* Bibliography Citation-specific Metafields

[ColeValleyGirl]

I don't think FH says that its Bibliography is intrinsically bound with EE concepts, so its features could be expanded.

They do however say a Source Record is created for every Source... which currently exclude Citation Level fields. I'm not arguing those shouldn't be an option, just pointing out it's a fairly significant architectural change.

[arthurk]

I have serious practical problems, however, with EE's idea of recording the year of access in the Bibliography, not least because some of my parish registers and censuses (in particular) might have a bibliography entry for every year since I started this lark, were I to follow the same process. So 20 different bibliography entries all for "1901 Census, Ancestry, accessed yyyy" (substitute actual year for "yyyy"). Hm. (NB - I would indeed lump my bibliography entries like that)

Similarly, Adrian, I'm afraid I have reservations about including "Ancestry" as part of a bibliography entry. To me, the census itself is the source, and (in spite of what EE says and has been mentioned above) the website is the equivalent of a repository. Were I to do it your way, each of my census years might have 4 or more entries covering Ancestry, FindMyPast, The Genealogist and FamilySearch - to say nothing of the record offices or PRO (as it then was) where I viewed census returns on microfilm before they could be seen online.

As an aside to this discussion, you'll see that I'm still using FH6.2. With census sources I have so far been a lumper (and I currently have no images or full transcriptions of entries within FH). When I move to FH7 I am quite tempted to move to split census sources and include a tabulated transcription of each entry. If I do this, clearly I'll have dozens of sources for each census year rather than just one, but would FH then be able to combine these into a single entry in a bibliography (eg "1851 census", with or without Ancestry etc), or would I have to edit any output manually after FH has done its stuff?

[tatewise]

arthurk, you can combine them into a single entry in a Bibliography (e.g. "1851 census", with or without Ancestry, etc) simply by omitting the finer details from the Bibliography Format template.

Helen, how do you know it would be "a fairly significant architectural change"?
It would appear to be no more complex than the Footnote and Short Footnote that do allow Citation-specific fields.
The Sources footnotes are just a computed list of items very similar to the Bibliography computed list of items.
The computational algorithms would appear to be identical.

[ColeValleyGirl]

By verification, I mean verifying that the source has been recorded accurately, not that the source information is correct. For me, an individual source citation is more about the former, not the latter.

I'm not sure you can ever be certain that a source has been recorded accurately. Partly burnt or illegible records, for example -- it isn't always possible to be certain what the source content is.

[Mark1834]

Exactly- the census is the source, and the website/url/date is how and when you accessed it. This discussion seems to be between two camps - those who want to record full details of how and when, and those who aren’t bothered (my analogy of which shelf in which library the printed book was on). Again, neither is right or wrong, they’re just different approaches, and ideally FH would cater for both.

[ColeValleyGirl]

Helen, how do you know it would be "a fairly significant architectural change"?

If you move away from 1 source == 1 source record, the citation details change but what else does? The help file would need a revamp; but how embedded in the internal architecture (which we can't see -- only the gedcom output architecture) is that one-to-one correspondence? How are you going to construct a query that matches the contents of a report for example?

There would also be a fair bit of cognitive dissonance for users accustomed to the old one to one correspondence if they inadvertently used the 'new feature'.

Long term FH user: a source is a source! What is this rubbish about including Citation fields? If I'd wanted to use RM I would have bought it!

I'm not saying it shouldn't be implemented but it does need to be thought through?

[AdrianBruce]

... Similarly, Adrian, I'm afraid I have reservations about including "Ancestry" as part of a bibliography entry. To me, the census itself is the source, and (in spite of what EE says and has been mentioned above) the website is the equivalent of a repository. ...

Yes, Arthur - I'd refer to your idea of just having the census as the source, without the web-site, as referring to the logical source of the information, whereas I prefer to put the web-site in and then I have the physical source of the information. I've never really compared the two methods because my heart says that the source is what I consult and I'm not consulting (say) the 1861 census itself - that, IIRC, is down a salt-mine at Winsford! But if you can work it through, and get all the information needed, then fine.

Incidentally, the EE view that a web-site isn't a repository but a publisher is an over-simplification, so far as I remember. What she actually said in the bit that I read, was that it was quite easy to imagine a case where FamilySearch, say, was the author, publisher and repository for a source. She suggested that was a degree of overkill! I suspect that's where the idea that a web-site can't be a repository comes from - the desire to cut down on repetition.

