* Help Understanding Sources in FH

[USMC7312]

General Patton one said, “we are two peoples separated by a common language.” So I have been reading the wiki and this board trying to understand how best to implement and convert my existing sources and citations to best be rendered into reports and books.

So I will ask a few questions in hopes of making sure I comprehend the best means to implement sources on FH.


1. I think I understand the concepts regarding the two methods of entering sources, ie Splitting versus lumping.
A. My basic take is that you split when your source has media that you can attach, contains lots of information that can be linked to more than one events and attributed. On the other hand you lump things are indexes that don’t contain media.
B. So things like certificates (BMDD - Birth, Marriage, Death and Divorce) get split by person and things like Census data gets split by family. Books, newspapers, magazines and other traditional media from paper get split.
C. Things like indexes of birth, marriage, death and divorce get lumped. Find a grave would also get lumped.

2. I am struggling most with how to insert data in the right fields of the source and citation property boxes so that the source comes out looking like a proper bibliography or footnote.

A source has these fields
Title
Type
Short Title
Author
ID
Publication Data
Repository
Text from Source

A citation has these basic fields
Entry Date
Assessment
Where within source

Now if I want to enter my data so that FH creates a proper bibliography and footnote (assuming all footnotes are listed as first and subsequent footnoting is ignored), how should I add my source and citation info?

3. In looking at my source and citations that were imported I see some interesting things. For example, when looking at the raw data I see this:

[See 1st attachment]


If you look at this source and see the templates section, how was that added? More importantly how does the source get composed.

4. When I look at the source below it has a template that has fields for footnote, short footnote and bibliography? How was that created?

[See 2nd Attachment]

[USMC7312]

Then again I may be completely over reading the entire concept.

[tatewise]

Firstly, all the entries with an underscore (_) prefix such as _TMPLT are custom UDF (see below).
They were exported in the GEDCOM from your original product, which presumably has templates, etc.
Those entries have been faithfully imported by FH but totally disregarded as it does not recognise them.
See how_to:handling_unrecognised_data_fields|> Handling Uncategorised Data Fields (UDF).

That original product has used Citations to hold the details such as Where within Source and linked Media.
i.e. It has the 'lumped' style, but contrary to the guidelines you have correctly identified in 1.
Those Citations may need converting if you want to adhere to those guidelines. (Discuss later.)

There are examples of entering Source details in how_to:index#recording_facts_and_sources|> Recording Facts and Sources.
The are example Sources in the Family Historian Sample Project.

Different users desire different styles of Source reporting in their Reports.
You say you want a proper bibliography or footnote, but that means different things to different people.
FH uses essentially the same principles in every Report that holds Sources.
So use the Family Historian Sample Project existing Sources or create some Sources using the fields.
Then produce a Report and experiment with the Report > Options > Sources tab settings.

[USMC7312]

Ok, so I watched the video and add sources and citations seems pretty easy using the automatic feature, BUT .. I still don’t have a good answer to where should Put my source and citation data. Moreover, when using default reports does FH just publish a bibliography or is it a citation. The two are different concepts for sure.

1. Given these fields for input, what data in my bibliography should go into what fields to ensure a proper bibliography.

A source has these fields
Title
Type
Short Title
Author
ID
Publication Data
Repository
Text from Source

** I have seen some folks say that the best practice is to put the citation completely formed as desired in the TITLE field. Then creat a short but descriptive short title. Then ignore the rest. This would be the effective example of “splitting?” This would also allow for a Evidence Explained Citation minus markup.
•• On the other hand if you wanted to lump all BMDD records by type.place.title, then that requires to correctly form the common parts of the citation in the source fields and then rely on the citation fields to implement the one off portion or unique portion of the citation in the citation property box which contains even fewer fields.

A citation has these basic fields
Entry Date
Assessment
Where within source
Text from Source

None of which seems to make any sense at all from any style type description of source and citation perspective .

It almost seems easier if you just Simplified things down to this:
1. In the source property box provide these fields:
Full Bibliography:
Title:
ID
Repository
Text from Source

2. In the citation property box provide these fields:
Full Citation:
Short Citation
Text from Source

I understand that the goal though is to have source and citations that can be exported via GED and be properly imported.

Honestly, this type of source and citation strategy is common amongst European family tree software, but I just don’t get it? This data is critical to any good family tree that you plan to share publicly with anyone. Some would suggest if you don’t have it sourced it didn’t happen.

