Quantcast

Use of dates in the place hierarchy.

classic Classic list List threaded Threaded
12 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Use of dates in the place hierarchy.

John W. Kitz-3
Hi,

I was tinkering with the possibilities provided by the use of dates in the
place hierarchy.

I set a residence event between 1870-11-03 and 1873-05-10 in Street 149,
City, Province, Country, where Street 149 is enclosed by City, which is
enclosed by Province, which is enclosed by Country.

With an after date of 2000-01-01 for the enclosure of Province by Country
this expands to "between 1870-11-03 and 1873-05-10 in Street 149, City,
Province.". Similarly with a before date of 2000-01-01 for the enclosure of
Province by Country this expands to "between 1870-11-03 and 1873-05-10 in
Street 149, City, Province, Country."

However with a regular date of 2000-01-01 for the enclosure this expands to
"between 1870-11-03 and 1873-05-10 in Street 149, City, Province.", i.e. a
regular date is interpreted the same as an after date, which I don't quite
understand the rational of.

Unless I'm doing something wrong or misinterpreting the results could
someone please explain why a 'regular date' works the same as an 'after
date'?

Regards, Jk.


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Use of dates in the place hierarchy.

Dave Scheipers
Hi John,

It has taken me a while to figure having places change  based upon the event date.

First, the reason the generation of the place title stops is because based upon the event date, there is no longer an option for the next level. When you changed the date between the province to the country level with a specific date, the only way the country could ever show up would be if the event is the exact same date.
 
So for an event date  between 1870-11-03 and 1873-05-10

Number --> Street  --> City --> Province --> before  2000-01-01 Country
 Results: Number Street, City, Province, Country  because the event is before 2000

Number --> Street  --> City --> Province --> after  2000-01-01 Country
 Results: Number Street, City, Province   because the Country only shows for dates AFTER 2000

Number --> Street  --> City --> Province --> 2000-01-01 Country
 Results: Number Street, City, Province   because the date of the event is NOT 2000-01-01

And something to keep in mind as you set up your hierarchy, an event with no date has the same effect  as the earliest possible hierarchy path.

Number --> Street  --> City --> Province --> after 2000-01-01 Country A 
                                                                --> before 2000-01-01 Country B

An event with no date will create a path including Country B 

I hope this makes sense. I avoided having places show dynamically based upon the date until I figured out which path I want a place to follow when there was no date for an event and how to make that happen.

Regards, Dave

On Tue, May 2, 2017 at 7:42 PM, John W. Kitz <[hidden email]> wrote:
Hi,

I was tinkering with the possibilities provided by the use of dates in the
place hierarchy.

I set a residence event between 1870-11-03 and 1873-05-10 in Street 149,
City, Province, Country, where Street 149 is enclosed by City, which is
enclosed by Province, which is enclosed by Country.

With an after date of 2000-01-01 for the enclosure of Province by Country
this expands to "between 1870-11-03 and 1873-05-10 in Street 149, City,
Province.". Similarly with a before date of 2000-01-01 for the enclosure of
Province by Country this expands to "between 1870-11-03 and 1873-05-10 in
Street 149, City, Province, Country."

However with a regular date of 2000-01-01 for the enclosure this expands to
"between 1870-11-03 and 1873-05-10 in Street 149, City, Province.", i.e. a
regular date is interpreted the same as an after date, which I don't quite
understand the rational of.

Unless I'm doing something wrong or misinterpreting the results could
someone please explain why a 'regular date' works the same as an 'after
date'?

Regards, Jk.


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Use of dates in the place hierarchy.

John W. Kitz-3
Dave,

On 2017-05-03 02:37, Dave Scheipers wrote:

> Hi John,
>
> It has taken me a while to figure having places change  based upon the
> event date.

Although it's quite a bit of work to look up and enter the relevant
dates I think it's a pretty cool feature.

> First, the reason the generation of the place title stops is because
> based upon the event date, there is no longer an option for the next
> level. When you changed the date between the province to the country
> level with a specific date, the only way the country could ever show
> up would be if the event is the exact same date.

OK

> So for an event date  between 1870-11-03 and 1873-05-10
> __
>
> Number --> Street  --> City --> Province --> before  2000-01-01
> Country
>  Results: Number Street, City, Province, Country  because the event is
> before 2000
>
> Number --> Street  --> City --> Province --> after  2000-01-01 Country
>
>  Results: Number Street, City, Province   because the Country only
> shows for dates AFTER 2000
>
> Number --> Street  --> City --> Province --> 2000-01-01 Country
>
>  Results: Number Street, City, Province   because the date of the
> event is NOT 2000-01-01

The first two behaviours are apparent from the documentation, but for
the last one the documentation may need some updating?

> And something to keep in mind as you set up your hierarchy, an event
> with no date has the same effect  as the earliest possible hierarchy
> path.
>
> Number --> Street  --> City --> Province --> after 2000-01-01 Country
> A
>                                                                 -->
> before 2000-01-01 Country B
>
> An event with no date will create a path including Country B

Meaning setting no date is interpreted as a date with a value of
00-00-0000?

