I can see that support for layered citations would be useful. A hierarchy of sources is one way that they could be implemented, but not the only way.
Arranging sources into collections could also make source management easier. We could organise our sources into a catalogue in a similar way to that done by archives.
I think that this topic is worthy of further discussion.
Enno
I am using this forum to explore options and learn how other people
approach an issue.
So for me the logic of GRAMPS in this area is difficult and indeed for a
long time I did not use Citations as such putting all the information I
required into Event Notes other than Census.
My thought process can handle the Places Structure easily ie
Country
County
Town
Urban District
transferring this to Citations
Repository
Indexer/Transcriber/Aggregator
Source
Citation
Example
The National Archives(Kew)
FamilySearch
1851 UK Census
HO107/ā¦
Because there is no 4th tier in GRAMPS this is why now I put Ancestry
etc as Repository. Which is a work round for me which may or may not be
worth exploring further or not.
Phil
To answer your question, my approach is to just record some info that shows I found a relevant source/citation. I use Ancestry, FMP and FS for most of my work and these are the Repository and then their index and then the record. The reference material gets copied into the Page field, and I also append the URL in this field. I need the URL in the Page field because it is needed in the gedcom I export. I also copy the transcribed record into the Note field except for Census records.
I only use Census records to verify family members and to record where the person is at that time. All my Place locations have the Lat/Log entered so that they all show on the map. I donāt keep images of the records. I like having a copy of the transcribed record because many times I have needed to go back and look at it again when dates donāt match. And just because I have the URL doesnāt mean I will get back to the same record. Iām a minimalist and lazy so I only record what I feel I might need and no more.
Hello Phil,
I uderstand, and I went through the same questions years ago. And compared to the 11 fields that I find on our national portal for genealogists, your puzzle looks a bit easier.
In a case like these, where youāre dealing with indexed data, one can argue that you donāt really need to cite the reference numbers of the original documents, because you didnāt really look at those. You can store them in a note, so that you can use them when you have time to visit the archives in Kew, or when they have browsable images on their own site. Putting them in the citation suggest that you actually looked at those, which is probably not true if you just used the index.
If you just used the index, the fields will probably fit quite well in Gramps, like for this record:
āEngland, Yorkshire, Parish Registers, 1538-2016ā, , FamilySearch (FamilySearch.org : Thu Oct 26 20:34:01 UTC 2023), Entry for Philip Lesley Wharram and Robert Sharpe, 6 Dec 1908.
For such an entry, I put the 1st part in the source title, and add FamilySearch as the publisher, not the repository. And by adding it there, I make sure that the FanilySearch name is included in the bibliography, in the endnotes.
And unlike Dave, I donāt put the URL in the page field, but in the citation notes. And I do that, because I want to see the āEntry for ā¦ā part in the citation, so I put that in the Page field. I use the date field for the date that the record was made, which is equal to the event date in case of a baptism.
In this case, there is an extra layer, which you can see when you follow the link that was included in the citation. And that gives some clues about the original, like these:
| Document Type | Baptisms |
|---|---|
| Page Number | 29 |
| Reference | PE 49/4 |
| Affiliate Name | East Riding Archives & Local Studies Service |
And if youād cite the original instead of the index (database) the choice Iād make would be creating a repository for the affiliate, adding the reference to the link that connects the source to the repository, and put the page field in the page field of the citation.
The source title would probably be the thing that is behind the reference.
Elizabeth Shown Mills has a lesson for this here:
And it would of course be nice if Gramps would support these kind of layered citations. They are supported by GedcomX, and sometimes you see that the formatted citation on FamilySearch shows the layered part too, often preceded by the term ācitingā.
P.S. I didnāt link this source to a person yet, but if you do, please make sure that you correct his birth date. His current birth date is the date of the baptism.
P.P.S.S. ESM suggests that you treat the web site as a book, and the database as a chapter, but we donāt have a field for that chapter, and I see the database as the real source, so I put that in the title on purpose, with Ancestry or FamilySearch as the publisher.
Hereās a citation that Ancestry made for me, when access to those censuses was free:
As you can see here, the database is in the title, just like I suggested, and the repository can be safely removed, because thereās enough mention of the site in the title already.
Another thing is that they put all the info about the original document in the page field, as key value pairs, separated by semicolons. And that helps a lot if you run out of fields.
Hi Enno
I think we are singing of the same hymn sheet generally here and doing
the same or similar things.
One major difference is I only use a URL in two places
Re updating/correcting FamilySearch I just donāt it is a complete waste
of my time, I used to do it years ago and then somebody else would come
along and undo/change it.
Thanks for using the Wharram Yorkshire link where the name obviously
originates, unfortunately I can only get back to the 1780ās and that was
in Cambridge and London.
Phil
Hi Phil,
I think that weāre much on the same path indeed, and the only URLs that I store are for wikipedia articles about notable relatives, because they are quite stable. And in general, the ones on FamilySearch are quite stable too.
Thereās one thing I want to show you though, and thatās an entry on the Dutch WieWasWie site, which is a consumer portal to our archives. In that, you can see 11 fields, and Iām curious whether you see similar things in the UK.
The screenshot comes from the English version of the site, and it shows the amount of fields that we see in most of our finds. These fields have been agreed upon by all participation institutions (repositories), and in some rare cases, one may see even more than these.
Iām showing this, because entries like these seem to call for a flat model, because thatās the easiest for data entry, even though they do represent 4 layers. And one thing you see here is the Collection layer, represented by the Archive number and the Collection name.
Note that the Sourcenumber shown here is actually a volume/page.
It would be lovely if other users see similar things that can be used as an inspiration for a new form in Gramps.
Enno
I am going to be really impertinent here and say OTT, however if this is what required in Holland I can see why you are against additional complications in GRAMPS.
I have seen many but not all UK systems and I cannot think of any that come anywhere close to this. Obviously someone will set out to prove me wrong but thatās life.
Phil
I thought that the new (for 5.2) support for CITE (Citation formatter) plugins was going to be overrides to handle the exceptions where the basic legacy Citation data structure was insufficient. (And that might be true. We wonāt know until someone releases the 1st CITE addons.)
But from looking at the interface, it looks like the choice of CITE plugin will be a global setting in Preferences, not a granular approach like the overrides for Name grouping and format.
