Dennis Lee Bieber wrote: > On Sun, 14 Oct 2012 18:11:50 +0100, Ian Goddard<goddai01@hotmail.co.uk> > declaimed the following in soc.genealogy.computing: > > >> I'm in the UK. But a 1st edition 6" map is the best part of a century >> older than the current coordinate system. The maps use Cassini rather >> than Mercator (and, according to information received, have different >> meridians for different counties). This means that either the old maps >> have to be used as is with an arbitrary set of coordinates which are >> simply pixel coordinates or reprojected onto OSGB. >> > Ouch: > http://www.ordnancesurvey.co.uk/oswebsite/freefun/geofacts/geo1164.html Ouch indeed. Although this reference implies that there may only be two meridians to contend with a recent email from OS customer service suggests more. Forty more. -- Ian The Hotmail address is my spam-bin. Real mail address is iang at austonley org uk

Dennis Lee Bieber wrote: > On Wed, 10 Oct 2012 10:15:58 +0100, Ian Goddard<goddai01@hotmail.co.uk> > declaimed the following in soc.genealogy.computing: > >> pblair wrote: >>> Be safe. >>> >>> I add a latitude and longitude (which seem to have been reasonably stable over the years) >>> >> >> I'm beginning to wonder about that, and I don't mean continental drift. >> >> I've just spent several days trying to georeference maps using QGIS& >> failing to get consistency between coordinates obtained by different means. > > Well... You have to take into account the datum used... There is a > difference between NAD27 (okay, I'm US biased -- most USGS maps still > use NAD27), NAD83 (short-lived I believe), WGS-72, WGS-84, UTM... And > for the UK, the OSGB grid system, etc. > > I'm in the UK. But a 1st edition 6" map is the best part of a century older than the current coordinate system. The maps use Cassini rather than Mercator (and, according to information received, have different meridians for different counties). This means that either the old maps have to be used as is with an arbitrary set of coordinates which are simply pixel coordinates or reprojected onto OSGB. Although individual sheets have lat/long grids on the edges without knowing the parameters of the original projection the only alternative is to use a lot of reference points. I can use a plugin for Google Earth to give coordinates in OSGB which is one alternative. I can also use the OpenLayers plugin to show Google Maps hybrid layer and on-the-fly reprojection to OSGB. And these two give results which differ by a hundred yards or so. And trying to overlay the downloaded OS district vector images on Google Maps introduces a whole new level of weirdness. I can also use the raw lat/long data from GE or GM & use some calculator to calculate the OS coordinates. And the calculator on the current stable Linux release of QGIS gives different answers from the previous stable or the current development or the current Windows version. In fact I've used the highest zoom level of Streetmap which I believe is based on OS data but represents buildings as simplified polygons. The features I've relied on to be unchanged over a century are mostly churches and the simplified polygons are simple rectangles but almost all the churches are irregular with porches and/or chancels which the original shows; in one case it even represents the buttresses. So does the rectangle simply represent the nave or does it enclose the whole irregular outline? In short I have a reprojection of sorts and in the middle of the map it fits the downloaded OS District raster reasonably well allowing for the fact that the much poorer resolution of the latter but near the edges, even at control points, the coordinates are out by several tens of metres. -- Ian The Hotmail address is my spam-bin. Real mail address is iang at austonley org uk