Similarly would entering a Regular date of yyyy result in expansion
being triggered for all events with a date ranging from 00-00-yyyy to
99-99-yyyy, i.e. effectively 01-01-yyyy to 31-12-yyyy?

> I hope this makes sense. I avoided having places show dynamically
> based upon the date until I figured out which path I want a place to
> follow when there was no date for an event and how to make that
> happen.
>
> Regards, Dave

While tinkering, so far I entered "Street (and) Number" enclosed by
City, etc. However I could have used Number enclosed by Street enclosed
by City, etc. since I understand from previous emails that it should be
possible to have that expand to Street Number, City, etc. instead of to
Number, Street, City on reports by changing some setting.

Correct?

If yes, could you please (once again) briefly explain what setting I
need to change to make that happen?

Finally, could there be a potential problem in this working in
conjunction with the use of dates.

In that case one would enter a number once and have that enclosed by all
streets in which that number occurs in one's data?

How would the expansion logic cope with that, particularly if the
expansion were set up as conditional, based on time frames in which the
number was or wasn't enclosed by Street?

Regards, Jk.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Use of dates in the place hierarchy.

Dave Scheipers
Hi John,

Yes a cool feature and yes the problem of finding what changed when is the challenge. Fortunately for Europe, there is the GetGOV tool. I'll get back to that.

I can think of only one reason to have an enclosed by date as a specific date. All other enclosed by dates are either a range, from ... to... or after ....    I try to avoid before.... . Places are not enclosed by another higher locality on just one date. They were enclosed by the higher locality over a span of time (from... to...) or they are currently enclosed by and on going (after....)

I use a specific date to prevent adding the enclosed by place to the place generation. As an example, I have various oceans all enclosed by World Oceans. I enclose them so that they are all easy to find in the hierarchy list but I do not want "World Oceans" added to the place title when setting the ocean when a relative died while at sea. The enclosed by date I use is 1 Jan 1. 

For events having no date, I think of Gramps using 1 Jan 1 as the date and what path the place hierarchy automatic generation will use to display the place. This is why I am very careful not to use before...  as an enclosed by date. Events with no dates follow that path which may not be wanted.

before YYYY would be understood as before 1 January YYYY
after YYYY would be understood as after 31 December YYYY
Similarly, MMM YYYY would be either before 1 MMM YYYY or after 31 MMM YYYY

before or after a specific date 3 May 2000 means exactly that, before or after but excluding the specified date.
from... to... includes the specific dates listed while between... and... excludes the specific dates (I'm not 100% sure of this and could be wrong but this is how I think of the differences.)

As to the street and Number order. Gramps has been coded to have the Number enclosed by the Street and suppress the comma between them. I remember seeing either a bug fix or feature request to accommodate other display formats but have no idea what's being done if anything. If you need the address shown as Street Number, I would keep them together as one entry under the City.

The Number will always be enclosed by the street. 23 Main St is the location and can be located on a map with GPS  coordinates. These coordinates would be recorded in the Number place. 23 Elm St would have different coordinates. So the same 23 could never show under different streets.

In general, I cannot think of a reason to have enclosed by dates at the street or number level. Enclosed by dates occur for things greater than the city although some sections of cities may have been independent of their current city so one path goes through the city to the province and at some time, the 'neighborhood' went directly to the province.



This is a gramplet  that takes the information from http://gov.genealogy.net/search/index and imports the results into gramps with all of the hierarchies and the enclosed by dates. You use the web site to find the codes. If you don't want to use it on your database, you can at least create a new empty database to see what it does and what the enclosures are. You can also see how multiple enclosed by's are set up, as well how to set up different names and languages. 

I hope I covered everything and maybe raised some other issues. Dave


On Wed, May 3, 2017 at 3:56 AM, John W. Kitz <[hidden email]> wrote:
Dave,

On 2017-05-03 02:37, Dave Scheipers wrote:

Hi John,

It has taken me a while to figure having places change  based upon the
event date.

Although it's quite a bit of work to look up and enter the relevant dates I think it's a pretty cool feature.

First, the reason the generation of the place title stops is because
based upon the event date, there is no longer an option for the next
level. When you changed the date between the province to the country
level with a specific date, the only way the country could ever show
up would be if the event is the exact same date.

OK

So for an event date  between 1870-11-03 and 1873-05-10
__

Number --> Street  --> City --> Province --> before  2000-01-01
Country
 Results: Number Street, City, Province, Country  because the event is
before 2000

Number --> Street  --> City --> Province --> after  2000-01-01 Country

 Results: Number Street, City, Province   because the Country only
shows for dates AFTER 2000

Number --> Street  --> City --> Province --> 2000-01-01 Country

 Results: Number Street, City, Province   because the date of the
event is NOT 2000-01-01

The first two behaviours are apparent from the documentation, but for the last one the documentation may need some updating?

