Mapping Facility in FH


Robert Jordan
 

 

There have been a number of different threads on this site concerning the Maps facility in FH. As this is a slightly different angle I decided to start a new thread.

Some people are trying to merge Places to be consistent and we now learn that the AI process of geocoding really only recognises the latest Place automatically and not historical places.

I give some examples.

Early in 1974 I lived in a place Keynsham, Somerset. On 1 April that year when I woke up I found that without moving house I lived at a Place Keynsham, Wansdyke, Avon.  Some time later, after I had actually moved, Avon and Wansdyke were abolished and my old Place became Keynsham, Bath and North East Somerset, Somerset. So in the space of 20 or so years the Place had changed twice.

Keynsham itself is interesting. Whilst being in Somerset I know for a fact that some of my ancestors were born and married in Mangotsfield, Gloucestershire some 10 or so miles from Keynsham. If however one looks up the event on the BMD website it shows that the Registration event took place in Keynsham. It seems this was as a result of the Poor Law of 1834 which designated Keynsham as the poor law Union for 19 Parishes or Places 14 of which were in Somerset and 5 in Gloucestershire.

Another Place is Tilehurst, a parish, which before April 1974 would have been Tilehurst, Berkshire (actually in the District of Bradfield). Then it changed to Tilehurst, Newbury Berkshire.  Later in 1998 The County Council of Berkshire was abolished it changed again. Part of Tilehurst became Tilehurst, Reading another part of Tilehurst became Tilehurst, West Berkshire (Newbury having changed its name to West Berkshire).

As a general rule if I find a document with a Place I will record it exactly as the document shows because historically that is the fact.

I would urge caution about using the latest geocoding when the primary task of a Family Historian is to record history not mapping.

Robert Jordan


Mike Tate
 

Yes, place names regularly change all over the world for political and administrative reasons.

Also, quite often the BMD formal GRO Registration District place in the records will be different from the actual BMD event place.

Automatic geocoders are designed for modern-day users and current modern place names.

They cannot cater for the needs of the relatively small community of family history researchers and support all historic place names.

That is why FH has a Standardized field and the Map Life Facts plugin has a Substitute field.

Those fields hold the modern place name to use for geocoding while the main Place name field holds the historic name for Reports, etc.

 

Mike Tate

 

From: family-historian@groups.io <family-historian@groups.io> On Behalf Of Robert Jordan
Sent: 08 January 2021 10:58
To: family-historian@groups.io
Subject: [family-historian] Mapping Facility in FH

 

 

There have been a number of different threads on this site concerning the Maps facility in FH. As this is a slightly different angle I decided to start a new thread.

Some people are trying to merge Places to be consistent and we now learn that the AI process of geocoding really only recognises the latest Place automatically and not historical places.

I give some examples.

Early in 1974 I lived in a place Keynsham, Somerset. On 1 April that year when I woke up I found that without moving house I lived at a Place Keynsham, Wansdyke, Avon.  Some time later, after I had actually moved, Avon and Wansdyke were abolished and my old Place became Keynsham, Bath and North East Somerset, Somerset. So in the space of 20 or so years the Place had changed twice.

Keynsham itself is interesting. Whilst being in Somerset I know for a fact that some of my ancestors were born and married in Mangotsfield, Gloucestershire some 10 or so miles from Keynsham. If however one looks up the event on the BMD website it shows that the Registration event took place in Keynsham. It seems this was as a result of the Poor Law of 1834 which designated Keynsham as the poor law Union for 19 Parishes or Places 14 of which were in Somerset and 5 in Gloucestershire.

Another Place is Tilehurst, a parish, which before April 1974 would have been Tilehurst, Berkshire (actually in the District of Bradfield). Then it changed to Tilehurst, Newbury Berkshire.  Later in 1998 The County Council of Berkshire was abolished it changed again. Part of Tilehurst became Tilehurst, Reading another part of Tilehurst became Tilehurst, West Berkshire (Newbury having changed its name to West Berkshire).

As a general rule if I find a document with a Place I will record it exactly as the document shows because historically that is the fact.

I would urge caution about using the latest geocoding when the primary task of a Family Historian is to record history not mapping.

Robert Jordan


Barry1936
 

"As a general rule if I find a document with a Place I will record it exactly as the document shows because historically that is the fact.

