsweb.com... > Steven > > Just had a look at GENUKI regarding what you said and I was right they > were > moved .... in circ 1844 see below from GENUKI .... > > Ancient Parishes and Registers > With the exception of a few parishes, Bedfordshire formed the Archdeaconry > of Bedford, part of the diocese of Lincoln, until 1837, from 1837-1914 the > Archdeaconry was in the diocese of Ely and in the diocese of St Albans > from > 1914. > An exchange of parishes in 1844 saw part of Everton move to > Huntingdonshire, > Ickleford to Hertfordshire and parts of Meppershall and Studham into > Bedfordshire. A further exchange in 1899 saw Swineshead transferred into > Bedfordshire and Tilbrook into Huntingdonshire. err, no. I know about these boundary changes and they are not relevant to what I said. Steven

To understand boundaries one must be an historian. If a genealogist truly wants to learn the nature of all boundaries of all geo-political-eclesiastical entities that surrounded the areas where their ancestors lived, either when their ancestors actually lived there, or as things are today, or both they must invest in learning the necessary history. Isn't that simply the nature of the beast? It's fun to bring up examples for the areas that we ourselves are experts, but it doesn't accomplish much in the long run. Some genealogists care deeply about understanding places and recording the data by some system; some don't. Some are happy just entering a name in field on a computer screen. Doesn't do them any harm after all if things aren't correct. What would be very handy, and someday we will get there, is software that will take as input any point on the globe (specified by either latitude and longitude, or just clicking on an adequately scaled map), that then outputs the compete history of all the many overlapping, possibly conflicting, multi-lingual, set of geopolitical entities that ever surrounded that point. Such gazetteers are showing up with limited scopes (e.g., history of counties in the United States; history of the changing boundaries and "provinces" making up Canada over its full history), so one can anticipate the wonder we will feel when such an application appears. Other than that, it's just slog, slog, slog to learn the history that is needed. We all become experts on small parts of the globe as we push back and find where our ancestors came from. I would not have guessed that I would ever know a lot about Andreas Parish on the Isle of Man, or the towns in Mongomeryshire, Wales, or the history of the fishing villages on Newfoundland, but when I found my ancestors there is was simply the logical result of my research.

"Tom Wetmore" <ttw4@verizon.net> wrote in message news:5669673b-9d86-430f-ba7f-74e19bd044b3@googlegroups.com... > To understand boundaries one must be an historian. If a genealogist truly > wants to learn the nature > of all boundaries of all geo-political-eclesiastical entities that > surrounded the areas where their > ancestors lived, either when their ancestors actually lived there, or as > things are today, or both they > must invest in learning the necessary history. Isn't that simply the > nature of the beast? It's fun to > bring up examples for the areas that we ourselves are experts, but it > doesn't accomplish much in > the long run. > > Some genealogists care deeply about understanding places and recording the > data by some system; > some don't. Some are happy just entering a name in field on a computer > screen. Doesn't do them > any harm after all if things aren't correct. > > What would be very handy, and someday we will get there, is software that > will take as input any > point on the globe (specified by either latitude and longitude, or just > clicking on an adequately > scaled map), that then outputs the compete history of all the many > overlapping, possibly conflicting, > multi-lingual, set of geopolitical entities that ever surrounded that > point. Such gazetteers are > showing up with limited scopes (e.g., history of counties in the United > States; history of the > changing boundaries and "provinces" making up Canada over its full > history), so one can anticipate > the wonder we will feel when such an application appears. > > Other than that, it's just slog, slog, slog to learn the history that is > needed. We all become experts > on small parts of the globe as we push back and find where our ancestors > came from. I would not > have guessed that I would ever know a lot about Andreas Parish on the Isle > of Man, or the towns in > Mongomeryshire, Wales, or the history of the fishing villages on > Newfoundland, but when I found my > ancestors there is was simply the logical result of my research. Hi Tom. I entirely agree with the concept of a Place Authority, and for historical research in general rather just family history. I really hope that we'll eventually have a consistent specification for such an Authority that can be implemented in a federated way across the globe. I'm one of those people that is almost as interested in the places as the people. However, I would find little use for specifying a location's coordinates and asking for a history of it. I view my data as a 2+1 dimensional set of places - 'places' being named points or areas deemed to have significance to people, and hence distinct from mere 'locations'. Ian already mentioned the temporal aspect. To give a few examples. The county of Staffordshire is a recognised entity, although its boundaries (and hence its centre point) have shifted over time. If my data mentions Staffordshire then it refers to the 'Staffordshire' at that time - I don't give the modern-day associated county in place of the historic designation since it's just wrong. I have many instances of a place that is a street (or even a house) that no longer exists. The location still exists but the place has long gone, and usually replaced by a new place. The point locations may have coordinates in 2-D but the places have a name and a representation in 2+1 D. What I would really like is an Authority that could resolve names in the context of a partial hierarchy (e.g. you may know the county but not the town, village, street, or whatever), and then be given the associated map reference, history, and full hierarchical designation. There is so much local knowledge of the fine-grained places that currently has nowhere for it to live. Imagine an Authority that allowed that local knowledge to be added to different parts of a Place Hierarchy (analogous to a collaborative family tree), and then being able to link place references from your family tree to the appropriate reference held by the Authority. I'm probably out on a limb here so I'll go back to sleep now Tony Proctor