And something to keep in mind as you set up your hierarchy, an event
with no date has the same effect  as the earliest possible hierarchy
path.

Number --> Street  --> City --> Province --> after 2000-01-01 Country
A
                                                                -->
before 2000-01-01 Country B

An event with no date will create a path including Country B

Meaning setting no date is interpreted as a date with a value of 00-00-0000?

Similarly would entering a Regular date of yyyy result in expansion being triggered for all events with a date ranging from 00-00-yyyy to 99-99-yyyy, i.e. effectively 01-01-yyyy to 31-12-yyyy?

I hope this makes sense. I avoided having places show dynamically
based upon the date until I figured out which path I want a place to
follow when there was no date for an event and how to make that
happen.

Regards, Dave

While tinkering, so far I entered "Street (and) Number" enclosed by City, etc. However I could have used Number enclosed by Street enclosed by City, etc. since I understand from previous emails that it should be possible to have that expand to Street Number, City, etc. instead of to Number, Street, City on reports by changing some setting.

Correct?

If yes, could you please (once again) briefly explain what setting I need to change to make that happen?

Finally, could there be a potential problem in this working in conjunction with the use of dates.

In that case one would enter a number once and have that enclosed by all streets in which that number occurs in one's data?

How would the expansion logic cope with that, particularly if the expansion were set up as conditional, based on time frames in which the number was or wasn't enclosed by Street?

Regards, Jk.


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Use of dates in the place hierarchy.

John W. Kitz-3
Dave,

On 2017-05-03 15:53, Dave Scheipers wrote:
> Hi John,
>
> Yes a cool feature and yes the problem of finding what changed when is
> the challenge. Fortunately for Europe, there is the GetGOV tool. I'll
> get back to that.
>
> I can think of only one reason to have an enclosed by date as a
> specific date.

If I have an event e.g. between 1870-11-03 and 1873-05-10 that involves
a 'date based expandable location' what would happen is the 'trigger
date' were 1870 or 1873?

Would the expansion be triggered or should the date that triggers the
expansion always be outside of the range on the date(s) associated to an
event?

> All other enclosed by dates are either a range, from
> ... to... or after ....    I try to avoid before.... . Places are not
> enclosed by another higher locality on just one date. They were
> enclosed by the higher locality over a span of time (from... to...) or
> they are currently enclosed by and on going (after....)
>
> I use a specific date to prevent adding the enclosed by place to the
> place generation. As an example, I have various oceans all enclosed by
> World Oceans. I enclose them so that they are all easy to find in the
> hierarchy list but I do not want "World Oceans" added to the place
> title when setting the ocean when a relative died while at sea. The
> enclosed by date I use is 1 Jan 1.
>
> For events having no date, I think of Gramps using 1 Jan 1 as the date
> and what path the place hierarchy automatic generation will use to
> display the place. This is why I am very careful not to use before...
> as an enclosed by date. Events with no dates follow that path which
> may not be wanted.
>
> before YYYY would be understood as before 1 January YYYY
> after YYYY would be understood as after 31 December YYYY
> Similarly, MMM YYYY would be either before 1 MMM YYYY or after 31 MMM
> YYYY
>
> before or after a specific date 3 May 2000 means exactly that, before
> or after but excluding the specified date.
> from... to... includes the specific dates listed while between...
> and... excludes the specific dates (I'm not 100% sure of this and
> could be wrong but this is how I think of the differences.)
>
> As to the street and Number order. Gramps has been coded to have the
> Number enclosed by the Street and suppress the comma between them.

This I don't understand as it seems to imply that I can enter Street and
Number separately, which I can but only when I manually enter their
enclosed by relationship.

If I do the resulting address will expand to Number, Street, City, etc.,
which is normal in most English speaking areas, but not for many other
parts of the world.

The only way, that I can think of to avoid that is to enter Street and
Number in one field.

> I
> remember seeing either a bug fix or feature request to accommodate
> other display formats but have no idea what's being done if anything.
> If you need the address shown as Street Number, I would keep them
> together as one entry under the City.
>
> The Number will always be enclosed by the street. 23 Main St is the
> location and can be located on a map with GPS  coordinates. These
> coordinates would be recorded in the Number place. 23 Elm St would
> have different coordinates. So the same 23 could never show under
> different streets.
>
> In general, I cannot think of a reason to have enclosed by dates at
> the street or number level.

One would be Amsterdam that prior to 1796 didn't have any method to
identify individual buildings/lots other than street names. According to
an article on the site of the Amsterdam City Archives it was decided to
implement a method to uniquely identify individual buildings/lots from
1796 onward, but in the following decades the system was revised twice,
resulting in two distinct points in time when, at least in Amsterdam,
the Number changed.

