* [Wish List entry 599] Allow citation fields in bibliography on sources

[rcpettit]

I finally had to give up on Rootsmagic 8 and moved to FH. So far not so bad but I don't know if this has been asked before but I can't find it in searches. I'm a lumper and I found out I can't use citation fields in the biography bibliography for sources, getting error messages say that field can't be used in biography bibliography. Is there a reason for this restriction? It would be nice if this restriction was removed.

[AdrianBruce]

We've had a long discussion on this somewhere.... (Given Murphy's Law, it might even have been in the Mailing List).

Best guess at the basic reason for the restriction is that if you imagine a source as a physical book, with citations referring to pages within the book, then the bibliography is the list of book titles and if you include the pages, you haven't got a list of sources (books), but a list of citations (pages).

It also becomes much more complex to assemble a bibliography using citation details, since you need to go much deeper, closer to the facts concerned, whereas a non-citation bibliography can be compiled just from the source-records.

I hesitate to invoke her name, but I'm fairly certain that Elizabeth Shown Mills refers to Source Lists, not Bibliographies, in Evidence Explained. This would again suggest where the expectation over the contents lies - i e., it's about Source (Records).

[davidf]

Elsewhere I have seen reference to an "Index of Citations"

In such a (lumper) index, the top level is each (lumped) source and under that is each citation of that source with the page numbers of the pages on which the citation appeared.

I've not seen one in genealogy.

[rcpettit]

So basically, The biography would just contain something like "1920 U.S. Census" instead of " 1920 U.S. Census, Florida, Lee County"?

[LornaCraig]

I notice that you have referred in the topic title and in your posts to a ‘biography’. In case there is any misunderstanding, the section which (optionally) can be generated at the end of a report is a bibliography. In most academic works a bibliography is a list of sources (typically books, or archive collections of documents) which have been consulted.

As Adrian has said, there’s a difference between a list of sources (books, or collections of documents) and a list of citations, which are references to specific pages, or particular documents within a source.

As a ‘lumper’ you will have made multiple references (citations) to each of your sources, citing particular pages/documents/names within the source. Those details are shown in the footnotes, not the bibliography. So the bibliography might give the full title, author and publication details of a book. The footnote (or short footnote) might then refer to the book simply by the name of the author plus a page reference for example “Smith, page 162”.

In your example you would list the US 1920 census in the bibliography, including its full official title and, perhaps, the form in which you accessed it (online? In a physical archive? Transcriptions only or images?). The footnotes could then refer to it simply as “1920 US Census” followed by the state, district, page, household. You could if you wish have a separate source for each state and county so that "1920 US Census Florida, Lee County" is a source in its own right, leaving only the district/page/household to go in the citation details, but then you are moving towards being a 'splitter'.

[cwhermann]

I too am a lumper for the most part and switched fro RM and had to come to terms with this concept. After running several trials with the help of CP help desk, even though though RM allowed the use of Bibliography fields in the citations, it does not work when one creates reports in RM. FH and RM both allow Bibliography fields to be used in citations, so I have developed a work flow where I decide at what “level of lumping” I want for a given record set and create the Bibliography template accordingly. I the create the footnote template using whatever bibliography fields and citation level fields as required.
My US Census records are “lumped” at the county level, e.g. I have a Source Record for each State and County. Some of the same information (or fields) used to create the Bibliography, are also used to create a citation for a specific household, in a specific precinct, in the county and state.

[davidf]

I have developed a work flow where I decide at what “level of lumping” I want for a given record set and create the Bibliography template accordingly.

The level at which you chose to "lump" is very much up to you - for me it really comes down to how I want to describe "what I looked at" (source) against "where in that source" (citation) I found the specific information I wanted.

I lump the England and Wales Census for say 1881 into a single source, but I lump the Scottish Census for 1881 separately - because even though on say Find My Past accessing looks similar (through a Census search), the Scottish census is different as there are no images available on FMP - so the source is qualitatively different to me (I could not "look at" the images, so reliability is dependent on the transcription). The Scottish Census on ScotlandsPeople which is pay-per-view has images, so as far as I am concerned it is a different source.