I would urge caution about using the latest geocoding when the primary task of a Family Historian is to record history not mapping"

Spot on Robert - a key 'raison d'être"  that is often forgotten


Robert Jordan
 

Thanks Mike you have made my point exactly.
If you find a BMD registration in Keynsham then it would be natural to show that as the birth, marriage or death Place. The Geocoding would presumably place a mark on Keynsham on the Map. If subsequently you find that the event took place in Mangotsfield in a different County Then some people might assume that Mangotsfield is in Keynsham. An understandable mistake to make unless you are conversant with every place in the Country and I doubt whether there are many who would profess to be that knowledgeable.
I do not know whether the Geocoding system being used would pick up the Place Mangotsfield, Keynsham, Somerset and would place a map marker on Keynsham or Mangotsfield or neither.
That is precisely why I advocate using the original document to record a Place and exercise caution in dealing with the mapping facility which is after all incidental to the main purpose of FH which is History.

Robert Jordan


Virus-free. www.avast.com


Robert Jordan
 

Thanks Barry 1936 for your comment and support

Robert Jordan

Virus-free. www.avast.com


Adrian Bruce
 

On Fri, 8 Jan 2021 at 15:41, Robert Jordan <robacis1@...> wrote:
...If you find a BMD registration in Keynsham then it would be natural to show that as the birth, marriage or death Place. The Geocoding would presumably place a mark on Keynsham on the Map.
 
Ah but what I always do in a case like that is show the BMD event as taking place in (say) "Nantwich Reg. District ,Cheshire, England". That makes it clear to a human reader that the event takes place *somewhere* in the Registration District - which is rather larger than the town of Nantwich. I'm pretty sure that any Geocoder will still mark the event as happening in the town of Nantwich, mind.

If subsequently you find that the event took place in Mangotsfield in a different County Then some people might assume that Mangotsfield is in Keynsham.

Another reason for writing "Keynsham Reg. District" to remind yourself that we're not talking about Keynsham itself. And if you ended up knowing that the birth was in Mangotsfield, I'd suggest that there is zero reason to continue referring to Keynsham, so it's **probably** not a significant problem. (County can come from the GENUKI Gazetteer, for instance, unless it's one of those places that even the Victorians "moved"...)
 
 
I do not know whether the Geocoding system being used would pick up the Place Mangotsfield, Keynsham, Somerset and would place a map marker on Keynsham or Mangotsfield or neither.
 
The answer is probably "Yes"!

That is precisely why I advocate using the original document to record a Place and exercise caution in dealing with the mapping facility which is after all incidental to the main purpose of FH which is History.

Indeed, and there are all sorts of excellent reasons to record a place using the contemporary source, as you suggest. Mind you, as I belong to the "Picture is worth a thousand words" school, I'd never consider the map as a mere incidental but potentially a vital part of telling the history. The issues to which you rightly refer are exactly why the Standardized version of the Place Name exists - stick today's version of the Place Name in there and you should be better off.

Adrian


 


Robert Jordan
 

Adrian
Your comments do not entirely make sense to me.
Firstly when I enter an event the registration district shows up in the source so I know what it is.

I know that Mangotsfield is in Gloucestershire and Keynsham is in Somerset is because I lived in the area. If the same situation occurred in say Northumberland then I would not have a clue and I think most people would take the same view as me. To look up some gazetteer to check every place name used would I suggest take the whole enjoyment out of studying family history.

I posed a question as to what marker would be used on the map for Mangotsfield, Keynsham, Somerset and gave three optional answers. Your answer was "probably yes"  I do not have a clue what your answer means!!!

I'm afraid you have not convinced me. The map facility was introduced in V6 and I can count on the fingers of one hand the number of times I have used it. From memory it was to look at the map for myself as I had travelled to all continents of the world, the second time was when I entered Banjol, Gambia in a place and found that the marker showed Banjul which is the correct spelling and when I changed my place entry to the correct spelling the marker jumped about 50 miles away!
This is not in my opinion a reliable facility and I for one will not be using it

Robert Jordan

Virus-free. www.avast.com


Adrian Bruce
 

"I posed a question as to what marker would be used on the map for Mangotsfield, Keynsham, Somerset and gave three optional answers. Your answer was "probably yes"  I do not have a clue what your answer means!!!"