> Enclosed by dates occur for things greater
> than the city although some sections of cities may have been
> independent of their current city so one path goes through the city to
> the province and at some time, the 'neighborhood' went directly to the
> province.
>
> Now, back to GetGOV.
> https://gramps-project.org/wiki/index.php?title=Addon:GetGOV
>
> This is a gramplet  that takes the information from
> http://gov.genealogy.net/search/index and imports the results into
> gramps with all of the hierarchies and the enclosed by dates. You use
> the web site to find the codes. If you don't want to use it on your
> database, you can at least create a new empty database to see what it
> does and what the enclosures are. You can also see how multiple
> enclosed by's are set up, as well how to set up different names and
> languages.
>
> I hope I covered everything and maybe raised some other issues. Dave

Thanks and regards, Jk.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Use of dates in the place hierarchy.

John W. Kitz-3
Dave,

On 2017-05-03 15:53, Dave Scheipers wrote:

>> I remember seeing either a bug fix or feature request to accommodate
>> other display formats but have no idea what's being done if anything.
>> If you need the address shown as Street Number, I would keep them
>> together as one entry under the City.
>>
>> The Number will always be enclosed by the street. 23 Main St is the
>> location and can be located on a map with GPS  coordinates. These
>> coordinates would be recorded in the Number place. 23 Elm St would
>> have different coordinates. So the same 23 could never show under
>> different streets.
>>
>> In general, I cannot think of a reason to have enclosed by dates at
>> the street or number level.
>
> One would be Amsterdam that prior to 1796 didn't have any method to
> identify individual buildings/lots [by an address] other than street
> names. According to an article on the site of the Amsterdam City
> Archives it was decided to implement a method to uniquely identify
> individual buildings/lots [by an address] from 1796 onward, but in
> the following decades the system was revised twice, resulting in two
> distinct points in time when, at least in Amsterdam, the Number
> changed.

I was thinking about my comment a bit more and it may need some
clarification in that I do believe that (a precursor to) cadastral
registration may (or may not) have existed to identify individual lots,
but AFAIK no addresses that include some number as we known them to day.

That said, I also realized my reasoning was incorrect in that the issue
I described can only be addressed by some form of time dependent
conversion of addresses to geographic coordinates and not by means of
time dependent expansion of the addresses in one's data.

I.e. to solve the issue described the software that enriches the data
with geographic coordinates would need to be 'time aware' and insert one
from a number of possible different geographic coordinates that existed
for every address of the form 'Street and Number'.

I hope this explains the point I'm trying to make sufficiently

Regards, Jk.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Use of dates in the place hierarchy.

Dave Scheipers
Hi John,

I think you are over thinking the separate issues.

Issue one are the events and activities of your relative. These events have dates (or a span of time) and a location where it occurred.

You then have a place database. It consists of Countries, provinces, cities, villages, neighborhoods, streets and houses on some streets.

In a simple world, the city's genealogical tree is a stick.


While others have convoluted intertwining branches.


Just because cities and provinces have dates of when they were settled or formally recognized does not mean that these dates have to be included as an enclosed by date. In the Amsterdam tree above, no dates are needed. Amsterdam was always under Nord-Holland which was always under Nederland. The enclosed by dates are only important when the locale has more than one enclosed by place. Then you need to add the dates to indicate which path upwards to take based upon an event date. between these dates take this branch, between these date go this way and after this date go this third way.

And the date a city or village was settled or formally recognized is unimportant. If they didn't exist before a specific time, you will not find records saying a relative lived there at that time. I live in a town that broke off from another town. It doesn't matter what that date was because no document could be found before that date saying someone was born there. Before the split date people were all born in that older larger town.

In some cases, we learn that event will include a street that can be found in the city. And sometimes there will be an actual address. And we can find them on a map today and plot their coordinates. Sometime, we have information that the event occurred at a specific address. But after extensive search, we cannot find where that spot existed. In that case, I'll attach a note to the place record stating that and put the coordinates at the location that can be identified (usually the city).

We now come to the issue of what a particular place was called at various times through history. I'm not sure if you realize that the Alternative Names tab includes a date field. I have a local hospital that has changed names. When dates are added to the name (invoke place name editor) and the alternative names, the name of the place will modify based upon the date of the event. And this name change will work on any place record that changed names at some point in time.

Using a combination of having different enclosed by places with their enclosed by dates and different names with their dates, you can get the location to generate appropriate name for the date of the event

I really hope this helps. Dave

On Thu, May 4, 2017 at 10:51 AM, John W. Kitz <[hidden email]> wrote:
Dave,

On 2017-05-03 15:53, Dave Scheipers wrote:

I remember seeing either a bug fix or feature request to accommodate
other display formats but have no idea what's being done if anything.
If you need the address shown as Street Number, I would keep them
together as one entry under the City.

The Number will always be enclosed by the street. 23 Main St is the
location and can be located on a map with GPS  coordinates. These
coordinates would be recorded in the Number place. 23 Elm St would
have different coordinates. So the same 23 could never show under
different streets.

In general, I cannot think of a reason to have enclosed by dates at
the street or number level.