When "citing a source" in a document (whether FH produced or crafted on a Word Processor - my preferred way) the referencing has to "work" in that you need to be able to follow it to unambiguously get to the Source (the what) and the specific information in the citation (the where).

Before everything become heavily computerised, it was "obvious" how this worked. You have a list of your sources (the bibliography). This list is usually sorted in some "useful way". I use the "short title" in FH to hold this information. A line of Bibliography might therefore read (Short title in bold):

1861-04-07 England & Wales Census, "RG9 General Register Office: 1861 Census Returns" The National Archives, Census Transcript Search, 1841-1911 [database online]. TheGenealogist.co.uk 2022.

One of the benefits of lumping is that this list is kept manageable and you can quickly scan it to see the sort of sources that have been consulted.

Within my document, I need a reference to this source and a citation to tell me "where within the source" I found the specific information. Many refer to this combination as a "citation". In the case above my "citation" might read:

1861-04-07 England & Wales Census, RG9 PN:3929 FN:? Page:27

That is too big to sit unobtrusively in the text (say in brackets), so I will put it into a foot-note or an end-note (or for an illustration, in the caption). Note that the beginning of my "citation" is the short title of the source and it is then followed by what I have put into the "Where within Source field" - the latter usually pasted direct from the on-line transcription - hence the ? for folio number in the above case (it was not indexed and is not on the image).

Anyone wanting to check my source, can read the citation, extract the short name for the source, find that source within the bibliography and then using the citation details get to the actual page involved.

If you let FH "do the work" it should ensure that this "chain" works from the point in the document where a fact is reported, via the citation to a line in the bibliography.

Sometimes (as above) the citation can be complex and hard to decode. As a lumper, I have the source notes available to hold information such as.
"1861 Census:
Example 'where within source': RG9 PN:aaaa FN:bb Page:cc
RG9 is the archive series code
PN: is piece number
FN: is folio number (not always indexed/transcribed)
Page: is the actual page number
To identify the contents of a particular piece, put the reference [series code]/[piece number] into the TNA advanced search

[tatewise]

I've changed biography to bibliography in the Subject line as that is what Source Templates define.

[AdrianBruce]

So basically, The biography would just contain something like "1920 U.S. Census" instead of " 1920 U.S. Census, Florida, Lee County"?

If your source-record refers to just the 1920 US Census, then yes.

It's really only just occurred to me (because I'm a splitter, 98% of the time) that many lumpers who use other software, must use their intended bibliography to determine what their source-records will be. So if their intended bibliography is down at the county level for US censuses, then they create their source-records / master-sources / whatever-they-call-them at the county level. This is presumably way too late for yourself, of course - sorry.

FH can have bibliography entries that are higher level than source records - but not the other way round. E.g. it can have "1920 U.S. Census" in the bibliography and "1920 U.S. Census, Florida, Lee County" in the source-record (provided the templates distinguish the state and county separately, so far as I can see). But not the other way round.

[LornaCraig]

Should this topic be moved to the General forum? I suspect that the OP can achieve what he wants by choosing a suitable ‘level of lumping’, e.g. for 1920 US Census use a separate ‘lumped’ source for each state and county so that ‘1920 US Census Florida, Lee County’ is a source in its own right, as several of us have explained, with separate citations to details within it.

That may leave the OP with a problem if his existing sources imported from RM are ‘lumped’ at a higher level, but whether and how they could be subdivided is a matter for a different discussion.

Before V7 there was no option for a bibliography so everythning had to be shown in the footnotes. IMO if citation fields were allowed in bibliographies then a bibliography and a list of source citations (footnotes) ends up being almost identical, although perhaps arranged in a different order, which defeats the point of distinguishing between them.