So this leaves me struggling with my decision to give FH a try. Honestly, I love the data mining, UI, the reporting and workflow of FH, but if I can’t document my data using some common style: Evidence Explained, CMOS, APA, etc... I am screwed.

To the other extreme I have RM which does sourcing like no other. Any style any time any format works. But then the downside there is workflow, UI, and reporting.

So how in simple terms can I get my bibliography and citations added to FH so they are effectively and accurately reflected in reports and books?

[Jane]

V7 will gain source templates, bibliographies and source templates similar to those you are used to, but with the added flexibility which FH includes as standard. Adding "sentence" like construction to producing citations and other entries.

[tatewise]

Despite FH V7 adding those features mentioned by Jane, underneath I suspect it will still use just standard GEDCOM fields.
That is crucial if you want portability to other genealogy products via GEDCOM.
As you have have seen, the template (_TMPLT) features of your earlier product use non-standard GEDCOM tags that are not recognised by any other product including FH.

The Source and Citation fields presented by FH are almost exclusively the standard GEDCOM fields.
It leaves their usage up to you, and no prizes for using every field. Just use the ones that work for you.

The snag with using the Citation fields is the same citation text will get repeated against every Fact derived from the same source. That is why 'splitting' is attractive, as all the details exist only once in the Source record.

Does FH just publish a bibliography or is it a citation?
Please do check out the Report > Options > Sources tab where you will see you can choose whichever of the Source and Citation fields you prefer. So the answer to your question is it publishes almost whatever you want.

As you've recognised, to follow Evidence Explained use Title &/or Publication Information for a full Citation/Bibliography.
Author, Repository, Text From Source and Note are self explanatory.
Type is non-standard GEDCOM, but useful for internal grouping of sources.
Short Title is useful for internal searches.
Custom Id is also useful for searches and although standard is not widely supported in other products.
There are several other standard GEDCOM fields available via the All tab.
When using 'splitter' mode the only Citation field widely used is Assessment to establish the credibility of that citation in supporting the fact.

[AdrianBruce]

V7 will gain source templates, bibliographies and source templates similar to those you are used to ...

For clarity, I was about to add the statement that FH does not produce a separate bibliography. However, that tantalising glimpse of v7 from Jane means I should rephrase to "FH v6 & earlier does not produce a separate bibliography".

Previously, if I had ever wanted to produce a bibliography, I had it in mind that I would run a query on source records to get as close to a bibliography as I could (no details because I never got that close to doing it) and copy and paste it into a Word document along with the narrative text. So v7 looks very interesting!

Mike's caveat about "non-standard GEDCOM tags [for templated citation / bibliography etc stuff] that are not recognised by any other product" is important. When I was working with the BetterGEDCOM group, one of their aspirations was to put all the required extra tags into their new, expanded GEDCOM successor. These were usually tags as required by Elizabeth Shown Mills citation & bibliography formats. However, new source types come along all the time, requiring new "citation tags" (for want of a better term) - those would, naturally, require tags not yet known to other people.

Apologies to everyone who regards all this as self-evident but I feel it important to point out that even Utopia can have bits hidden in the long grass.

[tatewise]

It might help if we could establish a clear definition with examples of bibliography and citations in this context.
How do they relate to 'lumper' source documents?
How do they relate to 'splitter' source documents?
My understanding is each 'splitter' Source record identifies one citation (but maybe I am wrong), and that is why the FH/GEDCOM Citations hold little or no information.

[Jane]

There is a difference between recording and outputting information. So the citations for output, may combine many citations in the data, so for example if a source is cited 10 times in a report FH will combine these into a single source citation on a report, if you have it set to combine identical ones.

A Bibliography lists all the "Major" sources so you would expect to see one reference to each Census year with it's repository.

This will become much clearer once the new version is released and as I said the format for all the output formats are under the control of the user so they can pretty well be what the user wants.

[ColeValleyGirl]

From Evidence Explained:

bibliography: a list of sources relevant to the subject at hand, citing each source in full. An annotated bibliography is one that discusses the sources in addition to providing full citations. A bibliography typically does not cite individual manuscripts or documents; rather it cites a collection or series in which the manuscript appears. See also source list

citation: the statement in which one identifies the source of an assertion. Common forms of citations are source list entries (bibliographic entries), reference notes (endnotes or footnotes), and document labels.

reference note: a citation (q.v.) or comment placed at the bottom of a page or at the end of a piece of writing and keyed to a particular statement in the text; its purpose is to identify and/or discuss the source of the specific statement made in the text.

source list: a bibliography or list of sources used for an essay or in a research project.

