* FH lack of color coding capability

[BevSmallwood]

Am I correct in saying that there is no color coding other than on charts?

As a former TMG user and current RootsMagic user, I do use color coding to keep different things straight in my mind.

I'm wondering if you used a program prior that had color coding and are now using FH, can you tell me what you are doing instead.

By the way, my heritage is mostly Swedish so there are a TON of couples with the exact same names.

Looking for suggestions.
Thank you
Bev

[tatewise]

For what kinds of 'things' would you use colour coding?

Every record has a unique Record Id, usually shown in [ square brackets ] and differentiates records with the same name.
Alternatively, you can assign Custom Id containing your own codes. See Record Identifiers.

The record level Note in the Main tab of the Individual Property Box can have coloured text that might help.
Otherwise, I suspect users employ Record Flags or Named List membership, which can be shown in the Property Box and the Records Window. See Adding Flags to Main Tab in Property Box (20795) and Record Flags and Named Lists.

See also Wish List Ref 548 Conditional Colour and Font Options and Vote for it.

[Vyger]

Bev,

I use Record Flags and Named Lists membership as Mike Tate suggested. I have to admit I am now doing more of my work in Diagrams and did not make much use of Colour Coding in Rootsmagic due to the singular last colour preference.

In FH I find I am starting to work more in Diagrams where I can use multiple highlights, it's not something RM users would feel natural with but I am becoming more accustomed to working that way as time goes on.

Jackson

[BevSmallwood]

Hi Mike
I have been tracking ancestors on several DNA kits. When I'm looking at a list of Lars Larssons it shows up as being an ancestor of my cousin by the color code. This has helped me make sense of connections.

I do a query before adding people to see if they are already in there in a different context. Frequently a child in one family winds up as a husband in a different family. Lots of intermarriage in these small communities.

I'm not sure if I'm explaining this properly but with so many people with similar names, it is sometimes helpful to see them differentiated on query results.

EDIT
To be clear, I'm talking about coloring the NAME text in all displays - not just the Box fill on a diagram.

[BevSmallwood]

In FH I find I am starting to work more in Diagrams where I can use multiple highlights, it's not something RM users would feel natural with but I am becoming more accustomed to working that way as time goes on.

Hi Jackson
I enjoyed your video on this but I'm talking about searching the data for possible matches before adding a new person, figuring out how different lines connect, etc. All this happens before I'd get to the stage of looking at a diagram.
- Bev

[davidf]

Bev, Jackson

An interesting exchange:

In FH I find I am starting to work more in Diagrams

and

All this happens before I'd get to the stage of looking at a diagram.

I don't know if the difference between Diagrams and Diagram is significant, but I have found as my working methods evolve that I am using Diagrams more and more - particularly when doing family assembly or reconciliation.

I might have a custom diagram for a particular surname in say part of Cumberland (England) over a particular time period (say the UK census years 1841-1911/21). In that diagram I will add lots of partial trees derived from individual census schedules. If I find that someone on any of those trees was born in a different part of say again, Cumberland, I can add them to another diagram (switching out ancestors and descendants as appropriate).

I find looking at these diagrams is a more productive way of finding matches (and occasional duplicates).

The consequence of this evolving work method is that I am now reconsidering my previous practice of maintaining separate projects for apparently distinct clusters - in say North West Cumberland (UK), Hertfordshire (UK), Canada, USA. I can have them in a single project but by having "master" custom diagrams (one per cluster), when I do find a migration link, I just add the individual into the relevant diagram and do any necessary merge. That avoid using the hard to handle "Everyone" diagram (V6).

Now do I want some means of defining a surname-place combination and attaching it to the relevant custom diagram and then to be prompted if there are others matching that surname-place combination that I should consider adding to the diagram?

Custom Queries could do it - but it would be nice to have some form of "hint". Mapping may help - but certainly in V6 I find it more useful for displaying results and for actually deriving the,

Trouble is defining the "place" - "North West Cumberland" is not a "place" in FH but could be approximated to Cockermouth and Wigton Registration Districts - but they are not FH "places" either! I feel a need for some integrated gazetteer functionality possibly that enables me to define "meta places" possibly using an existing hierarchy such as that behind the Family Search England 1851 Jurisdiction map (or GENUKI?) - but that would still require me to ensure that my FH places were in those hierarchies?