"Tony Proctor" <tony@proctor_NoMore_SPAM.net> wrote in message news:... > > "Tom Wetmore" <ttw4@verizon.net> wrote in message > news:d0200e0f-68c2-4451-b47b-51c3cbbddb7a@googlegroups.com... >> Tony, >> >> The language field is not for the type but for the name itself. Sorry for >> the confusion. >> >> Alternate spellings in the same language can be handled by allowing name >> fields to to contain >> comma-separated lists. Alternate spelling in different languages can be >> handled by either: >> >> 1) adding language namespaces to names; or >> 2) adding a new line in the gazetteer for each language. >> >> I would suggest that "official" names and "short form" names be handled >> by new columns. >> >> The example is only a suggestion intended to get some discussion going. I >> was interested >> is demonstrating how easy it is to specify multiple-containment data in a >> way that would >> allow a sophisticated place authority to operate. > > Comparing the STEMMA spec > (http://www.familyhistorydata.parallaxview.co/home/document-structure/place/place-names) > with your gazetteer format, Tom, the only substantial difference is that > STEMMA supports date ranges on the alternative names in order to cope with > renames as opposed to simply alternative spellings or colloquialisms. > > There are some smaller differences such as: STEMMA provides a canonical > name to be used for display purposes (as distinct from the alternatives > accepted during input and matching algorithms), and STEMMA's dates can be > from alternative calendars (not just Gregorian). > > Tony Proctor > ...oops! Almost forgot Tom - STEMMA's temporal ranges (used for both hierarchies and alternative names) can include the IDs of events as well as plain dates. This may be less useful for places but the relevant STEMMA structures are designed to support both place names and personal names in a consistent way. A good example for a person is where a person's name changes at the event of their marriage. Simply duplicating the date may lead to errors later if the date is modified, and it also loses the association with the event itself. Tony Proctor

"Tom Wetmore" <ttw4@verizon.net> wrote in message news:d0200e0f-68c2-4451-b47b-51c3cbbddb7a@googlegroups.com... > Tony, > > The language field is not for the type but for the name itself. Sorry for > the confusion. > > Alternate spellings in the same language can be handled by allowing name > fields to to contain > comma-separated lists. Alternate spelling in different languages can be > handled by either: > > 1) adding language namespaces to names; or > 2) adding a new line in the gazetteer for each language. > > I would suggest that "official" names and "short form" names be handled by > new columns. > > The example is only a suggestion intended to get some discussion going. I > was interested > is demonstrating how easy it is to specify multiple-containment data in a > way that would > allow a sophisticated place authority to operate. Comparing the STEMMA spec (http://www.familyhistorydata.parallaxview.co/home/document-structure/place/place-names) with your gazetteer format, Tom, the only substantial difference is that STEMMA supports date ranges on the alternative names in order to cope with renames as opposed to simply alternative spellings or colloquialisms. There are some smaller differences such as: STEMMA provides a canonical name to be used for display purposes (as distinct from the alternatives accepted during input and matching algorithms), and STEMMA's dates can be from alternative calendars (not just Gregorian). Tony Proctor

Thanks for providing this info. However, there is no type or containment information in the files that I check. We don't know what kind place a line represents, nor do we learn was larger kind of place it belongs to. There are other on-line files for the United States at least that does have this necessary information.

pblair wrote: > Be safe. > > I add a latitude and longitude (which seem to have been reasonably stable over the years) > I'm beginning to wonder about that, and I don't mean continental drift. I've just spent several days trying to georeference maps using QGIS & failing to get consistency between coordinates obtained by different means. -- Ian The Hotmail address is my spam-bin. Real mail address is iang at austonley org uk