[tatewise]

I've returned this to the New Wish List Requests forum, where it was this morning. Don't know why it got moved.
Existing users who may have imported 'lumped' Source Citations cannot easily change to 'splitter' just to get more granular Bibliography Formats.

This is my perspective...

Method 1 'splitter' Source records hold all the details and the Citation-specific fields are mostly empty.
Most FH supplied Source Templates use 'splitter' mode.
So any level of granularity can be chosen for the Bibliography Format and may be virtually identical to the Footnote Format if the user wishes.

Method 2 'lumper' Source records hold fewer details and the Citation-specific fields hold the rest.
It is perfectly feasible to define Source Templates for 'lumper' mode.
So why not allow the same level of granularity for the Bibliography Format as would be possible in 'splitter' mode for the same documents, even if they are virtually identical to the Footnote Format?
Surely that is a reasonable choice to offer users. They should not have to jump through hoops to convert their existing Source Citations (that may have been imported) in order to achieve the same style of Bibliography as 'splitter' mode.

I don't see why 'split' Sources should have any different Bibliography options than 'lumped' Sources.

[ColeValleyGirl]

Mike, Lorna suggested it be moved and I agreed.

[tatewise]

Sorry, but there was no indication of who moved it.
In fact, I was about to draft a Wish List proposal along the lines of my posting just now.
I don't see why 'splitter' Sources and 'lumper' Sources cannot be allowed to set the same Bibliography granularity if users wish.

[ColeValleyGirl]

Sorry, but there was no indication of who moved it.

Well, it had to be a moderator, didn't it? So it's in the moderator logs...

[ColeValleyGirl]

I don't see why 'splitter' Sources and 'lumper' Sources cannot be allowed to set the same Bibliography granularity if users wish.

Depends how far you want to redefine the English language...

Seriously, I have no problem with that, although the result wouldn't actually be a bibliography in the strict sense of the word...

So maybe ask for something additional, with a different name, rather than confuse with misuse of a word.? But not perhaps the 'Index of Citations' that davidf mentioned, as that's yet another beast.

Edited to add: When bibliographies were added to FH, a major part of the argument in favour was to lend credibility to FH's reports. It would be a backward step to lose this credibility by misusing the word, or confuse potential users who expect standard behaviour.

[ColeValleyGirl]

Related:

Field codes for Source Template Definitions (20743)

[LornaCraig]

So any level of granularity can be chosen for the Bibliography Format and may be virtually identical to the Footnote Format if the user wishes.

Yes, but as I said that rather defeats the purpose of introducing bibliographies in V7, which were (presumably) intended to be significantly different from the footnotes. I can’t image why any one would want to have virtually identical formats for the two.

They should not have to jump through hoops to convert their existing Source Citations (that may have been imported) in order to achieve the same style of Bibliography as 'splitter' mode.

But unless this suggestion is adopted by CP at record speed, a one-off solution (plugin?) to convert heir existing lumped sources into lower level lumped sources (not necessarily fully ‘split’) might meet their needs. There would then be no need for the wish list request.

When bibliographies were added to FH, a major part of the argument in favour was to lend credibility to FH's reports. It would be a backward step to lose this credibility by misusing the word, or confuse potential users who expect standard behaviour.

Exactly. Personally I find the distinction between bibliography and citations (in footnotes) intuitively sensible and it mirrors the way a bibliography works in academic publications. If they were reduced to the same level of granularity I would think “Er, why are they doing the same thing? What’s the difference? What’s this so-called bibliography for?”

[tatewise]

I understand all the sophisticated arguments, but as FH stands, a 'splitter' can effectively use any field to produce the Footnotes and a Bibliography of any granularity (even with a Bibliography more granular than the Footnotes if they wish).
Maybe it goes against all the concepts but the FH software does not prevent it and there is no explanation of the concepts.
If the Bibliography is not meant to be more granular than the Footnotes then why does the software allow it?
The only reason such users have that flexibility is that there are few if any Citation-specific fields.

