Place talk:Scotland


Proposal for Organizing Places in Scotland

Currently, the Place records under Place:Scotland are an amalgam of places automatically populated from multiple sources, and do not reflect a coherent organization. For example, looking at the list of places contained in Place:Scotland, there is a mix of traditional counties, unitary authorities, villages, and other entities, though no parishes per se. Villages should all belong to a more local grouping (i.e., a parish or a county) and should not be directly under Scotland. The important layer of parishes is not really represented here. And there is the messy issue of traditional counties vs. unitary authorities (or modern council areas). The gist is that Scottish local government was substantially reorganized in 1975 and again in 1995, such that the official geographic subdivisions of Scotland used by the government today are rather different from those used prior to 1975.

I think that most of the places have a Place page already, but that in many cases the "Type" field is missing (or "unknown"), and/or the "Located in" field is not consistently filled out. (Note that the "Located in" field is important, because it drives the automatic population of the lists of "contained places" for the higher-level place pages. I.e., in order for the Place:Scotland page to properly list all of the traditional counties, it requires that every "traditional county" place page identify its Type as "Traditional county", and is "Located in" as "Place:Scotland". Obviously, this is not happening at the moment. Also, note that case must be consistent. Currently, some are typed as "Traditional county" and others as "Traditional County" with capital C, so they get listed separately.)

The first phase of the project would include construction of the necessary geographic "containers" (i.e., counties and parishes):

  1. visit the page for each "Traditional county" (I believe there are 33 of them), creating a new page if one doesn't already exist, and make sure the following:
    • Type: "Traditional county"
    • Located in: "Scotland" from <blank> to 1975
    • validate that Wikipedia, FHLC, and Getty refs are correct (if we decide to keep Getty refs -- see comment below)
    • add link to appropriate GENUKI page for county
    • include a {{place-stub}} tag if appropriate
  2. visit the page for each "Unitary authority" (there are thirty-something of them), creating a new page if one doesn't already exist, and make sure the following:
    • Type: "Unitary authority" (or do we prefer "Council area"?)
    • Located in: "Scotland" from 1996 to <blank>
    • See also: relevant Traditional county pages
    • validate that Wikipedia, FHLC, and Getty refs are correct (if we decide to keep Getty refs -- see comment below)
    • include a {{place-stub}} tag if appropriate
    • include at least cursory content explaining relationship of this present area to the specific Traditional counties that cover the same or overlapping area
  3. visit the page for each "Parish" (there are about 900 of them), creating a new page if one doesn't already exist, and make sure the following:
    • Type: "Parish"
    • Located in: <traditional county> from <blank> to 1975 (or appropriate "from" date if you know)
    • validate that Wikipedia, FHLC, and Getty refs are correct (if we decide to keep Getty refs -- see comment below)
    • add link to appropriate GENUKI page for parish
    • include a {{place-stub}} tag if appropriate

The second phase of the project would be grooming the specific Place pages (i.e., towns, cities, and other specific locations) for the following:

  • validate Type - should be one of:
    • "City" (i.e., Edinburgh, Aberdeen, Glasgow, and Dundee)
    • "Town"
    • "Inhabited place" (e.g., a named farm)
  • Located in: <unitary authority> from 1996 to <blank>
  • Formerly located in: <parish> from <blank> to 1854
  • Formerly located in: <traditional county> from 1855 to 1975
  • validate that Wikipedia, FHLC, and Getty refs are correct (if we decide to keep Getty refs -- see comment below)
  • validate (or provide) lat/long

Note on the type: I think there may be technical distinctions between a "town", "village", "hamlet", etc., but I don't think we should try to get into that, especially as the categories, as well as how they applied to specific places, have surely varied over time. I think it's mostly just important to distinguish the "town" sort of places from the "farm" sort of places, and hopefully that's a bright enough line.

For the grooming phase, we'll need to figure out an organized way to find all of the other Place pages that are in Scotland, and are something other than a "Traditional county", "Unitary authority", or a "Parish". (We'll also need to check that there aren't any Place pages inappropriately typed as a "Traditional county", "Unitary authority", or "Parish".) I'm sure some Wiki wizards know how to do come up with such lists easily.

Finally, we'll want to discuss the other sorts of Place pages folks may wish to create (e.g., "Cemetery", "Church", "Estate"), and to establish guidelines for how those should fit into the "containment hierarchy". We'll just want to make sure things are "typed" and "located in" consistently.

Note on Getty refs: I have mixed feelings about these. I've browsed thru the Getty info, and I believe that it is the source of much of our current Scottish placename muddle. They don't have any clear sense of traditional counties versus unitary authorities, and they use the type "unitary authority" inconsistently. Do people see value in preserving the ref to the Getty? If one cared about the answer to the question "where did this page originally come from?", one could always look to the revision history to find the Getty ref. But I'm not sure how useful it is going forward once these pages have been "groomed".