One would be Amsterdam that prior to 1796 didn't have any method to
identify individual buildings/lots [by an address] other than street
names. According to an article on the site of the Amsterdam City
Archives it was decided to implement a method to uniquely identify
individual buildings/lots [by an address] from 1796 onward, but in
the following decades the system was revised twice, resulting in two
distinct points in time when, at least in Amsterdam, the Number
changed.

I was thinking about my comment a bit more and it may need some clarification in that I do believe that (a precursor to) cadastral registration may (or may not) have existed to identify individual lots, but AFAIK no addresses that include some number as we known them to day.

That said, I also realized my reasoning was incorrect in that the issue I described can only be addressed by some form of time dependent conversion of addresses to geographic coordinates and not by means of time dependent expansion of the addresses in one's data.

I.e. to solve the issue described the software that enriches the data with geographic coordinates would need to be 'time aware' and insert one from a number of possible different geographic coordinates that existed for every address of the form 'Street and Number'.

I hope this explains the point I'm trying to make sufficiently

Regards, Jk.


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Use of dates in the place hierarchy.

John W. Kitz-3
Dave,

On 2017-05-04 20:13, Dave Scheipers wrote:

> Hi John,
>
> I think you are over thinking the separate issues.
>
> Issue one are the events and activities of your relative. These events
> have dates (or a span of time) and a location where it occurred.
>
> You then have a place database. It consists of Countries, provinces,
> cities, villages, neighborhoods, streets and houses on some streets.
>
> In a simple world, the city's genealogical tree is a stick.
>
> http://gov.genealogy.net/item/show/AMSDAMJO22LI [1]
>
> While others have convoluted intertwining branches.
>
> http://gov.genealogy.net/item/show/FROSENJN48KK [2]
>
> Just because cities and provinces have dates of when they were settled
> or formally recognized does not mean that these dates have to be
> included as an enclosed by date.

True, however (see below).

A historically accurate graphic representation of Amsterdam over time,
may be as convoluted as the one for Frommenhausen, DE that you refer to
above. The fact that some database does not contain data for
Amsterdam[1] similar to the data that results in the graphic for
 Frommenhausen[2] does not mean that it does not exist.

> In the Amsterdam tree above, no dates
> are needed. Amsterdam was always under Nord-Holland which was always
> under Nederland.

True, but it doesn't result in an historically complete representation.

> The enclosed by dates are only important when the
> locale has more than one enclosed by place. Then you need to add the
> dates to indicate which path upwards to take based upon an event date.
> between these dates take this branch, between these date go this way
> and after this date go this third way.
>
> And the date a city or village was settled or formally recognized is
> unimportant. If they didn't exist before a specific time, you will not
> find records saying a relative lived there at that time.

It may be considered a matter of opinion, but I think that's partially
true and partially an incorrect representation of reality. Over the
course of time one or more villages may have been merged into bigger,
sometimes previously non-existed, villages or municipalities, so chances
are you'll find records that pertain to communities that no longer
exist.

Likewise when people try to find past relatives they may be looking for
records of communities for which no records existed yet during the time
frame when those relatives are assumed to have lived there because the
community itself didn't exist yet.

Similar situations may exist for provinces and even entire nations,
particularly in the geographic area that we today refer to as Europe.
One of the best examples are all entries in genealogy data that suggest
people originated from Russia, while in actual fact they originated from
Prussia.

> I live in a
> town that broke off from another town. It doesn't matter what that
> date was because no document could be found before that date saying
> someone was born there. Before the split date people were all born in
> that older larger town.
>
> In some cases, we learn that event will include a street that can be
> found in the city. And sometimes there will be an actual address. And
> we can find them on a map today and plot their coordinates. Sometime,
> we have information that the event occurred at a specific address. But
> after extensive search, we cannot find where that spot existed. In
> that case, I'll attach a note to the place record stating that and put
> the coordinates at the location that can be identified (usually the
> city).
>
> We now come to the issue of what a particular place was called at
> various times through history. I'm not sure if you realize that the
> Alternative Names tab includes a date field. I have a local hospital
> that has changed names. When dates are added to the name (invoke place
> name editor) and the alternative names, the name of the place will
> modify based upon the date of the event. And this name change will
> work on any place record that changed names at some point in time.
>
> Using a combination of having different enclosed by places with their
> enclosed by dates and different names with their dates, you can get
> the location to generate appropriate name for the date of the event

This last issue IMHO depends on the goals that one is trying to achieve.
I think it may be pretty cool if it would be possible to covert address
level data to geographic coordinates with the same level of accuracy to
ultimately be able to visualize migration.

Obviously as a first step the same may, probably more easily, be
achieved using a lower degree of accuracy, i.e. at the community level,
but while I realize that it will be quite a job to collect, enter and
convert all data necessary to achieve that I think the end result might
be pretty cool.

For such an effort to result in an accurate representation of migration
e.g. within the limits of a community the conversion of address level
data to geographic coordinates does have to be time aware, given that
there may be situations, such as in the case of Amsterdam, in which more
than one geographic coordinate applies to one and the same address.