OK - apologies if my slightly flippant answer confused but I was trying to convey that Geocoding from place-names - especially when the place-names don't actually make sense, such as "Mangotsfield, Keynsham, Somerset" - is not an exact science and that different Geocoding systems **will** come up with different answers. For instance, if you try to locate "Mangotsfield, Keynsham, Somerset" on a Google Map, you get sent straight to Keynsham Police Centre for Avon & Somerset Police. No, I've no idea why either. Try it on Bing Maps and it says "No results". So for any given mapping & geocoding system, while one of your options is probably the answer, it might not be the same one for all such systems.

"when I enter an event the registration district shows up in the source so I know what it is."
Evidently we work in different ways - you're happy to take the fact and go and look at the source to see what it means, whereas I'd prefer to just have everything in front of me in the diagram or report. That's why I advocate entering the Place Name as "Keynsham Reg. District" and not "Keynsham" if, and only if, all you have to hand is the BMD Registration Index.

As for checking place-names using Gazetteers, etc, well, normally you don't need that because it's there in the source to be copied into the fact. But if the BMD Index first says that the birth was registered in "Keynsham Reg. District, Somerset, England"and *then* a census says that they're born in "Mangotsfield, Glos" then, not knowing the area, I'd need to check the Gazetteer first to see whether the two values are in complete contradiction (which can happen, of course). If the Gazetteer tells me that Mangotsfield is in Keynsham Reg. District (say), then that's fine, I'll replace the previous place-name by "Mangotsfield, Gloucestershire, England". 

As for using the Map Facility - that's entirely up to you - I struggle with how to choose and then physically incorporate (frozen) maps into my Reports, so I'm not much of a one for using them as output myself.

I wouldn't say, however, that it's not reliable - provided I check what it's giving me (so I won't bulk encode places) and tweak the results where it's picked up the wrong place (e.g. by adding the modern equivalent as the Standardized name), then it's as reliable as any genealogy data source. Providing I do that check, it's fine.

In the end, mapping is a personal choice. Like many things.

Adrian




johnfirr@...
 

I feel slightly responsible for starting the passionate debate on geocoding as it was me that had a query on "tidying up". So probably worth sticking another ten pennorth in!.

I have found the debate on the various threads both interesting and educational. I have learned a lot so thanks to all contributors. My query about tidying up was not focussed on trying to get my place names aligned with modern geocoding systems as I agree with the sentiments in this thread and always like to record the place as written in the original document so for instance "London and Kent" as opposed to just London or Kent for more modern entries. My "tidying up" was a reflection of my own sloppy habits when first using FH. I ignored typos, used different formats ( ie with commas, without commas etc) and critically in the early days didnt include the country. I have learned better discipline since I grew up!.

To this end I have found the comments about using the standardised box to record the modern usage of places really helpful and will incorporate that into my way of working from now on.

Are maps useful. Well I agree that they are probably secondary but they probably have a more direct application for people like me who are using FH to compile data for a one name study as surname distribution analysis is quite a feature of my study. Having said that I agree that the facility should be used with caution but I welcome its inclusion and development as part of the software.

In conclusion, it is this forum and the FHUG that is a major reason for my choosing Family Historian as my software. I value the knowledge and energy of all the regular contributors and thankyou for your support whenever I have a query, I only hope that one day I become expert enough to help other users.
regards to all
John Firr


Robert Jordan
 

Dear all

Thanks for your contributions. Please note that I only said Users should exercise Caution when using the Map function.

Just to illustrate I will use my own record.
I was born in Bristol at The Bristol Royal Infirmary (although that information was not shown on my Birth Certificate) right in the heart of the city. Bristol itself has for most of its existence been independent of either Somerset or Gloucestershire. My birth was confirmed using BMD records. My Birth certificate shows my fathers address as Staple Hill, Bristol. This was commonly used as a postal address but is not accurate as Staple Hill is in Mangotsfield, Gloucestershire. To overcome this I have recorded a residence fact to show that Staple Hill is in Gloucestershire. I can do this because I know what is factually correct from experience and other documents, however if when researching someone from the 19th century I would only be able to rely on documents especially in localities not well known to me and so it would from my Birth Certificate appear that I was born in Staple Hill, Bristol. How Geocoding would deal with that is anyone's guess.