--TomChatt 02:42, 18 December 2006 (MST)

Tom, In general I think you did an excellent job of summarizing what needs to be done for place names in Scotland. My only disagreement was in the section on towns and cities, to include:

  • Formerly located in: <parish> from <blank> to 1854
  • Formerly located in: <traditional county> from 1855 to 1975

Even though record keeping shifted from the parish churches to the civil government in 1855, towns were still located in parishes and civil records were organized by parishes. And, of course, in some of the bigger cities the parishes were sub-units of the city, as opposed to smaller places were towns were sub-units of the parish.

--Lauren 08:11, 18 December 2006 (MST)

Good points. Nearly all of my Scotland work has been pre-1854 and in rural parishes, so it's great to get other perspectives added. (That's how the wiki works!) My cursory understanding of the shift to civil reg in 1855 is that there became "registration districts" under the Government Registry Office (GRO), which used the same names and boundaries as the parishes for the next 100+ years, but underwent some consolidation/reorg in the 1960s/70s. And of course there's the Church of Scotland's point of view that places that were in the parish still *are* in the parish to this day.
Lauren you're of course right about the bigger places vs smaller places. Cities may indeed be bigger than parishes. I think the present City of Aberdeen contains all of Aberdeen (St Nicholas), Old Machar, and several other parishes (though some it's only pieces and not the whole parish that are contained).
There are a couple wiki-technical issues we should get clarified by Dallan. The current Place template allows only one "Located in" with date range, and multiple "Previously located in" with date range (as well as multiple "See also" places without a date range). I believe that both the main "Located in" and the "Previously located in" items are used to do the automatic creation of place lists on the containing Place pages. One issue is whether the date ranges are expected to be non-overlapping. A related issue is whether a Place may be "Located in" more than one place at the same time. It may be argued that a Place may properly be "Located in" more than one other Place at the same time (e.g., a town may be located in a parish and also located in a county). It may be better to revise the template slightly, so that "Located in" allows multiple entry lines (same format as "Previousy located in" field now), do away with the "Formerly located in", and whether a location is "current" or "former" can be deduced from the date ranges given when needed. (Note that some places, such as the Traditional counties, don't exist anymore, and so are only "previously located" and not currently located at all!)
Given that, should we say of parishes that they are a "previous" location, with an ending date of abt 1970? Or should we take the view that parishes are a current location? For the traditional counties, one can probably point to various dates for specific counties as the date they began, but I think it's fair to say they all stop in 1975. At the moment, I think the dates don't have any special use, although they may in the future. (And if Dallan thinks the above is a good idea, the presence or absence of a closing date would be used to determine whether something shows up as a current or previous "Located in" when displaying the automatically generated box on the right of the Place page.)
--TomChatt 11:45, 18 December 2006 (MST)

"Located in" places

To make matters even more confusing, some parishes were divided between more than one county. At one time the parish of Arngask was divided between Fife, Kinross, and Perth! I know they did some shifting around at various times to try to get the whole parish in a single county.--Lauren 11:53, 18 December 2006 (MST)

Oh! I hadn't run across that myself, but I was wondering if there were parishes like that. All the more reason to allow multiple "Located in" entries. Probably best to show it as "Located in" all three counties, so that researchers see all the places they might need to look. --TomChatt 16:58, 18 December 2006 (MST)
When I originally created the place index, the "located in" field was intended to be used to derive the "standard" name of a place, while "previously located in" was intended to provide alternate names. (See also places were not intended to be used by the system; just looked at by users.) If town X used to be located in county Y but was currently located in county Z, then if someone were to enter a place of "X, Y, Scotland," it would be standardized automatically to point to the place "X, Z, Scotland." And if someone were to search for a place "X, Y, Scotland," the search would also include results for "X, Z, Scotland." We're currently changing things so that the "standard" name of a place is represented in the title of the place page. This means that once we have renamed the place page titles to their "standard" names, we could conceivably merge "located in" and "previously located in" into a single "located in" field with multiple entries and an optional date range for each entry, and we could determine whether a place displays as a current or previous contained place of its "located in" place by whether or not there is something in the "end date" corresponding to that "located in" entry as mentioned earlier. Alternatively, we could rename "located in" and "previously located in" to something like "located in" and "other located in" to better represent that "other located in" could include alternate/additional "located in" places, and not just "previously located in" places.
The date ranges can be overlapping. They're not used currently. I was thinking that in the future (1-2 years from now), the date rnages could be used by a "research guidance" module to recommend sources to people. That is, if someone believes their ancestor was born in "X, Z, Scotland", and that they were born during the time that X was located in Y, and that the Y jurisdiction recorded the birth records, then the research guidance module could recommend vital records sources for "Y, Scotland." This will only be possible though for countries where people take on a project like what you're proposing here, because we just don't have enough date ranges otherwise.
Ideally, I would like to have just one place page for each place, even if that place was concurrently located in multiple containing jurisdictions. This will help when we do record linkage (determining whether two person pages are in fact for the same physical person). Having the systm tell people when they have created a wiki page that is likely to be for the same person as an existing wiki page so that they can look at the other page and decide whether or not to merge, is important I think. I'm not sure how you would name those place pages though. Would you pick the main containing jurisdiciton and use that in the place page title, and leave the others as additional entries in a combined "located in" field or in an "other located in" field?--Dallan 23:23, 18 December 2006 (MST)