So, a bibliography is a list of documents or sets of documents, NOT linked to specific assertions (in FH6 think Source Summary Report, but customised depending on how you specify your Source records). And a citation is a reference note that links an assertion to the supporting source, with as much detail as necessary to locate the evidence for the assertion in the source (in FH6 think the Source section at the bottom of the page for an Individual in an Individual report).

[tatewise]

It would help to put that definition of bibliography in the context of FH Citations and Source records.
e.g. Is each 'splitter' Source record for a Birth Certificate listed in the bibliography or just the entire collections of the GRO England & Wales Birth Certificates, the GRO Scottish Birth Certificates, & etc?
I suspect the latter, and if so, how could FH V6 produce such a bibliography list in a Report and only include those relevant to that Report? Which is what the OP is asking I think.
Would the answer be different with FH V7 when it comes?
Do Repositories have a part to play in this?

Postscript: The 'splitter' Sources listed in a Source Summary Report are the same as those listed in the Sources section of other Reports (assuming all Citations of one Source are the same) so I do not understand how that is a bibliography of a collection of sources. Maybe I am just not seeing something obvious.
Perhaps a Record Details - Repositories report could be made to work?

[ColeValleyGirl]

FH6 isn't good at bibliographies (but it does do reference notes well).

As a splitter, if I wanted to generate a bibliography (which is something I never do), in FH6 I'd probably look at putting the collection/document information in the Publication Information field and using just that field in a Source Summary report. (The page notes or end notes in an Individual report aren't any use because they tie a source citation to a particular assertion or assertions and that is not what a bibliography does). Might need to write a custom query or plugin (to populate a named list of sources as the basis for generating the report) to identify sources referenced by a subset of individuals rather than every source in a project, and also to avoid duplication of multiple sources in the same report... Caveat: I haven't tried sorting out the query yet and am not much motivated to try -- I'd rather wait for FH7 and see what has changed.

To be fair, for a bibliography, lumpers have the same issue, except they don't have to record collection/document information separately but they still have to generate a Source report based on a subset of individuals.

[tatewise]

In the Source Summary Report just selecting the Publication Information field does not produce a subset of Sources but repeats the same Publication Information repeatedly in all Source records from the same collection.
But as you say, a Plugin could determine such a subset, and in fact could produce a bibliography list derived from whatever field of Source information is preferred.
A Query cannot determine such a subset, nor produce a bibliography without help from an external tool to avoid duplication.

[ColeValleyGirl]

I think a plugin is probably the right way to go for FH6 users (selecting a set of sources cited for a set of individuals and using a user-specified field in the Source records to avoid duplication). I shall add it to my to-do list (I have a rather large plugin project on the go at the moment, for which I'm doing a lot of hand-sketches and what if's and oh rats that won't works, but intend to come back to this area for FH6 in due course -- I may also do some work on the Add Source from Template plugin for FH6 users (to be honest, I've done the work, I just need to tidy it up a bit and also get permission from the original author to publish it). I don't think providing some limited functionality in this space for FH6 will impact on the market for FH7 -- there's too many goodies rumoured to be in it -- but it might help people who for whatever reason decide they can't upgrade.

[davidf]

It might help if we could establish a clear definition with examples of bibliography and citations in this context.
...

Do we need to step back and try to find some expression of "need" that is language independent (particularly FH or Shown Mills language independent)? Then we can try and find how FH can implement such needs.

I come from a background of tutoring students (mainly at UK master's level) who were very reluctant to "do the work necessary" to record their citations and sources and often had to go back over their work and do a (very) painful refit to meet submission standards and to avoid accusations of plagiarism.

So(!),

1. Summary Needs

Are there two types of "summary" needed?

1. Those authorities, records or resources specifically quoted or paraphrased
2. Those authorities, records or resources consulted which may have influenced the work and consequently may be of interest to others working in the field.