Tony, Our discussions often boil down to me thinking you over analyze things, and you thinking I under analyze things! In this case I think we're in basic agreement. What you call a location seems to be what I think of just a spot on the globe, completely devoid of name or religion or geo-politics. This is what you mean by the 2D aspect of place/location. Whether a location should be just a spot, or a small area, or a large area I don't think matters -- it's somewhere on the globe we are interested in. That spot/area is effectively timeless, it's been there throughout history without regard to name, political control, or religion. Then comes the name/"place" dimension, which is, dimensionally, a forest of overlapping containment- based, named location trees that vary over the dimensions of time, language, nationality and religion. This seems more complex than your +1 indicates. The point here is there is no simple containment tree that works. The time and "purpose" dimensions require a much more complex forest structure if we truly wish to understand what a location has been called and what places it has been considered to "be in" over all of history. A place authority would be a representation of all those forests in all those dimensions. Plug in a location, output a list (actually a forest) of named-locations that contain or contained that location. Specify the time period or the religious component, and the authority could limit its output to the appropriate places that would have encompassed the location at some historical epoch. I don't understand why you wouldn't be interested in entering a location into an authority and then getting a "history" of all the places that location has been "in" over history. For me this is the FUNDAMENTAL need for anyone interested in accurately understanding the places where genealogical events occurred. I have a Place data type that I use in a variety of software, genealogical and otherwise. A Place object is simply a node in one of these forests. It can be contained by any number of other Places and it may contain any number of other Places. In "technical terms" this forest that truly represents the "naming history" of a place, is a potentially very complicated directed acyclic graph (DAG), which must be the mathematical structure than any place authority would use as its "database.' A single, "low level" Place can therefore be a member of ANY NUMBER of hierarchies. This handles all issues of parishes that cross county boundaries, or in another context, national parks that are located in more than one county. The "gazetteer" itself, which can be thought of as the external data that creates these graphs, is nothing more than a text file that lists the thousands (ultimately millions) of containment relationship between JUST PAIRS of places, along with properties that specify type of place, time dimension, language, etc. One lack of my software is that it doesn't have a way of dealing with nameless locations, points or areas. One must start with at least a single name. Clearly a real authority must bridge the gap between locations and names. With my software you could enter, say, Georgia, and it would let you know that you could be talking about a state in the USA or a country in Europe. You could enter Kings with a type of county and it would let you know all the states/provinces/countries that contain places named Kings County. That is the software can then take partially specified names and then provide the lists of all possible completely specified Places that correspond. If you are lucky there would only be one such "resolved Place;" if you aren't lucky then your data is ambiguous until you learn more. Tom Wetmore

Tom Wetmore wrote: > What would be very handy, and someday we will get there, is software that will take as input any > point on the globe (specified by either latitude and longitude, or just clicking on an adequately > scaled map), that then outputs the compete history of all the many overlapping, possibly conflicting, > multi-lingual, set of geopolitical entities that ever surrounded that point. Such gazetteers are > showing up with limited scopes (e.g., history of counties in the United States; history of the > changing boundaries and "provinces" making up Canada over its full history), so one can anticipate > the wonder we will feel when such an application appears. Anyone dealing with an unfamiliar location in the UK would be well advised to Google for it in conjunction with genuki. In particular locations in Yorkshire are well served by the contributions of Colin Hinson - but please observe his T&Cs. -- Ian The Hotmail address is my spam-bin. Real mail address is iang at austonley org uk