West Cumberland by Registration District

Screenshot from 2022-07-24 12-07-01.png (487.65 KiB) Viewed 1663 times

[BevSmallwood]

I will have to look more closely at diagrams but I don't think it will replace the visual functionality I'm looking for.

[davidf]

I don't know if this helps to show how I am working visually. This is an extract of a custom diagram for a family in Cork Ireland.

Williamson in Cork - part diagram

Screenshot from 2022-07-24 12-57-06.png (261.26 KiB) Viewed 1647 times

Or in a bit more detail:

Williamson in Cork - detail

Screenshot from 2022-07-24 13-00-29.png (162.51 KiB) Viewed 1647 times

The orange boxes cover where I think I may have duplicates that I need to merge; within the blue box, I suspect they are all related.

[LornaCraig]

To be clear, I'm talking about coloring the NAME text in all displays - not just the Box fill on a diagram.

Bev, what would you do in a case where the same individual is an ancestor in more than one branch (because of intermarriage)? How would you decide which colour to assign to them?

Incidentally you might find the How Related tool useful. (You'll find it under Tools in the main toolbar). Select two individuals and the tool can display up to 9 different relationships, either graphically or in report form.

[Vyger]

The consequence of this evolving work method is that I am now reconsidering my previous practice of maintaining separate projects for apparently distinct clusters - in say North West Cumberland (UK), Hertfordshire (UK), Canada, USA. I can have them in a single project but by having "master" custom diagrams (one per cluster), when I do find a migration link, I just add the individual into the relevant diagram and do any necessary merge. That avoid using the hard to handle "Everyone" diagram (V6).

David, firstly I am very impressed by your custom diagrams, a long time ago I combined all my geographical fragment and surname projects into one and don't regret it, that was back in Rootsmagic days.

You appear to be a person who realises geography is King and it certainly the case in Ireland where Parish records are scant and I would hope a geographical approach will also help Bev with her particular challenge. Unfortunately FH does not facilitate geocoding of Addresses (yet) as I believe it will come. Regardless of Gazetteer systems and recognition my Irish Place Names are Parishes and the Place Details, which I still maintain in Rootsmagic, are Townlands, all manually geocoded, well with some help from townlands.ie. I would never be able to associate certain families if I was to use a broad modern recognized Place name.

The south Antrim Parish map below encompasses three distinct families I follow in my name studies and without precise geographic associations these clusters would quickly blur.



I have an unconnected death index extract from the Connor district inserted below and you can clearly see the typical Place designation as the Townland, these have become invaluable in my research, I know nothing of Swedish geography but I am hoping a more precise geographic study of associations might enhance the challenge Bev is facing. It doesn't address the problem of colour coding on Records and Focus windows but using expressions for combinations on Place and name combinations on Diagrams might be an acceptable and superior solution for Bev.

I know it's a different way of working but it might be time for Bev to abandon an old single level colour coding system for something more advanced and build a set of expressions to her particular needs.



And I want to learn more about how you create and maintain those multi branch diagrams.

[BevSmallwood]

I'm starting to see the possibilities. Those are some creative charts.

One thing I'm struggling with in terms of place goes back to TMG where each place part was a separate field. This worked famously for my Swedish places which are typically:

Orter (farm, village, house), Socken (Parish), Län (county), Country

In RootsMagic the struggle was whether the Orter was part of the place or place detail. Residence within the parish is often key to figuring out which Anders Andersson is being referenced.

I did a sample import to FH to see what happens and the Orter is put in as Address.

I think I need to go back to RM and edit all those places. I've never been happy with how they've been handled since I left TMG. I've been working on geocoding on the parish level. Doing it to the orter will be challenging.

As with my transition from TMG to RM, I need to change my way of thinking about things to meet FH to its capabilities or deficits.

Frankly, I'm toying with starting fresh in FH rather than importing and fixing. I have a royal mess already from having transitioned from RootsIII/IV into TMG into RM. That might resolve many of the difficulties and finally get the old mess cleaned up.