These summaries are often high level;

• Cabinet Papers at National Archives ref CAB/xx
• Ferguson, N; The Pity Of War, 1999, Penguin: London
• The Churchill Papers at Churchill College.
• BBC: The Secret War (DVD of 1977 series), Simply Media: London
• Imperial War Museum; Photographic Exhibition: Women at War, 19xx
• Conversation with Military History Expert x at the National Archives

If they are not at high level they cease to serve the purposes of being "summaries". That Purpose being?

That purpose or need can take many forms:

• To allow the reader to see if the author is aware of a particular resource (examiners like to see their works in a student's references!)
• To check if a resource has been used if it should subsequently prove to be unreliable (or replaced by a more authoritative resource)
• To allow the reader (or author) to get back to a particular resource

I have seen the first of the above two types (authorities, records or resources specifically quoted or paraphrased) usually referred to as "(a table of) references" and the later often referred to as "bibliography". The latter is pretty much stand-alone - a sort of "reading list" for the interested. The former however has to be linked to from the parts of the work that quote or paraphrase that resource. If links are provided back into the work to show where the resource has been used it may be referred to as an "index of references").

2. Specific Justification Needs

We need to be able to justify statements we make in our work. This can generally be done three main ways:

1. By experiment
2. By logical argument
3. By authority

Or a combination of the three.

In genealogy the first two are often combined:
Argument Proposition: "If the wife of John Smith is Jane Jones, we would expect to find a Jane Jones in census records prior to the marriage with a birth place and deduced birth year approximately the same as that of the Jane Smith (wife of John Smith) found in census records after the marriage"
Experiment: "We found two such records, but we also found a death record for a Jane Jones prior to the date of marriage and after the last census prior to the marriage"
Conclusion: "Give the place of death of one of the Jane Jones, we can probably conclude, that the other Jane Jones is the woman who married John Smith."
Where we might hold this information is probably not part of this topic (some form of note against the particular facts?)

In genealogy the "By authority" option is the one that principally concerns us. The "authorities" are resources and we cite (or link to) them as supporting our statements (if necessary with qualification).

How we do that is as much about conceptual database design as anything else.

1. We normally want the "work" (be it a thesis or genealogical report) to be readable without visual interruption, so normally the authorities are not completely specified within the main text of the work but are referred to within the text of the work by some form of shorthand. This might be by a footnote or an abbreviation for an authority.
2. We don't want to clutter the work with repeated detail, so the repeated bits are held in a separate "master" list and the shorthand referred to above has to be linked to this list.
3. We sometimes need to hold information about the link or resource that is distinct from the flow of the main work. This might for instance be a short discussion as to the reliability of the resource in supporting the "fact" in question.

The link to the master list is often called the citation - how the resource is being cited - and often takes the form resource; place in resource, for instance Ferguson, N (1990); p234. The citation in academic papers is either embedded in the flow of the text or is in a footnote or endnote. The benefit of the latter is that it allows the author to volunteer subsidiary information which would otherwise interrupt the flow of the main text.

In "relational database terms" (I know FH is not relational but the linking concept is common!) the tables take the form:

Code: Select all

Main Work Table
- contains foreign keys (footnotes) to the citation table
Citation Table
- Key: Citation ID
- Foreign keys to the resources table
- If required: Some format of pointer (eg. page number) to a position within the resource
- Other information pertinent to the citation
Resources Table
- Key: Resource ID
- Optional: Foreign keys to some form of "location of resource" table
- Resource Details
- Other information pertinent to the resource
Location of Resource Table
- Location ID
- etc

...
How do they relate to 'lumper' source documents?
How do they relate to 'splitter' source documents?
...

For a lumper (full disclosure: unapologetic lumper), the above fits quite nicely because the lumper's FH sources are the resources referred to at the top of the post as the entries that would typically appear in a "table of references". Because they are high level you then need to use "Some format of pointer to a position within the resource" - the FH where within source field in the citation.