Programming support

I think this is a terrific project and am ready to give whatever programming support I can. For example, it's easy for us to generate various reports. See for an example report of "suspicious jurisdictional paths" for Scotland. Also, we have the ability to make mass edits if needed (e.g., set the "located in" place to Y for all places where it is currently X). Just let me know what you need.--Dallan 23:23, 18 December 2006 (MST)

Getty References

One possible reason to keep the getty refs is that if we decide later to incorporate additional information from getty, the getty refs make it possible to attach the additional information to the existing places. We're planning to use the getty refs when we add places for cemeteries in the US for example -- we'll know what "located in" place to set for each cemetery by using the getty ref. Another reason is to provide a source for the information like alternate names and latitude/longitude that we obtained from getty. However, if you're going to be re-creating a place from scratch I don't think keeping the getty ref is essential.--Dallan 23:23, 18 December 2006 (MST)

Scottish Places Reorg - Step 1

Okay, I've now gone and visited all of the pages for the 33 "traditional counties", and standardized them to be "Located in" Scotland from blank to 1975. This causes all of them to be properly listed under "Traditional county" on the Place:Scotland page, as one would like.

I have noticed that there are a couple of duplicate pages, which I'll need Dallan to delete, I think. (If I'm able to do it myself, I don't know how.) There is a Place:Moray, Scotland page and also a Place:Morayshire, Scotland page. The latter one is redundant, and should be deleted. It contains no Place pages (everything points to the Moray page as far as containment goes), and has only a handful of pages that link to it. Ideally, we patch the handful of links and then delete the page. Similarly, we have both a Place:Perth (traditional county), Scotland and a Place:Perthshire, Scotland page. Here we should ditch the former and keep the latter. The "Perth (traditional county)" page has no links to it, except a bunch of Place containment that should get patched. If there's no nifty automated way to do the patching, I'm willing to do the tedious work. I think the patching needs to happen before we delete those pages, so that we can find what needs patched.

You need to be a sysop in order to delete pages. This gives you other rights as well, such as undeleting deleted pages, the ability to block users/IP addresses from editing pages, protect specific pages from being edited, and edit protected pages. So there is a responsibility to use these additional rights wisely. I think you've been doing a great job, so if you would like to be a sysop just let me know. Lauren, that goes for you as well. If you're going to be involved in fixing a lot of Scotland places, I think it makes sense for you to be a sysop.--Dallan 13:59, 21 December 2006 (MST)
Regarding deleting pages, before deleting a page you need to click on the "What links here" link at the bottom of the page to make sure that nothing points to it. This goes for sources as well - we'll need to edit the source pages to point them to the correct place. If there are more than just a few (say 20-30) places or sources that point to a page to delete, let me know and we'll write a robot script to point them to the correct page. The person who writes the robot scripts is getting married next week though, so we won't be able to run any scripts until he gets back mid-January.--Dallan 13:59, 21 December 2006 (MST)

The next step would be to do the same with all the present-day unitary council areas. An issue arises what to call those pages when the council area is (more or less) the same as the traditional county. We do need a separate page with a type of "Unitary council area" if we wish it to show up on the list under "Scotland". But in those cases, the page should do little else but refer to the traditional county page.

Note that I haven't done anything much about the content on those pages. I've just standardized the containment relationship with Scotland. I've got some ideas about organizing the content on traditional county pages which I'll come to shortly. --TomChatt 02:30, 21 December 2006 (MST)

What should appear in the place drop-down list?