"Tom Wetmore" <ttw4@verizon.net> wrote in message news:9581b106-022b-4e7f-89cd-ed0f654e15da@googlegroups.com... > In an associated thread I mentioned formats for gazetteers that could be > used as a "database" for implementing multiple hierarchy place > authorities. > > Here is a too simple example format, with a few examples to give an idea > of a > very simple approach. A complete gazetteer would obviously have many > millions of entries. > > Example Format: > > uniqueId : name : timePeriod : language : type : parentId* > > Example Gazetteer File: > > 10101: North America : : en : Continent > 34343: Europe : : en : Continent > > 11111: United States : 1776 to present : en : NationState: 10101 > 44444: United Kingdom: xxxx to xxxx : en : NationState: 34343 > 89999: Denmark : xxxx to xxxx : en : NationState : 34343 > > 22222: New England : : en : InformalRegion : 11111 > > 33333: Connecticut : 1776 to present : en : ProvinceState: 11111, 22222 > 55555: Connecticut Colony : 1636 to 1776 : en : Colony : 44444, 22222 > > 34543: New London : 1636 to 1776 : en : County : 55555 > 34544: New London : 1776 to present : en : County : 33333 > > 34643: New London : 1636 to 1776 : en : City : 34543 > 34643: New London : 1776 to present : en : City : 34544 > > 66666: Great Britain : : en : Island : 34343 > > 77777: Greenland : : en : Island : 10101 > 88888: Greenland : : en : Dependency: 89999 > > 80808: Isle of Man : xxxx to xxxx : en : CrownDependency : 44444 > 81818: Andreas : xxxx to xxxx : en : CivilParish : 80808 > > Omit dates where the name is "timeless" (e.g., North America). > There would be a relatively small number of types. I've suggested > some of them here. > > Some names are used twice as part of different > hierarchies. Names are ambiguous when not fully > specified. It's the way the world works. Deal with it. > In a real gazetteer some names would be used > hundreds of times, fortunately each with a uniqueID. > > Note that United Kingdom is a NationState whereas > Great Britain is "just" an Island in Europe. That's > the way of it. Homework: Add England, > Scotland, Wales, Northern Ireland. What > type would you give them? > > Try adding "British Isles" (what type would you > give them) and handling both > the NationState of Ireland and the Island of Ireland. > > Note how Greenland is both an Island (so in North > America) and a member of the Danish > Commonwealth (so "in" Denmark). Note how in > the case of Greenland we need to worry about > the difference between political containment > and geographical containment. That is, > Greenland, politically part of Denmark, is not > geographically part of Denmark or Europe! > Lots of complications, but not uncontrollable. > Smart people can figure this stuff out. > > I've shown the tip of a large iceberg here. > I've avoided certain issues, such as "official" > names, "short form" names, which would require > gazetteer entries to have more properties > than are shown in this simple example. So > the format would clearly need to have more > "columns." How many do you think there would > have to be? Political hierarchies will get > real nasty real fast. Try to imagine what would > be needed to cover all names found in what is > today Poland over the past three centuries! > > This would be a massive undertaking, but > I think the path is pretty clear. I already have > some massive gazetteer files very similar to this > format for a few of the software programs > I have written. > > Tom Wetmore The type does not need a language-code prefix Tom. It should be defined as part of a controlled vocabulary, and is hence part of the data syntax. That type should always be mapped to a descriptive term appropriate for the locale of the end-user, but that is separate from the syntax of the data itself. I see no support for alternative spellings of places, either in the same language or in separate languages. Many places have alternative spellings in the normal language of that locality (especially over time), but in dual-language regions then there will be different names in different languages too. Tony Proctor

On Tue, 2 Oct 2012 11:06:02 -0700 (PDT), Tom Wetmore <ttw4@verizon.net> wrote in soc.genealogy.computing: >In an associated thread I mentioned formats for gazetteers that could be >used as a "database" for implementing multiple hierarchy place >authorities. > >Here is a too simple example format, with a few examples to give an idea of a >very simple approach. A complete gazetteer would obviously have many >millions of entries. > >Example Format: There was an US army site that had all places on the world. Many places named with the local name and many kinds of places. I forgot the name of site but made a copy years ago. Something like nima.mil with some free FTP access. Denis -- Denis Beauregard - généalogiste émérite (FQSG) Les Français d'Amérique du Nord - www.francogene.com/genealogie--quebec/ French in North America before 1722 - www.francogene.com/quebec--genealogy/ Sur cédérom à 1780 - On CD-ROM to 1780

Tony, The language field is not for the type but for the name itself. Sorry for the confusion. Alternate spellings in the same language can be handled by allowing name fields to to contain comma-separated lists. Alternate spelling in different languages can be handled by either: 1) adding language namespaces to names; or 2) adding a new line in the gazetteer for each language. I would suggest that "official" names and "short form" names be handled by new columns. The example is only a suggestion intended to get some discussion going. I was interested is demonstrating how easy it is to specify multiple-containment data in a way that would allow a sophisticated place authority to operate.

Sorry to have misunderstood you. Good luck straightening out all those Swedish parishes!