> I really hope this helps. Dave

Regards, Jk.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Use of dates in the place hierarchy.

Dave Scheipers
Hi John,

I offer up Roxbury to hopefully illustrate how some dates are important while others or not.

Enclosed By
Boston    City          after 4 Jan 1868
Norfolk    County     from 20 Jun 1793 to 5 Jan 1868
Suffolk     County     from 1620 to 20 Jun 1793
Suffolk     County

The earliest date is 1620. On 1 Jan 1620, Roxbury did not exist, Suffolk County did not exist and nothing listed in the upper hierarchy existed. The first permanent European settlement would not set down roots until November 1620. But I use 1620 for all "starting" dates  regardless of when most of the places where either settled or became an official place.

But when Suffolk County was created, it's upper hierarchy was true and existed. When people started calling this locality Roxbury, it was in Suffolk County. From 20 Jun 1793 to 5 Jan 1868 Roxbury became a part of the new Norfolk County. And on 5 Jan 1868 Roxbury was annexed by the city of Boston so Roxbury became a Neighborhood and found itself back in Suffolk County.

The last Suffolk County with no dates is how I handle events with no dates. The default path which takes it back to Roxbury, Suffolk Co, Massachusetts which is how it is known.

Hopefully this illustrates what I was saying about starting dates. It also shows how to set up a place that was latter merged with another entity. In the case of Roxbury, records still say Roxbury and its name lives on. Some places merge with another and the name disappears from future records. In that case, the place record stays in the database for use on older records, it is just never used on more recent events.

In the Netherlands, whole communities disappeared under flooded land. The place record still may need to be created. But do you have to put that flood date in the gramps record? No. It will just not be added to an event record after its flood date. For these records I Tag them with Defunct. (opgeheven ?) [Dutch for others following this thread] . I try to include the Wikipedia page under the Internet tab for all places to provide all these other facts for those interested.

I hope this helps. One could spend a lifetime creating a totally accurate, covering all places, database. Dave

On Sat, May 6, 2017 at 6:07 AM, John W. Kitz <[hidden email]> wrote:
Dave,

On 2017-05-04 20:13, Dave Scheipers wrote:
> Hi John,
>
> I think you are over thinking the separate issues.
>
> Issue one are the events and activities of your relative. These events
> have dates (or a span of time) and a location where it occurred.
>
> You then have a place database. It consists of Countries, provinces,
> cities, villages, neighborhoods, streets and houses on some streets.
>
> In a simple world, the city's genealogical tree is a stick.
>
> http://gov.genealogy.net/item/show/AMSDAMJO22LI [1]
>
> While others have convoluted intertwining branches.
>
> http://gov.genealogy.net/item/show/FROSENJN48KK [2]
>
> Just because cities and provinces have dates of when they were settled
> or formally recognized does not mean that these dates have to be
> included as an enclosed by date.

True, however (see below).

A historically accurate graphic representation of Amsterdam over time,
may be as convoluted as the one for Frommenhausen, DE that you refer to
above. The fact that some database does not contain data for
Amsterdam[1] similar to the data that results in the graphic for
 Frommenhausen[2] does not mean that it does not exist.

> In the Amsterdam tree above, no dates
> are needed. Amsterdam was always under Nord-Holland which was always
> under Nederland.

True, but it doesn't result in an historically complete representation.

> The enclosed by dates are only important when the
> locale has more than one enclosed by place. Then you need to add the
> dates to indicate which path upwards to take based upon an event date.
> between these dates take this branch, between these date go this way
> and after this date go this third way.
>
> And the date a city or village was settled or formally recognized is
> unimportant. If they didn't exist before a specific time, you will not
> find records saying a relative lived there at that time.

It may be considered a matter of opinion, but I think that's partially
true and partially an incorrect representation of reality. Over the
course of time one or more villages may have been merged into bigger,
sometimes previously non-existed, villages or municipalities, so chances
are you'll find records that pertain to communities that no longer
exist.

Likewise when people try to find past relatives they may be looking for
records of communities for which no records existed yet during the time
frame when those relatives are assumed to have lived there because the
community itself didn't exist yet.

Similar situations may exist for provinces and even entire nations,
particularly in the geographic area that we today refer to as Europe.
One of the best examples are all entries in genealogy data that suggest
people originated from Russia, while in actual fact they originated from
Prussia.