Recently I noticed that in my own record I had not included an Occupational event that occurred some 20 years ago. I had the Map window open at the time and entered the Place as Banjol, The Gambia. A map marker immediately appeared right over the top of Banjul, The Gambia. I then noticed that I had misspelt and corrected Banjol to Banjul. Then right before my eyes the map marker jumped about 50 miles to the east.

I know many people wish to embrace automated technology however I would still exercise caution.

I would thank Adrian for his contribution as he took the trouble to put the Place I gave him into two different geocoding software One came up with no result and the other Keynsham Police Station!!! I know my Grandparents were definitely not wed in a Police Station

Robert Jordan


Virus-free. www.avast.com


Adrian Bruce
 

I would totally agree with the mantra "Exercise Caution"... 

Adrian


David Potter
 

Evening All.

I have been reading the posts re Mapping in FH and Place Names in general with great interest. But I'm finding there is a clear tradeoff between keeping True to the Source and determining where on a Map (if it is resonably obvious) the Place is and should be marked so in FH.
If I take the UK Wales and England Census records, quite often the Places recorded in the Where Born column are best guesses by the Enumerator; and quite often even the Head of the Household is unclear about where born information. A good number are accurate references but many are not.

My method is to try to establish the most reasonabe determination on where the Place might be and mark it so in the FH Mapping feature along with the Geocode values. The reason for this is if I were to enter each place Verbatim as recorded in the Census record, I would have a mess of Place records with no clear logic behind the true (at least reasonable) location. So it is a compromise for me. Of course having resolved where I believe the place should be mapped I could go back and rehash the Place details as entered in the Source record itself, much helped with FH7 and Templated Source records and Rich Text support. But that is a lot of additional work and to what gain other that to be 100% True to the Source.

I still find the need to enter simplified Place data such as ,, Herefordshire,, England if that is all I have to go on, but it then becomes a common Place value used wherever that same need arises. I tend to choose the City within the County deemed to be the 'Capital' as a Map marker.

I appreciate one could use the Standardized Place Field to help this along but that is not the value used when you click on a Place in the Map Window and select Locate in Map. I'm always looking to apply best practice - but with Places it is often a compromise with what you wish to achieve:

Be absolutely aligned with being True to the Source, or
Have a more manageable/practical Place database.

Thoughts/Comments always welcome.

David


Mike Tate
 

David, you are mistaken.

If you use the Standardized field then that IS the place used for plotting in the Map Window.

That is its sole purpose.

 

From: family-historian@groups.io <family-historian@groups.io> On Behalf Of David Potter via groups.io
Sent: 09 January 2021 21:00
To: family-historian@groups.io
Subject: Re: [family-historian] Mapping Facility in FH

 

Evening All.

I have been reading the posts re Mapping in FH and Place Names in general with great interest. But I'm finding there is a clear tradeoff between keeping True to the Source and determining where on a Map (if it is resonably obvious) the Place is and should be marked so in FH.
If I take the UK Wales and England Census records, quite often the Places recorded in the Where Born column are best guesses by the Enumerator; and quite often even the Head of the Household is unclear about where born information. A good number are accurate references but many are not.

My method is to try to establish the most reasonabe determination on where the Place might be and mark it so in the FH Mapping feature along with the Geocode values. The reason for this is if I were to enter each place Verbatim as recorded in the Census record, I would have a mess of Place records with no clear logic behind the true (at least reasonable) location. So it is a compromise for me. Of course having resolved where I believe the place should be mapped I could go back and rehash the Place details as entered in the Source record itself, much helped with FH7 and Templated Source records and Rich Text support. But that is a lot of additional work and to what gain other that to be 100% True to the Source.

I still find the need to enter simplified Place data such as ,, Herefordshire,, England if that is all I have to go on, but it then becomes a common Place value used wherever that same need arises. I tend to choose the City within the County deemed to be the 'Capital' as a Map marker.

I appreciate one could use the Standardized Place Field to help this along but that is not the value used when you click on a Place in the Map Window and select Locate in Map. I'm always looking to apply best practice - but with Places it is often a compromise with what you wish to achieve:

Be absolutely aligned with being True to the Source, or
Have a more manageable/practical Place database.