Tom Wetmore wrote: >> Isn't that where I came in this round? I did not know and >> was unable to discover whether "word" was a house-name, a >> church parish, a town, a district, or what-the-flip. I was >> all too aware that it could be any of 'em or none of 'em. >> Having a list that verified it could be any of 'em wouldn't >> have been helpful in the least. > > If you don't know anything about a place name, not even what > kind of a place it is, you don't. What is your point? All a place > authority can do in that case is give you a full list of > resolved locations that it knows about that has your word as > one of the locations' name components. If you're lucky the > list of locations may give you a clue. I found that most folks saw my point. It has zip-city to do with hierarchies and list authorities. It has quite a bit to do with communication between genealogists.

"Tom Wetmore" <ttw4@verizon.net> wrote in message news:39fb07d6-d331-4eba-aa79-7f550607e60d@googlegroups.com... > Tony, > > Thanks. The proof will be in the pudding! > > I would like to work on the multi-hierarchy place authority, but > I have major projects that take priority right now. > > How would you handle the differences between the geographical > and political distinctions? For example the Greenland example > I gave on the other thread. Or another case, the islands of > St. Pierre and Miquelon that are off the coast of Newfoundland. > > Geographically Greenland and St. Pierre and Miquelon are in > North America. Politically Greenland is "an autonomous > country within the Kingdom of Denmark" while > St. Pierre and Miquelon is an "overseas > collectivity of France". What should a place authority do in a > cases like this? > > I believe that there should be both a geographical hierarchy > and a political one. But from there I don't know exactly > where to go. > > Tom They're good examples Tom. When I started my research, I naively expected to have a purely geographic hierarchy. That quickly evaporated though.My hierarchies start at countries and blatantly ignore continents since there's no agreement on their definition, and not everything belongs to a continent. Hence, the top of my hierarchy is political, becoming administrative within the national boundaries (country, district, etc), and ending with purely geographical for towns, villages, etc., down to households. However, "country" is not really an adequate term. The UK, for instance, is a 'sovereign state' rather than a country, and hasn't existed as long as say 'England'. Northern Ireland is part of the UK (despite the Olympic "TeamGB" name) and is sort of a country but is better described as a dependent territory. On our side of the water we also have 'Crown Dependencies', 'British Overseas Territories', and possibly even the 'British Commonwealth' to consider, but then we're known to be eccentric. The ISO 3166 standard talks of "countries and dependent territories" for the very reasons you point out. However, we need to consider historical entities as well as modern ones which is why ISO 3166-1, ISO 3166-2, and NUTS are inadequate by themselves. My research defines a controlled vocabulary for representing entities in a place hierarchy, and tries to include as many as I personally know of around the world. The beauty of time-dependent hierarchies is that a territory that moves from one entity to another can be represented without having to adopt the modern applicable hierarchy, and without having to displace any parent entity. All entities exist in the database, even if they no longer exist in the real world, or their boundaries have moved, or they have alternate names. Two problems I'm still looking at: (a) how to link two independent places that just happen to occupy the same physical location. For instance, a demolished house or bulldozed road that have been replaced by entirely new entities. (b) an entity that has been split into two or more new entities. This could be a whole country, or even just a farmstead. The two problems are obviously related to some degree. Tony Proctor

Entering in the thread after it is overdebatted... I tried to build a database with old places in France. For your information, France was divided in provinces on the basis of formation (i.e. province of Britain was acquired as a whole at some time) or regionalism, in generalites for tax purposes, in government for military control, in religious dioceses, etc. Then, in 1790, everything was structured into departements. Moreover, the borders of the entities were changed. At some time, there were 14 generalites, later 15, 16, etc. Some religious dioceses were created from parts of 2 or more existing dioceses. And you have many places with the same name and located in the same area. So, I tried to do something with that. I got something like : - modern name of town "commune" - modern code for towns (INSEE) - previous name(s) of the same or of a part - diocese, encoded with dates, i.e. Langres>1731>Dijon to mean this place was moved to another diocese in 1731 - arcdeaconry - deanery - source of that information - parishes in that town - no. of departement - no. and name of arrondissement - no. and name of canton (township) etc. I think when you build some kind of gazeeters, you will get a lot of possible structures, with time frames, name changes, etc. Subdivisions of generalites depend on the area (i.e. can be election, civil dioceses, vigueries, etc.), so even the name of area can be different. The time frames can be fuzzy. We know that in a source from 1635, that place was in that diocese, but in 1725, it is in another, perhaps because the sources are wrong, or the bishops made some land exchanges. My conclusion would be it is not possible to get a universal model. Even if limited to one country, this is not possible. Denis -- Denis Beauregard - généalogiste émérite (FQSG) Les Français d'Amérique du Nord - www.francogene.com/genealogie--quebec/ French in North America before 1722 - www.francogene.com/quebec--genealogy/ Sur cédérom à 1780 - On CD-ROM to 1780