[arthurk]

arthurk, you can combine them into a single entry in a Bibliography (e.g. "1851 census", with or without Ancestry, etc) simply by omitting the finer details from the Bibliography Format template.

Thank you, that's reassuring to know.

[Mark1834]

arthurk, you can combine them into a single entry in a Bibliography (e.g. "1851 census", with or without Ancestry, etc) simply by omitting the finer details from the Bibliography Format template.

If Arthur' are like mine, all are generic sources with a typical title along the lines of "Census 1881: Details of location and address". I couldn't find a way of "easily" combining those to a single "Census 1881" reference in the Bibliography. How would you do that without making a global change that impacts all generic sources?

[AdrianBruce]

Exactly- the census is the source, and the website/url/date is how and when you accessed it. ...

Here's my personal take on date of access... It's pointless because it's not actionable. I did faithfully record them - still do where it comes to text based indexes such as FreeBMD (not totally sure why). But I gave up on date of access for images because I can't do anything with it. If I record that I accessed a census image in 2010, I can't do anything with that year - short of a TARDIS I can only access this year's census image. Yes, the image and / or index may have changed - but I don't need the date of access to tell me that. And if I thought that things didn't change - adding the date of access wouldn't convince me otherwise...

It's useful to have the year of publication because I can go to the appropriate library to find the 2003 edition of "My Life in Kenya", say. But I can't bring up the 2003 edition of the 1881 census....

Your opinions may vary...

[AdrianBruce]

... If Arthur' are like mine, all are generic sources with a typical title along the lines of "Census 1881: Details of location and address". I couldn't find a way of "easily" combining those to a single "Census 1881" reference in the Bibliography. How would you do that without making a global change that impacts all generic sources?

Oh bother. And there was me hoping to do that...

"Well done, Pike. I was waiting to see who'd spot that..."

Captain Mainwaring, Dad's Army....

[tatewise]

Mark, arthurk clearly stated that he would be converting his 'lumped' generic Source Citations into 'splitter' templated Source Citations, so that would allow any new restructuring that is necessary. With templated Sources, the Title is derived from the metafields just as is the Bibliography.

If you persist with generic Source Citations then your options are much more restricted as there is only one Bibliography template in Tools > Preferences > Sources > Generic Source Formats... so my statement does not apply.
It might be possible to use FH functions that detect the Titles that start with 'Census' and only include the first dozen characters in the Bibliography or split it on the colon (:). Similarly, other Source Title formats could be accommodated but it would get complex. You need to review the options described in the FHUG Knowledge Base for migrating from FH V6 generic Sources to FH V7 templated Sources.

[ColeValleyGirl]

And while we're on generic sources... Another area CP would need to reprogrammed to include Citation fields in reports, if they were going to do the job properly rather than just bodge a change to templates.

Re access dates… I find them useful:

Fred: you said site X has that source!
Me: it did when I looked at it but they don't have the rights any more.
Mary:your transcription is rubbish.
Me: they've provided a new image.

It does depend who your audience is.

[arthurk]

Here's my personal take on date of access... It's pointless because it's not actionable....

It's useful to have the year of publication because I can go to the appropriate library to find the 2003 edition of "My Life in Kenya", say...

Which leads me back to the thought that a bibliography shouldn't generally include anything other than a title or lumped category: if you look at the same book in more than one place, do you record "My Life in Kenya - Crewe Library" and "My Life in Kenya - Little Gloomsbury in the Marsh Library" etc etc?

The circumstances where I think it might be appropriate to add more would be if there is a unique repository, such as "Adrian Bruce letters and papers, Bruce Family Muniments, Bruce Park, Cheshire".

[ColeValleyGirl]

For non unique items I agree. You don't need to identify where you viewed a source.

For unique items you do.

And since providers like ancestry and fmp use different images and indexes very often I wouldn't call them non unique.

[Mark1834]

Mark, arthurk clearly stated that he would be converting his 'lumped' generic Source Citations into 'splitter' templated Source Citations

No he didn't. What he said was "When I move to FH7 I am quite tempted to move to split census sources and include a tabulated transcription of each entry". It would be perfectly possible to do that with generic sources as well - you have assumed he will also convert them to templated ones.

As it happens, it doesn't impact me as I don't use the Bibliography option, but I didn't want your reply to mislead anybody wanting to using a bibliography with generic sources.

[Mark1834]

For non unique items I agree. You don't need to identify where you viewed a source.

For unique items you do.

And since providers like ancestry and fmp use different images and indexes very often I wouldn't call them non unique.