The discussion in this section is excellent and has clearly gone beyond Scotland. I've moved it to the more general WeRelate talk:Place page titles page.--Dallan 14:35, 28 December 2006 (MST)

Parish Place naming

I'm about to try create/cleanup place pages for the parishes of Aberdeenshire, and I've quickly realized that we have a naming issue, in that most of the parishes have a town of the same name. So, we'll really want a separate Place page for the town and for the parish, since the town is a more specific and distinct place than the parish. So here are the options I see:

  • Option 1 - distinguish the parish
    • Place:Birse (parish), Aberdeenshire, Scotland -- for the parish of Birse
    • Place:Birse, Aberdeenshire, Scotland -- for the "inhabited place" called Birse
This option seems most consistent. Birse the "inhabited place" is named the same as any other inhabited place in the parish (e.g., Place:Finzean, Aberdeenshire, Scotland, which is in the parish of Birse).
  • Option 2 - distinguish the town / inhabited place
    • Place:Birse, Aberdeenshire, Scotland -- for the parish of Birse
    • Place:Birse (inhabited place), Aberdeenshire, Scotland -- for the "inhabited place"
This option gives preference to the parish name over the town name, and in most (but not all) cases, the parish may be the one you want more often. (Then again, my experience in that regard may not be representative.)
  • Option 3 - distinguish them both
    • Place:Birse (parish), Aberdeenshire, Scotland -- for the parish of Birse
    • Place:Birse (inhabited place), Aberdeenshire, Scotland -- for the "inhabited place"
This option may look best in a drop-down menu, as it clarifies the choice being presented.

For distinguishing the town, the parenthetic notation could vary as appropriate. In many cases "(kirkton)" may be appropriate, though some places (like Birse) may be too sparse to really call a "town" or "kirkton" and other places (like Aberdeen) are full-up cities. (Note that in many cases, the town is formally called "Kirkton of Bourtie", so the proper name clearly distinguishes it from "Bourtie", the parish.)

In option 2, the distinguishing tag (e.g., "(inhabited place)") would only happen in the cases where it was needed. In options 1 and 3, we'd want to consider whether to put the "(parish)" tag in all cases, or only where it was needed. (My feeling is that it would be needed in enough cases that people would see it often, and may be confused when some parishes don't have it. So I'd vote for always putting it on.) There's also the issue of whether we should name it "Birse (parish)", "Birse Parish", or "Parish of Birse".

A related issue is the county bit. As discussed on Werelate_talk:Place page titles, we're going in general for "fully qualified" place names, so for parishes, I'm thinking we should name them "parish, traditional county, Scotland". I'm aware that there are a few oddball parishes that straddle county boundaries or have been jiggered between counties over time. My recommendation for those is that we should pick one county for the name (based perhaps on how the GRO has mapped it), but include the other counties in the "See Also" field for that parish. Additionally, we should create page names for the parish with the alternate counties that redirect to the one we've chosen to be the actual page.

For towns, should we include the parish in the fully qualified page name? (E.g., Place:Finzean, Birse, Aberdeenshire, Scotland) Or is that too much? Note that for some towns with common names, we will need to do that to distinguish them. So should we include parish for all towns, or only those where it's needed? --TomChatt 16:13, 30 December 2006 (MST)

Unskilled Labor

So for those of us who have no familiarity with Scotland, and would like to assist without causing additional problems for you, what should we do? I've read through the proposal above, but the remainder of this page seems to second-guess it in places, so I don't want to make the matter worse. --Joeljkp 13:06, 11 January 2007 (MST)

Thanks for your kind offer. I've been busily trying to get things into a state where we can make use of such volunteering. I'd like to get a few good examples - a "model" county page, parish page, etc - that I can point people to, as well as some detailed instructions on ways volunteers can help. I wanted to work through a whole county myself to see what issues I'd run into, and get a more solid sense of what worked well and what didn't. I'm thinking by later next week I should have Aberdeenshire in decent shape, as well as the "how you can help" instructions. Thanks again, and stay tuned! --TomChatt 01:43, 12 January 2007 (MST)
Tom, I'm starting on Fife, Kinross-shire, and Perth. Right now I'm moving the parishes to the "parish, county, country" place name format and making notes on other information I would like to add, such as Old Parish Register availability dates, villages and towns, etc. --Lauren 06:14, 12 January 2007 (MST)
Lauren, excellent, your help is most welcome. I've just now sorted out the mess of the rival pages "Perth (traditional county), Scotland" and "Perthshire, Scotland", both of which represented the traditional county. I've remapped a boatload of towns and sources, making everything point to "Perthshire, Scotland", and changed the "Perth (traditional county)" page to just redirect to the other. Still all the work left to be done inside of Perth, but I wanted to at least get the county level pages sorted out. --TomChatt 16:29, 14 January 2007 (MST)

It might be helpful to use the GENUKI Church Database. The url is:

This nifty site shows the nearby churches, and places them on the various maps available, including multimap which gives the longitude and latitude.