For a splitter the citation is effectively redundant and the database schema above reduces to:

Code: Select all

Main Work Table
- contains foreign keys (footnotes) to the resources table
Resources Table
- Key: Resource ID
- Optional: Foreign keys to some form of "location of resource" table
- Resource Details
- Other information pertinent to the resource
Location of Resource Table
- Location ID
- etc

But the Resources are no longer high level so the summaries discussed at the beginning are not easy to produce, because you need to strip out the implicit citation details.
So a Low Level "splitter"'s resources such as:
RD Jones @ Huddersfield, District 19, Page 23, 1881 England and Wales Census
AP Jones @ Huddersfield, District 19, Page 23, 1881 England and Wales Census
RS Jones @ Huddersfield, District 19, Page 23, 1881 England and Wales Census
has to be cut back to:
1881 England and Wales Census
This might be achieved by means of short titles etc, but
then the inevitable duplicates have to be removed.
Mike has suggested (following good database practice) that another database table should be used to hold these high level resource details and that the summary reports could be produced from them. Unfortunately the only "table" available within FH that could do this is the Repository "table" - and borrowing a table for another use is not good practice!

Having another table would allow splitters to hold information that might be specific to a particular high level resource. Such information might be:
High Level Resource: UK Passport Applications (TNA)

• Repository: Foreign Key for The National Archives
• Other Information: TNA Reference FO 611/20-25, These are large collections that need to be eyeballed, so do early in a visit before you are tired.

High Level Resource: UK Passport Applications (FMP)

• Repository: Foreign Key for Find My Past
• Other Information: TNA Reference FO FO611/1-19, This index appears not to be complete

For Lumpers this is held on the actual source record's note.

If there is something specific about say the Passport Application for "Stephen Jones, 1946" (say it was later cancelled for some "interesting reason"), the splitter would hold that on the "low level" source note; the Lumper would hold it on the citation note.

...
My understanding is each 'splitter' Source record identifies one citation (but maybe I am wrong), and that is why the FH/GEDCOM Citations hold little or no information.

I think this would be right in respect of the "extreme" splitter, but consider a census record. My understanding is that for most splitters the "source" is the page on which the household is found (what happens if it spread over two pages?). But that source would be cited for each member of the household. Even if the extreme splitter defines the "source" as the line in the census return it will be cited for the Census fact and potentially for an occupation fact as well as place of birth, date of birth etc. I suppose you could go mega-extreme and make the source the row/column intersection (the actual data point) - but then you could end up repeating information common to the household (such as address).

As I think we all end up saying it depends; in this case "what do you see as the source?"

• The 1881 England and Wales Census (the "collection")
• The 1881 England and Wales Census for Huddersfield
• The 1881 England and Wales Census for the Jones Family in Huddersfield (Household)
• The 1881 England and Wales Census for RD Jones in Huddersfield
• The 1881 England and Wales Census for RD Jones in Huddersfield (Occupation Data point)

And that opinion is possibly moderated by whether you wish to hold information specific to any point within the above hierarchy.

[tatewise]

Thank you David ~ a bit long for me, but I think I got the points.

You say: "My understanding is that for most splitters the "source" is the page on which the household is found (what happens if it spread over two pages?)."
My understanding is that for most Census splitters the "source" is the Household Schedule that has its own number.
If that is spread over multiple pages, then they can all be added on the Media tab.
However, that is a diversion from the main thrust of your points.

I have realised there is an alternative database table that could hold high level resource details (bibliography).
Instead of linked Repository records it would be possible to use linked Note records.
They are easily added via the Source record Notes tab Add Note > Add Link to Note Record.
To differentiate these high level resource bibliography Notes from other Notes there are several options:

1. Start each such Note text with Bibliography:
2. Give each such Note a unique Custom Id
3. Add each such Note to a unique Named List

The Publish > Record Details - Notes report can use 1 or 3 above as selection criteria, but unfortunately its format is unattractive and repetitive. Perhaps a Query or Plugin could do a better job.

A 4th option is to link a Source record to each such Note record.
Those Source records would have their Type set to Bibliography.
A Source Summary Report filtered by Type = Bibliography could be customised to yield a bibliography list.
That would be a Project wide Bibliography and would need some more filtering (using a Plugin and Named List) to produce a Bibliography for a subset of Source Citations.
OR we wait and see what FH V7 has to offer.

[AdrianBruce]

For a splitter the citation is effectively redundant ...

Based on your 'schema' yes, but remember that there are other items in that yellow source pane, such as the assessment of whether the information in the source that is used to justify the assertion is primary or secondary (drawing a veil over the other values in that item, which are in a different 'dimension' to primary / secondary). Also the text from source which can be used to highlight specific text relating to the assertion (not a duplicate for me since I don't normally copy the text).