Thoughts/Comments always welcome.

David


Adrian Bruce
 

On Sat, 9 Jan 2021 at 21:00, David Potter via groups.io
<David.potter5=googlemail.com@groups.io> wrote:
...
Be absolutely aligned with being True to the Source, or
Have a more manageable/practical Place database.

Thoughts/Comments always welcome.
...

My view about being "True to the Source" is that this applies
precisely and only to the Source-Record and to any quotes from it
(e.g. "Text from Source" etc.). The place-names held against facts
such as Birth or Residence are **conclusions** and are therefore free
to differ from the text in the Source. If there is a difference then I
**should** be justifying that difference somewhere - usually in the
Yellow-pane Note against that citation but potentially in a longer
note somewhere.

Examples of differences from the Source include:
- the 1851 census seems to have a number of my relatives who entered
their *parish* of birth, rather than the town / village. My 4G-GF
entered his birthplace as "Great Budworth" in the 1851 but "Northwich"
in later censuses. His 1851 Source Record says "Great Budworth" (I
split my census sources down to the household level). The birth fact
says "Northwich", cites the 1851 against that birth fact but records
in the Notes for that citation, something along the lines of "Gt
Budworth is the par for Northwich";

- the contemporary common name of the town and the contemporary
official name differ. One of my ancestral towns is the railway town of
Crewe. The BMD certificates usually refer to an event taking place in
"Monks Coppenhall, Cheshire" and so the Source Record for the Birth
(e.g.) says "Monks Coppenhall". However, I enter "Crewe, Cheshire,
England" for the fact because at the time that they were born, the
physical settlement was the railway town and it *was* known as Crewe.
I probably don't enter that in the citation but I do enter it as a
note against the place record for Crewe - something along the lines of
"Officially Monks Coppenhall for Civil Registration purposes etc.".
The original usage was "Monks Coppenhall" for the south and centre of
where the town was later built, and "Church Coppenhall" for the north.
However the term "Coppenhall" is today used to refer to the **north**
end of Crewe - e.g. "Monks Coppenhall Junior School" is actually in
the former "Church Coppenhall"!

- the original is ambiguous. Look, they knew which St. George's
parish they meant in the Bristol area. But the rest of us? I can
usually work it out - so I will tweak the name in the fact to
distinguish between "St. George, Gloucestershire" and "St. George,
Brandon Hill, Bristol, Gloucestershire". Again, the Source Record
quotes (or images) the exact text in the original and the citation
data should have something useful.

Now, all of those are potential differences between my conclusion and
the source-record. I haven't referred to maps because, in my book, the
maps do NOT influence the choice of name in my conclusion.

When it comes to looking at the maps and the geocoding of the places
from the facts, then, I must emphasise again, I will not alter the
place-names from the facts. What might happen is that the mapping goes
to an odd place. For example, my places in the pre-1901 Australian
colony of Victoria have been mapped to British Columbia.... Note that
my pre-1901 Victoria place names terminate in Victoria, NOT Australia.
I respect the contemporary political arrangement in my Australian
place-names in my facts. What I will do there is to add a Standardized
Place-Name of "XXX, Victoria, Australia" for the place "XXX, Victoria"
in its Place-record.

That Standardized Place-name is where any tweaking / fudging /
simplification / whatever, goes on in my book.

That's what I do... (Just don't ask about London!)

Adrian


David Potter
 

Um, Okay Mike very Interesting.

As the Map Geocoding is still not operational just now I can’t test this. But if what you are saying is by way of this example:

I have say Cardiff, Glamorgan, Wales in the Place field, and Canterbury, Kent, England in the Standardized Place field. And no Geocoding already exists (that field is Blank). If I click on the Cardiff in the List Pane and choose Locate in Map it would find Canterbury is that right?

 

Kind Regards

 

David Charles Potter

 

From: family-historian@groups.io <family-historian@groups.io> On Behalf Of Mike Tate
Sent: 09 January 2021 22:09
To: family-historian@groups.io
Subject: Re: [family-historian] Mapping Facility in FH

 

David, you are mistaken.

If you use the Standardized field then that IS the place used for plotting in the Map Window.

That is its sole purpose.