A 'lumper' cannot produce the same Footnotes and Bibliography for exactly the same actual sources as a 'splitter' just because the citations are structured differently.
Several users have questioned why, especially those who migrated from a product that allowed such a Bibliography.
Many products effectively only offer 'lumper' mode and users wishing to migrate between products would want to retain that structure in FH.

It would seem an almost trivial change to the software to cater for the 'lumper' scenario above and avoid disappointing users.

The FH documentation states the rules about which fields can appear in which formats but gives no explanation along the lines Lorna and Helen have put forward, so users are left in the dark as to why the rules exist. IMO it gives the impression that FH is still 'splitter' focused.

[LornaCraig]

The FH documentation states the rules about which fields can appear in which formats but gives no explanation along the lines Lorna and Helen have put forward, so users are left in the dark as to why the rules exist.

Well I don't think any explanation is needed because as I said in my last post "Personally I find the distinction between bibliography and citations (in footnotes) intuitively sensible and it mirrors the way a bibliography works in academic publications. If they were reduced to the same level of granularity I would think “Er, why are they doing the same thing? What’s the difference? What’s this so-called bibliography for?”

I know we often have to explain to users that some terms have a specific meaning within FH which differs from their meaning in everyday language, but in this case FH has introduced the term 'bibliography' in a way which does match its use in the real world. You seem to be saying that it should be changed deliberately so that FH uses the term in a different way.

As Helen said, "maybe ask for something additional, with a different name, rather than confuse with misuse of a word.?"

[ColeValleyGirl]

Lorna just said what I was about to say -- bibliography is a 'technical term' in this context with a well-specified meaning (I could point to a large number of how-to books for genealogy, courses, etc.) that explain the term, but you'd accuse me of being sophisticated again.

I'm not arguing that what you're proposing shouldn't exist, for users who don't want/can't go to the effort of redefining the way their sources are lumped to get the same result as their previous package (if that's possible); I just don't want to muddy the waters about bibliographies.

(Re granularity, it might be worth pointing out that the standard Essential templates, based on Strathclyde, actually do some lumping on behalf of splitters, otherwise you'd end up as Lorna described with detail being duplicated for no good reason, and you might as well put all the Source Citations at the end of the report and call it a bibliography and leave the bibliography out.)

Screenshot 2022-12-17 181633.jpg (45.26 KiB) Viewed 1910 times

[AdrianBruce]

...
It would seem an almost trivial change to the software to cater for the 'lumper' scenario above and avoid disappointing users.
...

Unless I've missed something, I cannot see how allowing citation fields in a bibliography is almost trivial. A bibliography can be run off just by examining the Source Records. (In fact, Elizabeth Shown Mills refers to it as a Source List, rather than Bibliography - I rather wish we'd used her terminology - it would have been more explanatory). To include the citation fields, requires "dropping down a level" into the citations - a completely different sort of entity.

I understand all the sophisticated arguments, but as FH stands, a 'splitter' can effectively use any field to produce the Footnotes and a Bibliography of any granularity (even with a Bibliography more granular than the Footnotes if they wish). ...

But that Bibliography is only more granular than the printed footnotes. It can only be more granular than the printed footnotes if detail has been excluded from those footnotes.

I would agree with Lorna and Helen that CP should cater for the requirements that people clearly have, by producing a new report. The Sources and Citations Report appears to go through the data required - sources and then citations - maybe even further since it also looks at the facts. This is illustrated in the screen shot below:

Screenshot 2022-12-17 202134.jpg (97.19 KiB) Viewed 1799 times

Suppose I call this idea "Extended Bibliography" for the sake of simplicity (I'm not saying the the screenshot is what an Extended Bibliography should look like - only that it is possible to get to that sort of data in an existing report).

Then the Source Template Definition Editor would need an extra section for the Extended Bibliography Format, to go below the existing Bibliography Format.

The Extended Bibliography Format should then have the ability to include the citation fields.