I would suggest that your definition of bibliography deviates from the usual genealogical uses, particularly when you refer to "a sort of "reading list" for the interested". But it's probably perfectly fine for real life books and theses!!!

The issue of what goes in the bibliography is as much an issue for lumpers, I suggest, as splitters. The ESM format for the US Federal Census bibliography entry (actually, she calls it a Source List entry) is, in essence, text something like "Alabama. Pike County. 1860 U.S. census, population schedule. Local copy. Office of the Probate Judge, Troy, Alabama." for a locally held manuscript original. It is traditional in the US (presumably because it's in the ESM format) for census source list / bibliography entries to describe a census at a county level. Whether that makes much sense in the era of Ancestry etc images, I don't know - depends on how it can be used to evolve the full reference note (i.e. the foot-note / end-note or whatever).

Now, every once in a while I end up talking about Master Sources and Sources. It's a long time since I looked at this so I could be talking @@@@@ but software such as FTM has not just Source Records but Master Source Records and Source Records. The definition of MSRs and SRs is basically non-existent so even ESM's explanation of Master Sources makes little sense. ( ) So far as I can both see and remember, Master Source Records are just a computer concept and they are created to hold common details that can be inheritted by Source Records.

(NB RootsMagic has a similar concept except they have Source Records and Detailed Source Records respectively. RM's help file gives the game away when attempting to answer the question: "How much detail do you put in the top level and how much in the lower level?" with the answer: "Up to you..." )

The point about this is that, again, so far as I can both see and remember, the titles or descriptions in the Master Source Records are what appear in the Source List / Bibliography. And the titles or descriptions in the (lower level) Source Records are what appear in the reference / foot / end notes. But bear in mind that the titles or descriptions in the (lower level) Source Records (dynamically) inherit part of their contents from the (upper level) Master Source Record title. E.g. the detailed reference note for the census schedule includes the census type, county name from the MSR and the detailed location and schedule data from the (lower level) source record.

How Simon intends to do all this, I have no idea, to avoid the duplication that will come for splitters like myself if we list multiple census schedules, I dunno. Perhaps highlight something in our source records...

[USMC7312]

This is a great conversation and I seriously appreciate those that participated.

I must admit though that I find myself shocked that FH (and many other genealogy apps) fail to be able to capture the necessary data to present a bibliography (Source list), and citations (First and Subsequent). In fact, the only company that I have seen to do this effectively is Roots Magic and it does the job amazing well using Evidence Explained Style, but allows you to use any style you want by providing template functionality. This is what I am hoping for with FH7? FH is a great product and excels where RM fails especially in terms of reports, workflow and data mining. I love these features and just hope that FH7 takes the leap of providing a means to effectively manage source and citation data using Evidence Explained and other styles with templates. Inclusion of multi-layered source and citations would be great too.

I am just dumbfounded by this entire problem because when you share your genealogy with anyone even close family members, they always want to know the answer to the question, "Where did you get that information?"

To me the entire idea of having a Windows based application starts with this question. In fact, I would earnestly state that any genealogist that doesn't provide a bibliography (Source List) and citations (First and Subsequent) is not a genealogist at all.

So the take away from this thread and correct me if I am wrong? The only means of including source and citation data is a hack. Essentially, put the citation in the Title field of the source when splitting and when lumping place the bibliography in the Title Field of the source and the citation data in the Text from source. Then live with the massive duplication caused by splitting?

[USMC7312]

I might add that part of the problem in FH is this entire concept of "lumper" and "splitter." This drives the strategies used to assign evidence of a fact. Yes, "splitting" and "lumping" exist in Evidence Explained, ie. U.S. Census Records. In the U.S., we typically split our census data by county or at least that is the common expectation amongst genealogists that call themselves professional as I understand from conversations. That said, "lumping" still should allow media. In fact, I would suggest that media should be part of the citation not the source. But that is just my opinion...

[ColeValleyGirl]

Splitting doesn't cause massive duplication.

All the data common to all uses of the source is stored in the source record (or the associated media record); all the data specific to a single citation (the association between a fact and the source) is stored in the citation, including (if you want) media associated with the citation/fact (rather than the source).

Yes, splitters end up with more source records than lumpers, but FH is very good at dealing with large numbers of records and gives you the tools to search and filter as you need to; splitters end up typically with more information in their sources than in their citations and tend to associate media with sources and/or include source transcriptions in the source record.