> I live in a
> town that broke off from another town. It doesn't matter what that
> date was because no document could be found before that date saying
> someone was born there. Before the split date people were all born in
> that older larger town.
>
> In some cases, we learn that event will include a street that can be
> found in the city. And sometimes there will be an actual address. And
> we can find them on a map today and plot their coordinates. Sometime,
> we have information that the event occurred at a specific address. But
> after extensive search, we cannot find where that spot existed. In
> that case, I'll attach a note to the place record stating that and put
> the coordinates at the location that can be identified (usually the
> city).
>
> We now come to the issue of what a particular place was called at
> various times through history. I'm not sure if you realize that the
> Alternative Names tab includes a date field. I have a local hospital
> that has changed names. When dates are added to the name (invoke place
> name editor) and the alternative names, the name of the place will
> modify based upon the date of the event. And this name change will
> work on any place record that changed names at some point in time.
>
> Using a combination of having different enclosed by places with their
> enclosed by dates and different names with their dates, you can get
> the location to generate appropriate name for the date of the event

This last issue IMHO depends on the goals that one is trying to achieve.
I think it may be pretty cool if it would be possible to covert address
level data to geographic coordinates with the same level of accuracy to
ultimately be able to visualize migration.

Obviously as a first step the same may, probably more easily, be
achieved using a lower degree of accuracy, i.e. at the community level,
but while I realize that it will be quite a job to collect, enter and
convert all data necessary to achieve that I think the end result might
be pretty cool.

For such an effort to result in an accurate representation of migration
e.g. within the limits of a community the conversion of address level
data to geographic coordinates does have to be time aware, given that
there may be situations, such as in the case of Amsterdam, in which more
than one geographic coordinate applies to one and the same address.

> I really hope this helps. Dave

Regards, Jk.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Use of dates in the place hierarchy.

John W. Kitz-3
Dave,

On 2017-05-06 16:19, Dave Scheipers wrote:

> Hopefully this illustrates what I was saying about starting dates. It
> also shows how to set up a place that was latter merged with another
> entity. In the case of Roxbury, records still say Roxbury and its name
> lives on. Some places merge with another and the name disappears from
> future records. In that case, the place record stays in the database
> for use on older records, it is just never used on more recent events.

I above clearly illustrates the use of dates as it relates to enclosed
by.

> In the Netherlands, whole communities disappeared under flooded land.

It seems somewhat beyond the scope of this email list, but I don't known
what communities you might be eluding to, particularly since the Dutch
have made quite a name for themselves in taking land from the sea to be
able to build industrial zones (such as in the case of the Port of
Rotterdam) or new communities (such as e.g. the 'polders' in the
IJssellake), thus expanding the country in size over time (see e.g.
https://en.wikipedia.org/wiki/Flood_control_in_the_Netherlands#Modern_developments).

I think the steady disappearance of communities in The Netherlands,
particularly in the most recent decades, is more likely attributable to
legislature aimed at smaller communities merging into larger ones than
to flooding or anything else for that matter.

Unfortunately I couldn't find an English language article that explains
this better and in more detail than I can here, but in case anybody
proficient in Dutch reading this is interested the Dutch term to Google
for is "gemeentelijk herindeling".

> The place record still may need to be created. But do you have to put
> that flood date in the gramps record? No. It will just not be added to
> an event record after its flood date. For these records I Tag them
> with Defunct. (opgeheven ?) [Dutch for others following this thread] .
> I try to include the Wikipedia page under the Internet tab for all
> places to provide all these other facts for those interested.

I believe in an earlier reply I mentioned the issue that is related to
the introduction of addresses of the form 'Street Number' in the late
1700's and the subsequent changes to that method of addressing during
the course of the 1800's as a result of which historically one and the
same address of the from 'Street Number' may apply to more than one set
of geographic coordinates is one that most likely must be addressed at
the level of the geographic data and the tool that enriches addresses
contained within the Gramps with coordinates.

Regards, Jk.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Use of dates in the place hierarchy.

Dave Scheipers


On Mon, May 8, 2017 at 6:06 AM, John W. Kitz <[hidden email]> wrote:
Dave,

On 2017-05-06 16:19, Dave Scheipers wrote:

> Hopefully this illustrates what I was saying about starting dates. It
> also shows how to set up a place that was latter merged with another
> entity. In the case of Roxbury, records still say Roxbury and its name
> lives on. Some places merge with another and the name disappears from
> future records. In that case, the place record stays in the database
> for use on older records, it is just never used on more recent events.

I above clearly illustrates the use of dates as it relates to enclosed
by.

> In the Netherlands, whole communities disappeared under flooded land.

It seems somewhat beyond the scope of this email list, but I don't known
what communities you might be eluding to, particularly since the Dutch
have made quite a name for themselves in taking land from the sea to be
able to build industrial zones (such as in the case of the Port of
Rotterdam) or new communities (such as e.g. the 'polders' in the
IJssellake), thus expanding the country in size over time (see e.g.
https://en.wikipedia.org/wiki/Flood_control_in_the_Netherlands#Modern_developments).

I think the steady disappearance of communities in The Netherlands,
particularly in the most recent decades, is more likely attributable to
legislature aimed at smaller communities merging into larger ones than
to flooding or anything else for that matter.

Unfortunately I couldn't find an English language article that explains
this better and in more detail than I can here, but in case anybody
proficient in Dutch reading this is interested the Dutch term to Google
for is "gemeentelijk herindeling".