How You Can Help

I've written up an overview of how place pages are (ought to be) organized under Scotland, and a how you can help page where there is a sign-up for "adopting" a county as well as step-by-step instructions for how to get a county sorted out. Comments and volunteers welcome! I've also taken a stab at putting some initial content (beyond the stuff copied from Wikipedia) on the Scotland page. --TomChatt 01:57, 18 January 2007 (MST)

Scotland and the great Fall '07 WeRelate place review [14 November 2007]

Dallan left this message on my user talk page:

Hi Tom, I wanted to let you know that I'm thinking about running an automated process to add "previously-located-in" entries to all places with an FHLC link that corresponds to where the FHLC says the place was located in. When I did the initial merge, I unfortunately often threw this information away. It's been critical to get this information back for places like the Czech Republic and Wales so I've written a program to automatically update the place places. I thought I'd run it over Scotland as well. In preparation for this I'm swapping the located-in and previously-located-in places where the previously-located-in field is actually the current located-in place -- the program expects this. (I've modified the program to recognize your link-fhlc template.) Once I get the list of updates to be made I'd like to show them to and ask you to check to make sure they look reasonable before I actually apply them if that's alright.--Dallan 15:13, 7 November 2007 (EST)

Hi Dallan, I think that a located-in/previously-located-in would be pretty dodgy at this point. Our approach to Scotland is that places should be primarily "located in" where they were traditionally located, which means that the way we've been grooming places, the primary "located-in" will correspond to where the FHLC has it. The FHLC follows the traditional Scottish place organization, whereas the Getty follows the modern one. In pretty much every case where the FHLC and the Getty differ, we'll follow the FHLC. (The only possible issue between what we're doing and the FHLC is that for the name of the county, the FHLC tends to use the "shire-less" name, while we're tending to use the "-shire" name. For example, what FHLC calls "Stirling, Scotland", we are calling "Stirlingshire, Scotland". The "shire-less" form will be included as an alternate name, so you should be able to match up FHLC place names to WeRelate place names.)
Although this varies from what we had originally been proposed for Scottish places, I am becoming more and more convinced that the best thing to do with the modern container-places is to ditch them altogether. The modern container-places didn't exist until 1975 or later, and I'd venture that there are NO records or repositories of genealogical interest organized by the modern containers. The problem is that many of the modern containers have the same or similar names to the traditional places, but aren't the same thing, and thus cause an awful lot of confusion.
I really can't believe the Scots made such a jumble of things! For those Americans not familiar with the mess that is Scottish place names, imagine this: Imagine that in 1975, the 50 states were eliminated and replaced by 12 "regions" and 84 "districts", and then in 1996, they did away with that and replaced it with 36 "unitary authorities". Some states survived kind of intact, like "New York", which was a district in the New England region, and then became the New York unitary authority. Other states got merged, like "Iowa and Dakotas" unitary authority, others got split, like North California and South California, and others got hacked and merged, like the "Mid-Atlantic" unitary authority, which comprises what was Maryland, Delaware, most of New Jersey, and the eastern half of Pennsylvania. Now consider an automated process that constructed a set of place names, drawing both on the FHLC which used the traditional states, and the Getty, which uses the modern structure. So of course, many things are in there in a redundant (but non-obviously so) manner, like "Dover, Delaware", and "Dover, Mid-Atlantic". That gives you an idea of the mess that we have in WeRelate's Scotland place pages! --TomChatt 02:09, 8 November 2007 (EST)
I have to second Tom on this one! From a genealogy standpoint, parishes are THE most important level for information. Almost of the pre-1855 church records are listed by parish, and the vast majority of the post-1855 civil registration records are arranged by parish as well. 99% of the Scottish events that will be listed on person and family pages will be recorded in the form of: possibly town, parish, traditional county, Scotland. The post-1975 and post-1996 place arrangements are a real mess to deal with, but most importantly, none of the genealogy information that people will be looking for is arranged at that grouping level. Whatever we do about the currently and previously located in sections I think it is important to have the page title reflect the pre-1975 location. --Lauren 09:42, 8 November 2007 (EST)

I also moved

Place:East Ayrshire, Scotland and Place:South Ayrshire, Scotland out from under Dumfries and Galloway, Place:South Lanarkshire, Scotland out from under Scottish Borders, Place:City of Edinburgh, Scotland out from under Midlothian, Place:Argyll and Bute, Scotland and Place:Aberdeenshire, Highland area, Scotland out from under Highland area, and put them all directly under Scotland. I hope that's correct.--Dallan 16:06, 7 November 2007 (EST)