Tom Wetmore wrote: > Cheryl, > > The purpose of the place authority is to define correctness. Note the word "authority." > You seen concerned that creating one is too difficult to be possible. That's fine. My > points are based on the possibility that one can be built. > No, my basic problem is a bit more practical than that.

>Isn't that where I came in this round? I did not know and >was unable to discover whether "word" was a house-name, a >church parish, a town, a district, or what-the-flip. I was >all too aware that it could be any of 'em or none of 'em. >Having a list that verified it could be any of 'em wouldn't >have been helpful in the least. If you don't know anything about a place name, not even what kind of a place it is, you don't. What is your point? All a place authority can do in that case is give you a full list of resolved locations that it knows about that has your word as one of the locations' name components. If you're lucky the list of locations may give you a clue.

"Tom Wetmore" <ttw4@verizon.net> wrote in message news:e0eb94af-6026-4e0a-b16b-ff33960b7a04@googlegroups.com... > Tony, > > I disasgree that things can be done with a single hierarchy, say > based on modern times, that then points off to associated data > to handle earlier time periods and other dimensions. > > Here's an example. Poland. The history of Poland is complex. > Every location in Poland fits within an incredible number of > different hierarchies. Locations > have been in many countries over time (Poland, Germany [West Prussia], > Austria, Russia, Lithuania, ...). When in those nations every location was > in > a separate political hierarchy. And each of those hierarchies changed over > time as the political leaders rearranged boundaries. In parallel there > were the religious hierarchies. Poland has been divided into catholic > parishes, protestent arrangements, Jewish arrangements, all overlapping > in time and in space. > Towns and cities had/have a multiplicity of spellings depending upon the > nationality of the speaker or the purpose of the hierarchy. > > I don't believe this complexity can be handled by a single > hierarchical tree with associated information. If I am wrong, > however, the job might be a bit easier, but I don't think so. Your > "associated information" would have to become so complex > in a case like Poland that I bet you'd throw up your hands and > head for the graph solution. > > You can see the beauty of the solution that some people use, > that is, record place names in terms of the place names > in use today. It allows them to avoid this whole messy area! > > But I think it's pretty darn weird to record an event that > occurred in Connecticut before the American revolution > as having occurred in the United States. I record those > events as occurring in Connecticut Colony (and I would > add Great Britain as the country in the cases where it were > needed). > > Tom Wetmore > > >> I think we only disagree on the multiple hierarchy issue Tom. I believe >> that >> you can create a single hierarchy which includes geographic and >> administrative entities, and then represent the other information (e.g. >> ecclesiastical, political, judicial, etc) as attributes, or properties, >> of >> the leaves in that main hierarchy. > >> This is working well for me at the moment. I have seen "nameless places", >> such as cottages in very rural landscapes that had neither house name or >> number, but these were generally addressed (at the time) by the name of >> the >> family living there. >> >> Tony Proctor You've misunderstood me Tom. You're one of the few people who has read my research pages but maybe you've forgotten that my geographical/administrative hierarchy has time-dependent up-links. Hence, a village that existed in county A before one date, and county B from that date onwards, can be represented accurately. The village exists as named entity, as do the two counties, and properties/attributes/history can be attached to any of them. It's only the relationship between them that is time-dependent. If a street is deemed to be in a particular ecclesiastical parish then all I was saying before is that this information can be recorded as a property of the entity in the place hierarchy, as opposed to in a different type of hierarchy or some type of hybrid hierarchy. I'm not hoping to convince you that this is the right way (smiling to myself here) since this is still a research project. I just wanted to clarify my approach since I believe it certainly has some mileage (i.e. worth consideration). Tony Proctor Tony Proctor