Lumpers will typically end up with more data in their citations (including partial or complete source transcriptions) and may associate media with the citations. However, you can vary the approach depending on what makes sense for each source and fact. (For example, I use a single source for a collection of Phone Books online, rather than a source per entry; ditto civil registration BMD indices -- it's just common sense).

Re sharing data, of course we source it! Look at this or any other page for an individual in my tree and tell me it isn't sourced! It uses reference notes -- I've already said FH6 isn't good at bibliographies, -- and I choose not to use the ibid style of referencing but you can do that if you choose. My citations are produced using Templates (the Add Source from Template plugin).

You'll have to convince me why I would need multi-layered sources and (I assume you mean) multi-layered citations?

The 'hack' as you call it is deciding how to make use of the source fields available to you in the Gedcom specification to capture what you want to record. I'm hoping FH7 will make this easier with templates, but everything except a true Bibliography is possible right now.

[jbtapscott]

Perhaps I'm missing something here, or looking at things too simply, but I build my website from my FH project and anybody can see quite clearly on there what the Source (or Sources) for a Fact are and get the answer to "Where did you get that information?". I certainly don't "hack" my data to get these results!

[davidf]

...
So the take away from this thread and correct me if I am wrong? The only means of including source and citation data is a hack. Essentially, put the citation in the Title field of the source when splitting and when lumping place the bibliography in the Title Field of the source and the citation data in the Text from source. Then live with the massive duplication caused by splitting?

We may be on the same page:

For Lumpers the Title Field of the Source holds the unique common name of "the source" - which as a lumper is usually some form of "collection" - A parish register, A "box" at the National Archives with a distinct reference, An England and Wales Census on Ancestry, A "collection" on Family Search (although I think that gets murky with the merger of the old IGI transcriptions and the Patron Records - which in some places were not much more than wishful thinking). This gives the top level answer to your question "Where did you get that information?" and is what I would expect to see in a list of sources / references.

Then the specific position in the source "Where you got that information" is put in the citation "Where in Source" field (in the yellow right hand side of the Individual/family property box).

The full "citation" is then the combination of "Source Title" & "Where in Source" - which gives a complete answer to your challenge "Where did you get that information?". I don't see that as a "hack" - but making proper use of one of the means by which it is intended to use the program.

(Mike's ingenious suggestion for using Shared Notes to allow Splitters to have a "higher level source" in order to produce a "Source List", is also making use of a feature of the program - Shared Notes - which, carefully used, can get you out of a lot of problems. A hack? I don't think it "breaks" any rules of either FH or GEDCOM - although you may need a bit of explanation to understand what is going on.)

[davidf]

I might add that part of the problem in FH is this entire concept of "lumper" and "splitter." This drives the strategies used to assign evidence of a fact. Yes, "splitting" and "lumping" exist in Evidence Explained, ie. U.S. Census Records. In the U.S., we typically split our census data by county or at least that is the common expectation amongst genealogists that call themselves professional as I understand from conversations. That said, "lumping" still should allow media. In fact, I would suggest that media should be part of the citation not the source. But that is just my opinion...

I would agree that "fact" media (being specific to a source and a fact) should be part of the citation (a realisation that regretfully I only recently came to) and that is possible in FH if you use the media icon on the Right Hand side of the Property Box - this then asks you if you are adding citation media or source media. Splitters would use the latter; Lumpers should use the former.

In this respect I think the media icon below the fact panel might be removed (or at least disable-able in preferences). If you are adding media to a fact, you are adding "a source" so you should be using the yellow right hand side! Adding media directly to a fact that may have more than one source is inherently ambiguous as to the source of the media!

[ColeValleyGirl]

David, the are three places you can add media associated with a fact, a citation and its source.

1. To the source -- typically for splitters this would be one or more images of the source.
2. To the citation -- which might be used for lumpers to include an image of a particualr 'page' or 'schedule' or whatever within the source.
3. To the fact -- could be used for e.g. an image of the church where an event took place -- I'm not sure I'd use this for anything source or citation related.

So you need to be able to access media both under the facts list and for the source and/or citation in the yellow pane.

[USMC7312]

ColeValleyGirl, thanks for the correction as it is multi-layered citations. As for proving the necessity, I think the author of Evidence Explained has done that thoroughly already. You either agree or not with her. As for data size in citations, I don't mind that as long as it is pertinent.