[davidf]

And I want to learn more about how you create and maintain those multi branch diagrams.

Maintenance can be a problem!

Multi-branch is relatively easy.

1. Set up the main family with what ever-style of diagram is best- I usually use (V6) Ancestor and Descendant Diagram.
2. For multi branch I right click in the diagram and "insert into Diagram" another Ancestor and Descendant Diagram and manually position it. This helps with both other possibly connected fragments and with showing trees of in-laws etc.
3. If you have an individual in both trees you will get connecting ribbons (if switched on); these can be manually minimised by:
   • using the "pruning circles" on the vertical relationship lines (if you have them switched on) to hide ancestors and descendants that you don't want and then
   • moving one instance of the individual to lie on top of the other instance.
   • You then have to manually re-organise the tree using the "Enable Moving/Resizing" icon (2nd from right on V6) and selecting box and bar and selecting a horizontal relationship line and dragging it. Rather than describe in detail how to do this, it is probably best to experiment and you will find it becomes more intuitive - honest.
4. You can then (via right click) add other elements and again via right click but with the relevant element selected order them (back to front etc. - like on Powerpoint). Right click on the element and select format to change colours etc.
5. Remember to save the diagram with a useful name (Diagram, Save Diagram)
6. Retrieve via (View, Saved Charts)
7. (Diagrams vs Charts? Don't ask! I think it is historic and the implications are lost to many)

That creates them. Unfortunately as you add siblings, extra facts etc., the trees redraw themselves and you lose all the alignments!

Diagrams I think were designed for "displaying final results" rather than as a research tool. I think there is a small group of us that do most of our work using diagrams.

There is much I would like to do ("see done") with diagrams to help as a research tool:

• The ability to "pin elements together". "Group" allows you to do this with non-tree elements. I want to be able to pin Joe Bloggs on the original tree to Joe Bloggs (same guy) on another tree that I have added to the diagram. I want to pin an individual into a coloured box.
• The ability to control the "redraw" - quite how I am not sure
• Have an option to show a different text scheme for an individual in a "pop-up/over-lay" which shows when you hover over an individual for more than say 500 microseconds (configurable). Thus the diagram could be prepared with a fairly minimal set of facts (enough to identify), but then hovering could show more facts. Yes, you can click and look at the property box - but it won't for instance have the census facts grouped together as in a diagram text scheme. By hovering it is possible to look at one individual in detail, whilst having another showing in the property box. If you could somehow "impose" a text scheme on the property box, comparison between two possibly duplicate people becomes much easier than the "merge dialog".
• Relationship lines that have a "confidence indication"
• A "possibly the same" graphic element (yes you can draw a line, but you cannot pin its ends to the individuals!) Possibly something using the "Associated Person" (under Miscellaneous Facts - via the All Tab) to set up the relationship and then a way to control the graphical presentation.
• Attach a "diagram note" (like an Excel comment) to an individual, a relationship or any element. Yes, you can add a text box, but you cannot pin it to a tree element!
• Better/easier management of text schemes
  • Creating and Editing
  • Cascading variations from an Installation default, a Project Default, a Diagram Default, to possibly even an Individual
  • Easy "copying" of elements (groups of lines) from one text scheme and pasting into another
  • Ability to use a mouse to reposition text scheme elements moving them up and down
  • etc. etc.
• etc. etc.
  
  I tried to think through this a while ago (I think CSS as a technology would offer so much) - but the brain fog got me; I'm meaning to have another go at it if/when health allows.

[davidf]

...
One thing I'm struggling with in terms of place goes back to TMG where each place part was a separate field. This worked famously for my Swedish places which are typically:

Orter (farm, village, house), Socken (Parish), Län (county), Country

In RootsMagic the struggle was whether the Orter was part of the place or place detail. Residence within the parish is often key to figuring out which Anders Andersson is being referenced.

I did a sample import to FH to see what happens and the Orter is put in as Address.

Wait a short while and Mike will probably be offering advice:

1. You can bulk change the order and position of elements in the place and address fields - too many options some think!
2. Hints on importing from RM - what to do before import and what to do after import

Both beyond my pay-grade!