I recently had contact with another researching my family in the Netherland and Belgium. He pointed me to some family branches from Zeeland in the Netherlands and while adding those places, I thought I read that yes, more land was being created, there were some land flooded. If I'm mistaken, I am sorry.

But while writing the above, it reminded me that there were three towns in my state of Massachusetts that were flooded and went out of existence when a reservoir was created for more water for Boston.

These towns are in my database but I don't these flood dates on them. Events just stop linking to them after these date.


> The place record still may need to be created. But do you have to put
> that flood date in the gramps record? No. It will just not be added to
> an event record after its flood date. For these records I Tag them
> with Defunct. (opgeheven ?) [Dutch for others following this thread] .
> I try to include the Wikipedia page under the Internet tab for all
> places to provide all these other facts for those interested.

I believe in an earlier reply I mentioned the issue that is related to
the introduction of addresses of the form 'Street Number' in the late
1700's and the subsequent changes to that method of addressing during
the course of the 1800's as a result of which historically one and the
same address of the from 'Street Number' may apply to more than one set
of geographic coordinates is one that most likely must be addressed at
the level of the geographic data and the tool that enriches addresses
contained within the Gramps with coordinates.

I think you need to look at the place record starting with the GPS coordinates. That identifies a particular spot on the earth that is unchanged through time. The spot may have had different names throughout history but the spot stayed where it always was and will be. So all this activity is putting a name to the spot which may have changed throughout its history.

So most of what you describe above would be handled with the alternative name with dates to reflect these changes if you can find this information. A street name may have changed and the alternative places with the dates can be set. And if a street was renumbered, these can be reflected the same way. 

As your search reaches out to other communities, you'll become less aware of these changes. It is difficult tracking some of these name changes at the city level. Finding a database or source to track that Main Street became Elm Street at some point in time or that Willemstrasse 10 became Willemstrasse 1110 will be next to impossible.

When I encounter a location that a record says happened at a particular place, but after a search I cannot locate today, I'll create its record, place a note in the record that I cannot locate it and set the coordinates to the closest level I can locate.

Dave


Regards, Jk.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Use of dates in the place hierarchy.

John W. Kitz-3
Dave,

On 2017-05-08 15:56, Dave Scheipers wrote:

> I recently had contact with another researching my family in the
> Netherland and Belgium. He pointed me to some family branches from
> Zeeland in the Netherlands and while adding those places, I thought I
> read that yes, more land was being created, there were some land
> flooded. If I'm mistaken, I am sorry.

No problem, you were probably referring to
https://en.wikipedia.org/wiki/North_Sea_flood_of_1953.

A minor misunderstanding thanks to the use of email.

> I think you need to look at the place record starting with the GPS
> coordinates. That identifies a particular spot on the earth that is
> unchanged through time. The spot may have had different names
> throughout history but the spot stayed where it always was and will
> be.

What I was trying to convey is that while each and every set of
geographic coordinates may be unique the labels that we humans use to
refer to any unique set do not necessarily need to be, just as more than
one host name may refer to a globally unique IP address. Since these
labels, in the case of Gramps, addresses of the form 'Street Number' are
the basis for the conversion to, or if you will enrichment with,
geographic coordinates there must be a mechanism that takes time into
account when doing that conversion.

> So all this activity is putting a name to the spot which may have
> changed throughout its history.
>
> So most of what you describe above would be handled with the
> alternative name with dates to reflect these changes if you can find
> this information. A street name may have changed and the alternative
> places with the dates can be set. And if a street was renumbered,
> these can be reflected the same way.
>
> As your search reaches out to other communities, you'll become less
> aware of these changes. It is difficult tracking some of these name
> changes at the city level. Finding a database or source to track that
> Main Street became Elm Street at some point in time or that
> Willemstrasse 10 became Willemstrasse 1110 will be next to impossible.

Based on the information that I've looked at it is my understand that,
at least in Amsterdam, the changes that followed the introduction of
addresses in the late 1700's, are on a smaller scale (from a geographic
perspective) than in your example in that e.g. the address 'Street 123'
may have referred to a different house before or after the date of the
change to the method of addressing.

The Amsterdam City Archives do provide for a conversion tool. But it is
intended to figure out what the current (human readable) address is of
e.g. an early 1800's address, it does not involve conversion to
geographic coordinates.

For that one would need another source of data, such as e.g. historical
cadastral records, but if such data were available and it were possible
to get an addon, like the one you mentioned in an earlier email, to do
the conversion properly, it looks like, at least in theory, it ought to
be possible to visualize historic migration down to the street level.

> When I encounter a location that a record says happened at a
> particular place, but after a search I cannot locate today, I'll
> create its record, place a note in the record that I cannot locate it
> and set the coordinates to the closest level I can locate.
>
> Dave

It would seem we've gone of on a bit of a tangent, so let's leave this
topic for the time being.

Thanks for taking the time to share you points of view on this, regards,
Jk.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Gramps-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gramps-users
https://gramps-project.org
Loading...