Almost all of those are those confusing modern places. I'd prefer to see "East Ayrshire", "South Ayrshire", "South Lanarkshire", "Dumfries and Galloway", "Argyll and Bute" all deleted, and just keep the traditional counties of "Ayrshire", "Lanarkshire", "Dumfriesshire", "Argyll", "Bute", etc. I had originally thought that it would be useful to show that the traditional county (e.g., "Argyll") as the primary "located-in", with a date thru 1975, and to show the "district" of "Argyll and Bute" as a previously-located-in from 1975-1996, and "Highland area" unitary authority as a "previously-located-in" from 1996-. However, I'm becoming convinced that there is little genealogical value in keeping WeRelate place pages for those modern containers, and a lot of potential for confusion. I really think we're better off to get rid of them.
Also, note that the Aberdeenshire page that you moved represents the "unitary authority" of Aberdeenshire, which is similar but not the same as the traditional county of Aberdeenshire. Now, if you look on the Place:Scotland page, you'll see two Aberdeenshires! You can see the potential for confusion, and why I'd like to ditch those modern doppelgangers! (I think you'll also see why I liked to have the list of contained places sorted by place type.) --TomChatt 02:09, 8 November 2007 (EST)

Another question for you -- do you think we should put Place:Cromartyshire, Scotland and Place:Ross-shire, Scotland directly under Scotland, and add a see-also link to Place:Ross and Cromarty, Scotland? Were they ever located in Ross and Cromarty?--Dallan 16:44, 7 November 2007 (EST)

Re Cromartyshire and Ross-shire, those are ancient counties which became merged into the traditional county of "Ross and Cromarty". If it's useful to keep those pages (I'm not sure), they could go under "Ross and Cromarty". That was an interesting case. When Ross and Cromarty were separate, the territory of one was like Swiss cheese and the other was like the holes! For people researching in those areas, it would be useful to know whether a given village was anciently in Ross or in Cromarty, but perhaps it's better just to leave that to the descriptive text for the town, rather than having explicit Place pages for Ross-shire and Cromartyshire. Unless there are notable genealogical resources that belong distinctly to the two ancient counties, in which case I would keep their Place pages, and show parishes as "also contained in" the appropriate ancient county. That's the sort of thing I plan to sort out after I get all of the parishes sorted.
At the moment, we have proper pages for all of the traditional counties, and I am in the process of making sure all of the ~900 parishes (the next level of "container" for Scottish places) are properly located in their traditional county, and with redundant pages deleted, merged, or otherwise properly aligned. Fortunately, some of the work was already done. Lauren Snellgrove had done Perthshire and Fife (which are both large counties) as well as parts of a few other counties, and I had done Aberdeenshire (which is large), Banff, and Kincardineshire. I've been systematically going through the parishes for the remaining counties, and am probably 2/3 complete. If we can get the parishes into the right traditional counties before the automated renaming, then we'll be in decent shape. We'll still need to get towns into the right parishes, but at that point, we shouldn't have to rename any of the "container places". --TomChatt 02:09, 8 November 2007 (EST)

Thursday morning update: probably about 80% done with the parishes, hope to finish tonite. Looking at the Scotland placelist, it's looking a lot less unruly than it did when we started a few weeks ago!

These modern containers still contain places, which might be automatically patched up:

  • East Ayrshire -> Ayrshire
  • Ross and Cromarty (general region) -> Ross and Cromarty [traditional county]
  • South Ayrshire -> Ayrshire
  • South Lanarkshire -> Lanarkshire
  • Wigtown [DELETE-ME] -> Wigtownshire

These modern contains still contain places, but will require manual review to sort out, as they need to sort into more than one traditional county:

  • Argyll and Bute -> two counties
  • Dumfries and Galloway
  • Falkirk -> Stirlingshire (mostly, part in West Lothian; not too many, maybe leave these to manual review?)
  • Highland area -> covers most of five traditional counties, just gotta fix these manually
  • North Ayrshire -> part Ayrshire, but the island of Arran (which is half the area) goes to Bute, so manual review here
  • North Lanarkshire -> part Lanarkshire, part Dunbartonshire, so manual review
  • Perth and Kinross -> both counties
  • Scottish borders -> covers parts of several traditional counties
  • Stirling (unitary auth) -> all of Stirlingshire but also a fair chunk of Perthshire--TomChatt 11:40, 8 November 2007 (EST)

I can at least automatically sort out places in these counties with a source of FHLC by listing them as previously-located-in the containing place according to the FHLC. That process has been automated and applied to other countries already. Tomorrow I can send out the list of updates it would make.--Dallan 17:03, 8 November 2007 (EST)

The question was raised whether we should include islands in the primary naming hierarchy. At one point, I thought it was a good idea, especially for the large islands like Lewis and Skye, which contain many parishes. However, in other cases, a parish will contain several islands. And in some cases, they're split (eg a large island has several parishes, but those parishes may take in other smaller islands too). Maybe it's best to keep strictly to "town, parish, county, Scotland" for the naming structure, and put island information as a "see also" container. I think it's important to capture the islands, because for island areas, that may be a stronger "place identity" than the parish (i.e., it's more likely to be remembered that Great Grandpa came from the Isle of Lewis, than the particular parish). Any descendants of the isles out there? Opinions?--TomChatt 11:46, 8 November 2007 (EST)

Here's what I've been thinking for place titles; hopefully we can sort this out together.