[davidf]

Unfortunately FH does not facilitate geocoding of Addresses (yet) as I believe it will come. Regardless of Gazetteer systems and recognition my Irish Place Names are Parishes and the Place Details, which I still maintain in Rootsmagic, are Townlands, all manually geocoded, well with some help from townlands.ie. I would never be able to associate certain families if I was to use a broad modern recognized Place name.

I'm not into geocoding yet - I find it either too problematic or too time consuming.

One of the things that I hope will happen is the recognition that Places and Addresses are distinct. Broadly Places are areas:

So "County Durham" (England) is a place and should be shown as an area - as Google Maps does:

Country Durham AREA in Google Maps

Screenshot from 2022-07-24 16-28-43.png (285.62 KiB) Viewed 1576 times

Addresses however are more amenable to being shown as a map-pin as Google Maps does for Durham Cathedral (a big address!):

Durham Cathedral PLACE on Google Maps

Screenshot from 2022-07-24 16-30-42.png (232.28 KiB) Viewed 1576 times

When doing genealogical research, you are sometime interested in the general area (e.g. my previous example of North West Cumberland which I want to show as an area), but you are sometimes interested in knowing that someone lived "two doors down" or "just around the corner", which I would want to show as two pins on a map.

So sometimes you want to map at an area (Place) level, whilst at other times you want to map at a point (Address) level. I hope that one day mapping will make this easier. (V7 may be better than V6, but CP have not made much of it if it is!)
It would help if addresses had to "belong" to a place - "St Marys" is a valid address but could be anywhere. FH could when you type in "St Marys" offer a list of places in your file that already have a "St Marys" and store with the St Marys in the address field a place reference (in privacy brackets or similar to avoid printing).

[Vyger]

There is much I would like to do ("see done") with diagrams to help as a research tool:

I can identify with your well thought out list of possible enhancements, I hope CP are listening

I'm not into geocoding yet - I find it either too problematic or too time consuming.

One of the things that I hope will happen is the recognition that Places and Addresses are distinct. Broadly Places are areas:

So "County Durham" (England) is a place and should be shown as an area - as Google Maps does:

Addresses however are more amenable to being shown as a map-pin as Google Maps does for Durham Cathedral (a big address!):

When doing genealogical research, you are sometime interested in the general area (e.g. my previous example of North West Cumberland which I want to show as an area), but you are sometimes interested in knowing that someone lived "two doors down" or "just around the corner", which I would want to show as two pins on a map.

So sometimes you want to map at an area (Place) level, whilst at other times you want to map at a point (Address) level. I hope that one day mapping will make this easier. (V7 may be better than V6, but CP have not made much of it if it is!)
It would help if addresses had to "belong" to a place - "St Marys" is a valid address but could be anywhere. FH could when you type in "St Marys" offer a list of places in your file that already have a "St Marys" and store with the St Marys in the address field a place reference (in privacy brackets or similar to avoid printing).

Thankfully in FH geocoding is facilitated by drag to Map but sadly this does not apply to Addresses which I agree should be the pin on the map and FH also allows you to disable a map pin where desired.

What cannot be achieved at present is the visualization of Addresses/Sites within a Place and that Place does not need to be a City as you suggested. The image above shows a small area of Belfast and all those individuals do have an association worth further investigation. In the case of an emigrant branch of my family I have Wayne County, PA as the Place mainly because the area is very rural and trying to assign further detail is pointless.

The image above is from Rootsmagic and whilst it is a good visualization there are many dead ends which would be the natural desire of researchers. Heredis 2021 also support the natural need for relational sub divisions and the need for the ability to geocode them, add media and notes, I agree with your St. Marys example, same with Main Street or Greenlawn Cemetery, it means nothing without being relational to the Place.

Rootsmagic have supported this relational link since around 2008, I don't know when Heredis introduced so I believe it will come to FH at some point. Apart from anything else it can be very useful for micro managing research by geography of communities which would certainly help those with a challenge like Bev.

[ColeValleyGirl]

I hope CP are listening

CP won't be aware of it until it appears in the Wish List (not the New Wish List forum, the Wish List itself).

[AdrianBruce]

...
So sometimes you want to map at an area (Place) level, whilst at other times you want to map at a point (Address) level. I hope that one day mapping will make this easier. ...