 

From: family-historian@groups.io <family-historian@groups.io> On Behalf Of David Potter via groups.io
Sent: 09 January 2021 21:00
To: family-historian@groups.io
Subject: Re: [family-historian] Mapping Facility in FH

 

Evening All.

I have been reading the posts re Mapping in FH and Place Names in general with great interest. But I'm finding there is a clear tradeoff between keeping True to the Source and determining where on a Map (if it is resonably obvious) the Place is and should be marked so in FH.
If I take the UK Wales and England Census records, quite often the Places recorded in the Where Born column are best guesses by the Enumerator; and quite often even the Head of the Household is unclear about where born information. A good number are accurate references but many are not.

My method is to try to establish the most reasonabe determination on where the Place might be and mark it so in the FH Mapping feature along with the Geocode values. The reason for this is if I were to enter each place Verbatim as recorded in the Census record, I would have a mess of Place records with no clear logic behind the true (at least reasonable) location. So it is a compromise for me. Of course having resolved where I believe the place should be mapped I could go back and rehash the Place details as entered in the Source record itself, much helped with FH7 and Templated Source records and Rich Text support. But that is a lot of additional work and to what gain other that to be 100% True to the Source.

I still find the need to enter simplified Place data such as ,, Herefordshire,, England if that is all I have to go on, but it then becomes a common Place value used wherever that same need arises. I tend to choose the City within the County deemed to be the 'Capital' as a Map marker.

I appreciate one could use the Standardized Place Field to help this along but that is not the value used when you click on a Place in the Map Window and select Locate in Map. I'm always looking to apply best practice - but with Places it is often a compromise with what you wish to achieve:

Be absolutely aligned with being True to the Source, or
Have a more manageable/practical Place database.

Thoughts/Comments always welcome.

David


Brian Lummis
 

More or less, David. I just tried it in the Sample Project and the pointer was just south of Canterbury!

 

Brian

 

From: family-historian@groups.io <family-historian@groups.io> On Behalf Of David Potter via groups.io
Sent: 09 January 2021 23:05
To: family-historian@groups.io
Subject: Re: [family-historian] Mapping Facility in FH

 

Um, Okay Mike very Interesting.

As the Map Geocoding is still not operational just now I can’t test this. But if what you are saying is by way of this example:

I have say Cardiff, Glamorgan, Wales in the Place field, and Canterbury, Kent, England in the Standardized Place field. And no Geocoding already exists (that field is Blank). If I click on the Cardiff in the List Pane and choose Locate in Map it would find Canterbury is that right?

 

Kind Regards

 

David Charles Potter

 

From: family-historian@groups.io <family-historian@groups.io> On Behalf Of Mike Tate
Sent: 09 January 2021 22:09
To: family-historian@groups.io
Subject: Re: [family-historian] Mapping Facility in FH

 

David, you are mistaken.

If you use the Standardized field then that IS the place used for plotting in the Map Window.

That is its sole purpose.

 

From: family-historian@groups.io <family-historian@groups.io> On Behalf Of David Potter via groups.io
Sent: 09 January 2021 21:00
To: family-historian@groups.io
Subject: Re: [family-historian] Mapping Facility in FH

 

Evening All.

I have been reading the posts re Mapping in FH and Place Names in general with great interest. But I'm finding there is a clear tradeoff between keeping True to the Source and determining where on a Map (if it is resonably obvious) the Place is and should be marked so in FH.
If I take the UK Wales and England Census records, quite often the Places recorded in the Where Born column are best guesses by the Enumerator; and quite often even the Head of the Household is unclear about where born information. A good number are accurate references but many are not.

My method is to try to establish the most reasonabe determination on where the Place might be and mark it so in the FH Mapping feature along with the Geocode values. The reason for this is if I were to enter each place Verbatim as recorded in the Census record, I would have a mess of Place records with no clear logic behind the true (at least reasonable) location. So it is a compromise for me. Of course having resolved where I believe the place should be mapped I could go back and rehash the Place details as entered in the Source record itself, much helped with FH7 and Templated Source records and Rich Text support. But that is a lot of additional work and to what gain other that to be 100% True to the Source.

I still find the need to enter simplified Place data such as ,, Herefordshire,, England if that is all I have to go on, but it then becomes a common Place value used wherever that same need arises. I tend to choose the City within the County deemed to be the 'Capital' as a Map marker.