Reports should then have an option to show either Bibliography or Extended Bibliography or neither. The Extended Bibliography should not be called a Bibliography in the interface or in the outputs - because it isn't.

[davidf]

Adrian

How would that report look if the citations did not match the source but were in the nature of a time offset from the start of a recorded interview? The report seems to say that all 14 facts are supported by the Source in its entirety - with no "Where within Source" data - which is indicative of a lumped source.

The Source is "Oral testimony of Henry Dowling, May 2000"
Then each specific fact would be supported by a citation ("Oral testimony of Henry Dowling, May 2000", 00:02:35 from start)

With a large "Lumped" source say "Wikipedia" (!), this report would not really tell you very much?

Or am I missing the point in this convoluted debate?

[AdrianBruce]

The report is not intended to be a prototype - it's only my intention to illustrate that something already exists in that area

... The report seems to say that all 14 facts are supported by the Source in its entirety - with no "Where within Source" data - which is indicative of a lumped source. ...

It's a Split source - the template has nothing in the citation part other than the Primary / Secondary, etc. I didn't claim it was a good source!

... With a large "Lumped" source say "Wikipedia" (!), this report would not really tell you very much? ...

As I said, it's not meant to be an answer but here's Wikipedia citations from my production database in the same report format. Firstly the header (of this lumped source)

Screenshot 2022-12-17 230628.jpg (27.23 KiB) Viewed 1774 times

And then two detailed citations.

Screenshot 2022-12-17 230121.jpg (72.48 KiB) Viewed 1774 times

As you can see the report does give all the Where-Within, Assessment, an abbreviated(?) Text from Source, Notes, etc. This simply illustrates that the required items for the Extended Bibliography are there. I presume that a real Extended Bibliography wouldn't be printing details of the facts.

[jnunnally]

My background is RootsMagic and I'm a lumper. I'm getting familiar with FH, but I've run into this issue with citation fields in bibliographies, in particular, an "access date."

There is a rational reason to at least allow an "access date" in bibliographies when dealing with sources from the web. This concept is covered in Mills' "Evidence Explained" where she says, "The reorganization of a website could eventually make our cited keywords and paths unworkable. By recording the access date, we may have a reference point we can use to retrieve the material from an Internet cache such as WayBack Machine (http://www.archives.org)."

In other words, when I cite material from a web page, I am actually citing a snapshot of the material at a particular time. Unlike a book or a periodical, It is subject to change both in format and in content over time and that fact should be acknowledged. My bibliography entry is citing not only a web page but one particular iteration of that web page's evolution over time. An access date is very similar to a publication date in this context, but as a lumper, the access date is part of the citation, not the source.

Although I understand the reasoning for the restriction, it seems to be a judgment call based on the assumption that users would have to be naive to want to put citation fields in bibliographies. That is not always the case.

What about a warning instead of a blanket restriction? Or perhaps a preference setting?

[AdrianBruce]

My background is RootsMagic and I'm a lumper. I'm getting familiar with FH, but I've run into this issue with citation fields in bibliographies, in particular, an "access date."

There is a rational reason to at least allow an "access date" in bibliographies when dealing with sources from the web. ...

Interestingly, I've just done a quick scan of some major types of record in Evidence Explained and I notice something that I hadn't seen before.

The (full) Access Date is present in the Reference Notes, (i.e. at Citation Level) for all those good reasons that you mention.

But in the Source List (i.e. Bibliography) it's not a full date but just the year of access. Indeed, in the Source List entry that specifies a forum (the Reference Notes define the individual threads in that forum), the "year" is actually a range of years.

That would make a lot of sense because if I were to create a Source List entry for an online set of parish registers, the likelihood is that most of the register entries will be accessed on different dates - if I were to put the full access date into the Source List / Bibliography, then I'd have nearly as many Source List / Bibliography entries as Reference Notes (citations) - which kinda destroys the point of having a separate Source List. But if I include a year (which needn't be part of the citation but up at the source record level) then I get a much more manageable (because shorter) list for my Source List / Bibliography.