IMO, it's a case by case judgement, and more about whether the content is unique. For example, different repositories may have different, but equally clear and good quality, images of the same original document. It is then down to personal preference how much we record. If we follow that logic to its natural conclusion, we should always record the when and where, just in case a different image becomes available in the future.

It's a personal preference of style, not an absolute requirement of what we all "need" to do...

[cwhermann]

Great discussion and it has certainly expanded way beyond my "issue" with specific templates dealing with a specific type of record - online images of local jurisdictional records.
Below is an image from an EE quick check model for an online image with the website and year in the bibliography.

EE page 381.PNG (111.27 KiB) Viewed 4471 times

I think as a starting point, we have to remember that EE considers a website as a publication, (just like a book), not a repository and as Mark pointed out

"Neither is a website url from some point in the past, as it cannot be verified if the website has been reorganized (unlike a physical book, where old editions remain accessible). The science equivalent is citing “personal communication”, which is strongly discouraged for just that reason - it cannot be verified.

And if one is coming from the premise that the website is a publication, it is for this exact reason, EE believea the Year needs to be in the Bibliography, just like you would indicate the edition of a printed book. ( and we have not even started down the rabbit hole regarding URLs vs. way points )
As I see it, the drawback of taking this approach is what Adrian pointed out- you can potentially end up with multiple Bibliography entries, all identical except for the YYYY.

One of the benefits of using a genealogy database program like FH or RM is it forced me to decided on a methodology for citing various sources, where to draw the source/citation line for a given source and how to utilize the features of the genealogy program. And in my search for the "correct way", I had to remember one of the key points, in the early chapters of EE "Citation is an art, not a science.", page41.
When I looked at the number of sources from online images and the way I do my research, I decided that the advantage of having the year in the Bibliography out weighed the disadvantage of multiple Bibliography entries when it came to on line images.

Having said all that, I don't think RM or FH limits (or forces) their users to adopt any specific methodology/approach when it comes to citations. Both offer the ability to create customized templates and draw the Source/Citation line where desired. I have found nothing in the FH7 program that prevents me from continuing to use the methodology/approach I chose for citing online images. I can create the same citation, the template just looks different because FH and RM have different constraints around the template creation.
I happened to start this journey in RM and as such, became accustomed to what I now see as greater flexibility when it came to creating source/citation templates. In this one instance, I feel RMs flexibility provides better citation/source management and data entry for my records, research and work flow with certain online records.
Will see where the ticket goes - from my standpoint more flexibility is better - and let the users determine how to use it to fit their needs.

[arthurk]

Mark, arthurk clearly stated that he would be converting his 'lumped' generic Source Citations into 'splitter' templated Source Citations

No he didn't. What he said was "When I move to FH7 I am quite tempted to move to split census sources and include a tabulated transcription of each entry". It would be perfectly possible to do that with generic sources as well - you have assumed he will also convert them to templated ones.

As I'm still on v.6.2, anything I say about v.7 at this point is likely to be a bit unclear and provisional. I like the look of census sources with a tabulated transcription, but I haven't yet looked into the best way of implementing them, and when I wrote the above it didn't cross my mind to consider whether it would involve templated or generic sources.

I'm not even 100% sure that I understand the all differences between them, or what the implications might be of choosing one over the other. But please don't try to explain any of that to me now: I have a few KB items and forum threads bookmarked and I'll look into it when I'm ready. (Unless of course it's relevant to the present discussion and might help others.)

[cwhermann]

Although I have spent some time messing with data entry to get a feel for the user interface, most of my time in the past month has been focused on issues regarding the transfer from RM. So I decided to look at how I might approach source/citations going forward. Per the KB, the Advanced collection of built in templates consists of 169 highly-detailed templates based on EE.

I used the Local & State Records - Imaged Copies - Online from the Advanced collection to see how close I could come to re-creating the citations in my previous post without having to customize/edit the template. Despite the differences in field names, I was able to create a citation very similar - and noted the FH template entered the year of access in the Bibliography.

Now, if I am a lumper and I wanted to lump/create a Resource record for Nez Perce County Records found on FamilySearch and move Series, along with access date, vol page and item of interest to the citation level. FH makes it easy to do with a couple of "clicks". But as previously discussed, any metafield designated as citation specific, cannot be used in the Bibliography or Record Title. In this case, if I did not want to "lose EE compliance", I would need to add both a Series field and a YYYY Accessed field to the Source Record and adjust the Bibliography and Record Title formats accordingly.

This is what I submitted to CP as the issue is entirely contained within FH program vs an issue having to do with my transition from RM.