I appreciate one could use the Standardized Place Field to help this along but that is not the value used when you click on a Place in the Map Window and select Locate in Map. I'm always looking to apply best practice - but with Places it is often a compromise with what you wish to achieve:

Be absolutely aligned with being True to the Source, or
Have a more manageable/practical Place database.

Thoughts/Comments always welcome.

David


David Potter
 

Thank you Brian. That's good to know. I need to think how to put this into use.

Wow what an eye opener.

Thanks.

On Sat, 9 Jan 2021, 23:11 Brian Lummis via groups.io, <lummis=btinternet.com@groups.io> wrote:

More or less, David. I just tried it in the Sample Project and the pointer was just south of Canterbury!

 

Brian

 

From: family-historian@groups.io <family-historian@groups.io> On Behalf Of David Potter via groups.io
Sent: 09 January 2021 23:05
To: family-historian@groups.io
Subject: Re: [family-historian] Mapping Facility in FH

 

Um, Okay Mike very Interesting.

As the Map Geocoding is still not operational just now I can’t test this. But if what you are saying is by way of this example:

I have say Cardiff, Glamorgan, Wales in the Place field, and Canterbury, Kent, England in the Standardized Place field. And no Geocoding already exists (that field is Blank). If I click on the Cardiff in the List Pane and choose Locate in Map it would find Canterbury is that right?

 

Kind Regards

 

David Charles Potter

 

From: family-historian@groups.io <family-historian@groups.io> On Behalf Of Mike Tate
Sent: 09 January 2021 22:09
To: family-historian@groups.io
Subject: Re: [family-historian] Mapping Facility in FH

 

David, you are mistaken.

If you use the Standardized field then that IS the place used for plotting in the Map Window.

That is its sole purpose.

 

From: family-historian@groups.io <family-historian@groups.io> On Behalf Of David Potter via groups.io
Sent: 09 January 2021 21:00
To: family-historian@groups.io
Subject: Re: [family-historian] Mapping Facility in FH

 

Evening All.

I have been reading the posts re Mapping in FH and Place Names in general with great interest. But I'm finding there is a clear tradeoff between keeping True to the Source and determining where on a Map (if it is resonably obvious) the Place is and should be marked so in FH.
If I take the UK Wales and England Census records, quite often the Places recorded in the Where Born column are best guesses by the Enumerator; and quite often even the Head of the Household is unclear about where born information. A good number are accurate references but many are not.

My method is to try to establish the most reasonabe determination on where the Place might be and mark it so in the FH Mapping feature along with the Geocode values. The reason for this is if I were to enter each place Verbatim as recorded in the Census record, I would have a mess of Place records with no clear logic behind the true (at least reasonable) location. So it is a compromise for me. Of course having resolved where I believe the place should be mapped I could go back and rehash the Place details as entered in the Source record itself, much helped with FH7 and Templated Source records and Rich Text support. But that is a lot of additional work and to what gain other that to be 100% True to the Source.

I still find the need to enter simplified Place data such as ,, Herefordshire,, England if that is all I have to go on, but it then becomes a common Place value used wherever that same need arises. I tend to choose the City within the County deemed to be the 'Capital' as a Map marker.

I appreciate one could use the Standardized Place Field to help this along but that is not the value used when you click on a Place in the Map Window and select Locate in Map. I'm always looking to apply best practice - but with Places it is often a compromise with what you wish to achieve:

Be absolutely aligned with being True to the Source, or
Have a more manageable/practical Place database.

Thoughts/Comments always welcome.

David


David Potter
 

Thank you Adrian. Some good examples given and the explanation of your process has been very helpful.

I just need to get my head around what adjustments I need to make.