People will probably want to see places listed approximately according to the hierarchy that existed when an event that they want to record occurred. If we omit unitary authorities (which I'm not completely against), people will probably re-create them over time as they record death events for their near relatives. So I'm not sure how much we'd gain by omitting unitary authorities. It seems like we'll have to figure out how to work with them somehow. For example, suppose I had a grandfather born in Alness in the old Ross county in 1880, married in Alness under Ross and Cromarty county in 1905, and died in Alness under Highland area in 1996 (he lived a long time). Let's suppose for the sake of this example that we had separate county pages for Ross-shire and "Ross and Cromarty". What's the best way to handle place pages for these three events?

1. We could create (or allow users to create) separate place pages for Alness under Ross county, Ross and Cromarty county, and under Highland area, with see-also links between them. We already have a lot of this kind of duplication in Eastern Europe due to the changes that happened there after WWII. Pro's: this provides an easy-to-understand structure, and each page can have its own history separate from the other page. Con's: we have to repeat information (like latitude and longitude) on each page.

2. We could (automatically) create redirect pages from older places to newer places based upon the previously-located-in field. So "Alness, Ross-shire, Scotland" could redirect to "Alness, Highland Area, Scotland" and "Alness, Ross and Cromarty, Scotland" would also redirect to "Alness, Highland area, Scotland". The county pages for "Ross-shire, Scotland" and "Ross and Cromarty, Scotland" would list Alness as a (historical) contained place, and all three Alness pages would show up in the drop-down box when someone typed "Alness" in a place field, with the two redirects showing that they redirected to the non-redirected page. Pro's: Just one page to update and put research tips on. Con's: most people would probably expect to see the non-redirected place page listed with a title of "Ross and Cromarty", not "Highlands area".

3. Similar to option 2, but create redirects from "Alness, Ross-shire, Scotland" to "Alness, Ross and Cromarty, Scotland" and from "Alness, Highland area, Scotland" to "Alness, Ross and Cromarty, Scotland". Pro's: now the non-redirect place page has the most-common title. Con's: Is it confusing to have redirects from older to newer places and also from newer to older places? What if some countries decide to have some redirects from newer to older, while others decide to always redirect older to newer?

My preference would be for option 2, but I could be convinced to switch to option 1 or 3. What would we need to do to make option 2 more satisfying? (For example, we could rename same-named unitary authorities to "X (unitary authority)" to distinguish them from traditional counties.)

--Dallan 17:03, 8 November 2007 (EST)