Tom Wetmore wrote: >> Well, uh, but, I would have sworn "Great Britain" was an >> island, not a country. > > Mea culpa, but see the other gazetteer thread for the correction. > >> Things were happening in Plymouth Colony and in Virginia >> before the UK existed (170-something wasn't it?) so you >> can't claim that John Rolfe's child was born in Virginia UK, >> England or Great Britain. Virginia was a stand-alone real >> estate scam until 1625 when it became a Crown Colony. If >> saying something happened in 1616 in Virginia USA is wrong, >> so is saying it happened in 1616 in Virginia Great Britain >> or Virginia UK. At least by saying USA you put people on >> the right continent. No sneezable trick these days. > > The point being? Sounds like you're in the camp that wants to > label all locations with their modern hierarchies. I won't argue > against that. A lot of people want to do that. No, Tom, I'm not. I'm saying that there is considerable question about "correct" is on this side of the Atlantic. However, you /are/ arguing against it when you say it's wrong. To be technically correct, that 1616 event happened in Virginia. Period. End of locale, unless you want to tack on North America. > Some people want more accuracy. First, one needs to agree on what constitutes ACCURACY. Then, we can argue over whether accurate but foggy trumps inaccurate but clear. Is Greenland "Greenland" or "Denmark"? You can't call either of them "wrong" unless there is consensus on which is "correct".

Tom Wetmore wrote: >> How will this magic gazetteer *know* which "Somerset" is >> meant if the only word given it is Somerset? > > It will NOT know. YOU will provide as much information as you > know about the particular Somerset, and the place authority will > give you all the possibilities that match what you know. If you > restrict the Somerset to a country or a county or a "type" (eg, > city, county, parish, village) or a time period or more than one > of the above the gazetteer will give you a shorter list. There is > no magic here. > Isn't that where I came in this round? I did not know and was unable to discover whether "word" was a house-name, a church parish, a town, a district, or what-the-flip. I was all too aware that it could be any of 'em or none of 'em. Having a list that verified it could be any of 'em wouldn't have been helpful in the least.

Cheryl, The purpose of the place authority is to define correctness. Note the word "authority." You seen concerned that creating one is too difficult to be possible. That's fine. My points are based on the possibility that one can be built. (I use the word gazetteer simply to represent the "database" of information that the place authority refers to.) If Virginia were a stand-alone political entity in 1616 the place authority's gazetteer would know that. It is the purpose of the name authority to know that stuff. The Virginia example is a very interesting one. You seem to see it as some kind of an argument against a place authority, whereas I see it as one of the main reasons we require one. Thinking that a place authority would have to decide whether Greenland is Greenland or Denmark wholly misses the point. Greenland is Greenland. Geographically it is part of North America. Politically it is a nation state within the Kingdom of Denmark (just as England and Scotland are nation states within the United Kingdom). There is no concept of rightness or wrongness here. There is simply what is (and was), and the place authority would know both of these things. Tom

Tony, Thanks. The proof will be in the pudding! I would like to work on the multi-hierarchy place authority, but I have major projects that take priority right now. How would you handle the differences between the geographical and political distinctions? For example the Greenland example I gave on the other thread. Or another case, the islands of St. Pierre and Miquelon that are off the coast of Newfoundland. Geographically Greenland and St. Pierre and Miquelon are in North America. Politically Greenland is "an autonomous country within the Kingdom of Denmark" while St. Pierre and Miquelon is an "overseas collectivity of France". What should a place authority do in a cases like this? I believe that there should be both a geographical hierarchy and a political one. But from there I don't know exactly where to go. Tom