David


John Hanson
 

Robert

Whilst Mangotsfield might be in Gloucestershire it is in Keynsham registration district
There are many instances of registration districts crossing the county boundary

There are also instances where the registration district is not an actual place – Cosford in Suffolk that crops up many times in my family

 

These days I resort to an online gazetteer to sort them out and first port of call is the LDS mapping facility at https://www.familysearch.org/mapp/


Enter a parish and it will give you all the information that you want about it and links to its wiki which is one of the most underused part of the site these days yet one of the most important with link to all online information for a parish

 

Regards
John Hanson - researching the Halstead/Holstead/Alstead names
Researcher, the Halsted Trust - https://www.halsted.org.uk
Research website - https://www.halstedresearch.org.uk

2021 Family History Conference - https://www.elizabethanancestors.org.uk

 

 

From: family-historian@groups.io <family-historian@groups.io> On Behalf Of Robert Jordan
Sent: 08 January 2021 18:13
To: family-historian@groups.io
Subject: Re: [family-historian] Mapping Facility in FH

 

Adrian

Your comments do not entirely make sense to me.

Firstly when I enter an event the registration district shows up in the source so I know what it is.

 

I know that Mangotsfield is in Gloucestershire and Keynsham is in Somerset is because I lived in the area. If the same situation occurred in say Northumberland then I would not have a clue and I think most people would take the same view as me. To look up some gazetteer to check every place name used would I suggest take the whole enjoyment out of studying family history.

 

I posed a question as to what marker would be used on the map for Mangotsfield, Keynsham, Somerset and gave three optional answers. Your answer was "probably yes"  I do not have a clue what your answer means!!!

 

I'm afraid you have not convinced me. The map facility was introduced in V6 and I can count on the fingers of one hand the number of times I have used it. From memory it was to look at the map for myself as I had travelled to all continents of the world, the second time was when I entered Banjol, Gambia in a place and found that the marker showed Banjul which is the correct spelling and when I changed my place entry to the correct spelling the marker jumped about 50 miles away!

This is not in my opinion a reliable facility and I for one will not be using it

 

Robert Jordan

 

Image removed by sender.

Virus-free. www.avast.com


David Potter
 

Hi John

 

I tried that link below but it failed to load the Map having found the place I asked for. I have a FamilySearch account. Is there something else required to get this to work.?

 

Thank you.

 

Kind Regards

 

David Charles Potter

 

From: family-historian@groups.io <family-historian@groups.io> On Behalf Of John Hanson
Sent: 08 January 2021 18:39
To: family-historian@groups.io
Subject: Re: [family-historian] Mapping Facility in FH

 

Robert

Whilst Mangotsfield might be in Gloucestershire it is in Keynsham registration district
There are many instances of registration districts crossing the county boundary

 

There are also instances where the registration district is not an actual place – Cosford in Suffolk that crops up many times in my family

 

These days I resort to an online gazetteer to sort them out and first port of call is the LDS mapping facility at https://www.familysearch.org/mapp/


Enter a parish and it will give you all the information that you want about it and links to its wiki which is one of the most underused part of the site these days yet one of the most important with link to all online information for a parish

 

Regards
John Hanson - researching the Halstead/Holstead/Alstead names
Researcher, the Halsted Trust - https://www.halsted.org.uk
Research website - https://www.halstedresearch.org.uk

2021 Family History Conference - https://www.elizabethanancestors.org.uk

 

 

From: family-historian@groups.io <family-historian@groups.io> On Behalf Of Robert Jordan
Sent: 08 January 2021 18:13
To: family-historian@groups.io
Subject: Re: [family-historian] Mapping Facility in FH

 

Adrian

Your comments do not entirely make sense to me.

Firstly when I enter an event the registration district shows up in the source so I know what it is.

 

I know that Mangotsfield is in Gloucestershire and Keynsham is in Somerset is because I lived in the area. If the same situation occurred in say Northumberland then I would not have a clue and I think most people would take the same view as me. To look up some gazetteer to check every place name used would I suggest take the whole enjoyment out of studying family history.

 

I posed a question as to what marker would be used on the map for Mangotsfield, Keynsham, Somerset and gave three optional answers. Your answer was "probably yes"  I do not have a clue what your answer means!!!

 

I'm afraid you have not convinced me. The map facility was introduced in V6 and I can count on the fingers of one hand the number of times I have used it. From memory it was to look at the map for myself as I had travelled to all continents of the world, the second time was when I entered Banjol, Gambia in a place and found that the marker showed Banjul which is the correct spelling and when I changed my place entry to the correct spelling the marker jumped about 50 miles away!

This is not in my opinion a reliable facility and I for one will not be using it

 

Robert Jordan

 

Image removed by sender.

Virus-free. www.avast.com