Of these, I like option 1 the least. I think that page duplication is a bad idea. Alness was Alness is Alness, regardless of how the administrative "containers" got reorganized around it over time. If we have duplicate pages for the same place, we reduce the opportunities for collaboration. Someboby might add Alness information to one page that really should go on all three. If I link an ancestor to one Alness and you link your ancestor to another, we lose the opportunity to see that we're working in the same area.
Another complication to keep in mind: there are towns and villages under the parish. Alness is a bad example, as it's a rural parish and there's not much in Alness parish besides Alness town. But in Birse (an Aberdeenshire parish I know well), you'll find:
  • Birse, Birse, Aberdeenshire, Scotland
  • Ballogie, Birse, Aberdeenshire, Scotland
  • Finzean, Birse, Aberdeenshire, Scotland
  • Marywell, Birse, Aberdeenshire, Scotland
So if we had distinct pages for each parish under its various historical containers, we'd not only need to have three Birse parish pages:
  • Birse, Aberdeenshire, Scotland
  • Birse, Kincardine and Deeside, Grampian, Scotland
  • Birse, Aberdeen (unitary authority), Scotland
but we'd also need to have at least four distinct village Place pages under each of the three Birse parish pages, so twelve pages for four villages in the one parish. Having multiple villages in a parish multiplies the disadvantage described in the previous paragraph.
If forced to choose between option 2 and option 3, I'd definitely take option 3. I think you just need to pick an arbitrary point in time as your canonical reference, and for Scottish genealogy, I think the most useful consensus is "where was Alness in 1891?". (1891 was a magic year, in that they did a boundary clean-up. Not a turn-everything-upside-down reorg as in 1975 and again in 1996, but just a tidying up, eliminating detached bits, putting parishes cleanly in a single county, etc.) This is the choice that FHLC made, and the choice that GENUKI made. It would be obtuse to go with the current system just because it's current, when the overwhelming body of genealogical resources are organized along 1891 lines. (As a counterpoint, JewishGen made the opposite choice to go with modern names in their database of Jewish ancestral towns, and boy do people complain, get confused, and miss opportunities to connect because of it!) If some of the redirects go "forward" and some "backward" in time, so be it. I truly believe it will cause the least confusion to do it that way.
My preference would be for an option 4, which is to not have the post-1975 constructs at all (not even as redirects), or to minimize their interaction with the other place pages (e.g., just as a "see also" or "also contained" for parishes). If we created redirects for all the variant containments for each parish, if the redirects show up as options in the pulldown place menus, I think most people who didn't have a PhD in Scottish geography would be confused by having to choose between "Alness, Ross-shire", "Alness, Ross and Cromarty", and "Alness, Highland area". (I faced those kind of choices myself when I first started using this wiki, and it's taken me a long time to get to the point where I understand why those choices might be there.) Keep in mind that the confusion gets much worse when you hit one where there are actually several towns in Scotland with the same name, so instead of choosing between the three real Kilbrides, you'd be confronted with nine (not counting the East Kilbrides and West Kilbrides)!
I think the best way to deal with the other historical containers is not in the Place page hierarchy, but just in the text on the one canonical page for "Alness, Ross and Cromarty". There you can explain that Alness was in Ross-shire before the two counties were united in 1891, and that you'd find it in Lochaber district for 1975-1996, and in the Highland council area after 1996.
--TomChatt 00:47, 9 November 2007 (EST)
Let's keep thinking about this. You make some good points. I didn't know about JewishGen. I've decided to review every European country over the coming week and compare the place hierarchy that we have with to the one that's in FHLC to see how different they are. So far, the UK has made much more significant changes than France or Spain for example. Let's keep thinking/talking about this and gathering data. In the meantime, I think it would be best if we could put the "latest" place in the located-in field and put earlier places in previously-located-in. This will make things easier for the automated programs. If we decide to go with options 3 or 4, we can do that automatically during the renaming.
Also, here's a list of automated place updates for Scotland. By placing towns inside their parishes, you've made enough improvements to the FHLC structure that it may not make sense to automatically apply these updates in Scotland's case. You may just want to use this list as something to review in case there's anything that was overlooked. Just let me know whether you'd like them to be applied.--Dallan 12:30, 10 November 2007 (EST)
I took a look at the list of automated place updates, and it looks like the bulk of the updates would be unhelpful. As you observed, it would generally move towns out from their parishes where we've put them in. Also, it seems to be confused about matching some of the counties to the right Place page. I don't think we should apply these updates. --TomChatt 18:46, 10 November 2007 (EST)
That's what I expected. Could you tell me what counties it is confused on? This may mean that we have the wrong FHLC id on the county page. I'd like to make sure that we have the right FHLC id's on the county pages, because when we re-work the source pages next year I plan to use FHLC id for linking FHLC sources to the right place page. Thanks --Dallan 23:45, 12 November 2007 (EST)
Looking at the update list, it seems to be confused by the various incarnations of Ross and Cromarty. In terms of the county pages having the right FHLC links, I've just verified that. Dumbarton (the parish) and Dunbartonshire (the eponymous county) had their FHLC id's swapped, but I fixed it. All the other counties were fine. --TomChatt 03:14, 14 November 2007 (EST)
Thank-you! I'm just about done with my review of some of the other European countries. Over the weekend I plan to write up a proposal for handling countries that have had a lot of recent changes, but I wanted to let you know that I'm leaning heavily toward your suggestion of putting things into their historical structure when significant recent changes have occurred.--Dallan 12:36, 16 November 2007 (EST)

traditional county [4 feb 2013]

I do not have sufficient knowledge of Scotland, but I understand the WeRelatesystem well. I feel that the provinces/county are not tight on a date organized. I have added a map and it's not possible for me in the provinces/county there is a tight turn. --Lidewij 09:55, 4 February 2013 (EST)

(Ik heb niet voldoende kennis van Schotland, maar ik begrijp het WeRelatesysteem goed. Ik heb het gevoel dat de provincies/county niet strak op één datum georganiseerd zijn. Ik heb een kaartje toegevoegd en het is daar bij voor mij niet mogelijk om de provincies/county er strak in te zetten. --Lidewij 09:55, 4 February 2013 (EST)