Just as an aside that may be stating the obvious and / or things already known...

Mike Tate's Map Life Facts plug-in will geocode down at the address level, drawing its "locations" from any combination of Place and Address, including Address-only if you have munged everything into there. Naturally, you can't store Address level co-ordinates in Place Records, but the Plug-in gives various options about how and where to store stuff.

I did have a short play with working at the address level, with a vague idea of showing how peripatetic some of my ancestors were around their home town but rapidly decided that discretion was the better part of valour for a couple of reasons - firstly I still have only a vague idea of how I would like my maps to be presented and secondly I feared the workload. I can cope with geocoding Places "just in case" but geocoding Addresses "just in case" wasn't something I wanted to do.

Bear in mind, incidentally, that free geocoding interfaces have major load limitations and would therefore presumably cough at geocoding addresses in bulk. To say nothing of the impossibility of geocoding streets in places like Bristol or Manchester that have long since been flattened.

... It would help if addresses had to "belong" to a place - "St Marys" is a valid address but could be anywhere. FH could when you type in "St Marys" offer a list of places in your file that already have a "St Marys" and store with the St Marys in the address field a place reference (in privacy brackets or similar to avoid printing).

Well, this is where people like me have shot ourselves in the foot. To a degree. The GEDCOM Standard for Address does include the full place details - I hate that duplication so omit the town etc., but that, of course, leaves me with "St. Mary's". I did try adding the place details to the address in privacy brackets to suppress duplication but unfortunately diagrams fail to respect the privacy brackets even though reports do respect them.

In order to get "St. Mary's" with a full name and enable it to be geocoded, I am currently going through an exercise of promoting churches, chapels, cemeteries, hospitals, etc, from Addresses to be Places in their own right. That's a do-able workload in terms of editting and geocoding, albeit with a few issues of its own in places like Bristol where "St. Paul's, Bristol, Gloucestershire, England" might be a church or a suburb (with elastic boundaries) or even a parish, in-parish or out-parish.

[LornaCraig]

I did try adding the place details to the address in privacy brackets to suppress duplication but unfortunately diagrams fail to respect the privacy brackets even though reports do respect them.

I reported this anomaly to CP in January this year. They "logged it for investigation".

[davidf]

In order to get "St. Mary's" with a full name and enable it to be geocoded, I am currently going through an exercise of promoting churches, chapels, cemeteries, hospitals, etc, from Addresses to be Places in their own right. That's a do-able workload in terms of editting and geocoding, albeit with a few issues of its own in places like Bristol where "St. Paul's, Bristol, Gloucestershire, England" might be a church or a suburb (with elastic boundaries) or even a parish, in-parish or out-parish.

Presumably there is not too much wrong with "qualifying" the first place item when it is an "address", so you might have:

• St. Paul's, Bristol, Gloucestershire, England (locality implied by default)
• St. Paul's Church, Bristol, Gloucestershire, England (in England implied CoE by default)
• St. Paul's Baptist Chapel, Bristol, Gloucestershire, England
• St. Paul's Public House, Bristol, Gloucestershire, England
• St. Mary's, Southampton, Hampshire, England (locality implied by default)
• St. Mary's Stadium, Southampton, Hampshire, England

etc.

[AdrianBruce]

...
Presumably there is not too much wrong with "qualifying" the first place item when it is an "address", so you might have:

• St. Paul's, Bristol, Gloucestershire, England (locality implied by default)
• St. Paul's Church, Bristol, Gloucestershire, England (in England implied CoE by default) ...

Which is pretty much what I did to overcome (most of) the issues - excepting that I settled on "St. Paul's" by itself meaning the church and the area being "St. Paul's area", i.e. the reverse of your suggestion. I suspect I didn't give that much thought to your option because (a) mine was just a direct read-over of the addresses and (b) I was already deep into the conversion when I discovered the St. Paul's conundrum, so I didn't feel like backtracking and tweaking.

The remaining issues centre round what the phrase "St. Paul's" on a source document might actually mean - but sometimes the decision needs to be annotated with "Presumably the St. Paul's in the source means ..." caveats.