* Updating Source Template Fields already populated - issues

[AdrianBruce]

Background - See the thread viewtopic.php?p=132813#p132813

I raised in that thread the topic of updating the definitions of source template fields when they are already populated in linked source records. Mark Draper's comment (which I agree with) was that the issues deserved their own thread "if only to set reasonable expectation boundaries over what templates can and can't do, ... I think it needs its own thread and real examples to work through."

So I volunteered myself to type (some of) the words. Firstly I need to point out that I have no serious experience of using templated sources. I use generic sources where I have skeleton source records that get cloned for each new source record of that type. However, over the years, I know very well that those skeletons have changed. I have dabbled in templates but the sheer volume of my source records, plus inconsistent data contents stop me from going further. But I can see other issues - or I think I can - and (for the purposes of this thread) I'd like to discuss the issues raised by updating the definitions of source template fields when they are already populated in linked source records.

My initial thoughts (feel free to challenge) are that if I delete a template field, the field remains in the source record but it is not easily accessed. (So what?)

If I add a template field, this seems easy enough - my worry would be, would it make the source records created using the previous versions of the template, invalid in some way? I don't think so.... (Though obviously the work of populating the new item exists).

If I alter the definition of a template field - that would seem to have the possibility to cause issues. There must be type changes that work in one way but could cause issues another way - e.g. Date to Text might be OK but Text to Date might require work beforehand.

The real problems surely occur when stuff is updated in groups - e.g. Templated Item A is replaced by Templated Items B and C.

An example - I tried to set up a few example source records in a little project based on the Essentials collection and got the distinct impression that I'd need to alter lots of the templates - otherwise I'd be losing information. (Yes, I could mung everything into one templated item but then what's the point in having templates?). So for instance the Civil Registration Certificate went from 11 to 18 items. Most of those were extra items relating to how the copy in my hand had been produced (handwritten, facsimile, contemporary copy given to Informant, etc). But the major change was where I replaced the fairly amorphous item "Location" with "Event Place" and "Indexed Place" (Event Place = "Haslington", say, and Indexed Place = "Crewe Sub-District, Nantwich Registration District"). If I had source records already set up with the Essentials / Civil Registration Certificate (I didn't), then I'd be at risk of losing content in the original Location templated item.

In this case, I have the more detailed skeleton source record to set myself off on the right foot - but I know that I will refine stuff in the future (aka fiddle and shoot myself in the foot).

So - are there other issues? Are there ways of mitigating the issues? Are there sufficient warnings about the possible issues? (If someone wants to point me at a detailed KB article that I've missed, feel free to do so...) Anyone who feels confident in their ability to stick to the provided templates, fine - but that's not me. I know that I'll fiddle....

[ColeValleyGirl]

A general observation, orthogonal to the main question...

But the major change was where I replaced the fairly amorphous item "Location" with "Event Place" and "Indexed Place" (Event Place = "Haslington", say, and Indexed Place = "Crewe Sub-District, Nantwich Registration District").

Everyone works differently, but this feels to me as if you're using Source records for more than just identifying the source and where it can be found. For somebody else to access that certificate, they do not need to know the Event Place... which is (presumably) captured in the relevant Fact anyway.

I'm also wondering how you've managed to add another 6 extra fields. One for "how the copy in my hand had been produced", but the other 5?

[AdrianBruce]

... this feels to me as if you're using Source records for more than just identifying the source and where it can be found. For somebody else to access that certificate, they do not need to know the Event Place... which is (presumably) captured in the relevant Fact anyway. ...

I agree. But accepting the challenge...

My justification, such as it is, is inspired by what I believe Elizabeth Shown Mills says on the topic. Specifically on https://www.evidenceexplained.com/conte ... s-citation she says "Citations are an analytical tool we use to help ourselves reach accurate conclusions". And further:

... a thoughtful identification of each source, at the time we use it, will do two things:

• provide standard citation details for the source and its location; and
• identify the source in such detail that we know exactly its nature, its strengths, and its weaknesses.

It's that analysis angle that leads me to stuff more in the Source Record than others might. (Especially others who postpone analysis until a later stage than entering their data into FH. ) To be quite honest, I do wonder if I am doing the right thing by entering the actual place of a birth into the citation, but if I omit the physical event detail, the citation distracts me by being misleading. For instance, in the 1939 Register, my great-uncle would be cited as being in Crewe Municipal Borough - yet he wasn't - he was in a village a couple of miles outside the town. Similarly, if I cited "Crewe Sub-District, Nantwich Registration District" against my mother's birth certificate, that would instantly make me say - "Hang on, wasn't she born in Haslington? Recording in the citation that it was "Shavington" and "Crewe MB" for the 1939 and "Haslington" and "Crewe Sub-District, Nantwich Registration District", avoids the distraction by explaining the hierarchy.

One point there is that such analysis starts with being for my benefit and others might not be distracted. But perhaps that's what analysis is for...

[ColeValleyGirl]

I'm with you on the analysis front, which is why (for example) I agree that the 'format' of the document needs to be recorded, as well as 'the source of the source'.

But when you say the source/citation is 'misleading' , it isn't misleading about the source and how to access it; the place named in the source may not be the place of birth, but it's definitely the place of registration (the event that produced the source). It isn't your Great Uncle being cited in Crew Municipal Borough, it's the source bring created there. I'd be distracted by an extraneous detail such as the place of birth, but you're not creating your sources for my benefit

[AdrianBruce]

... I'm also wondering how you've managed to add another 6 extra fields. One for "how the copy in my hand had been produced", but the other 5?

• Location is replaced by Event Place and Indexed Place (nett gain of 1);
• Creator to record the organisation responsible for creating the original entry;
• Contemporary copy - Yes / No to indicate whether this was produced for the real world or for a genealogist;
• Contemporary no. - if it was a contemporary copy, what number does it display? That's in essence because if it's a contemporary copy, it wasn't produced by asking the office for Quarter, Volume, Page, etc, and the number is the closest thing I have to an identity;
• Certified Copy - Yes / No;
• Creation method - Handwritten, facsimile, etc. (Enum to ensure I get consistent terms);

Anyone is free to wonder about the significance of those extra items - me included - but when I created that template, the issues of creating / updating template items did weigh in my mind, so I thought I'd rather over-engineer than under and have to add or (worse) split items. The interesting thing about the last four is that they all sit within the single Publication field of my generic source skeleton record for this record type, where it says

This is a ?contemporary? ?certified? handwritten/facsimile copy, made dd mmm yyyy

That hint to me is backed up by further explanations in the Source Notes of the skeleton source record. It may be that a Jedi adept could therefore reduce my count of new template items, but frankly I'm less than keen on multi-line functions if that's what it takes.

Or that's my excuse...

[AdrianBruce]

...
But when you say the source/citation is 'misleading' , it isn't misleading about the source and how to access it; the place named in the source may not be the place of birth, but it's definitely the place of registration ...

True. As you indicate, it's possibly the psychological effect, which is personal to the researcher. I do wonder if I'm liable to miss finer distinctions around other towns - for instance, would a researcher from Oldham record a lower level detail than me and say (about me), "He clearly doesn't understand the first thing about the difference between Oldham Mumps and ...."?