Modern Software Experience

2011-06-27

the state of citation

sources

Many early genealogies are geneathologies; just a bunch of claims without sources to back these up. The Frenchman André Duchesne (1584 - 1640) is sometimes known as the father of French history, for basing history on critical collections of sources. He took the same scholarly approach to genealogy, basing genealogies on historic charters and other documents.
The German historian Johann Christoph Gatterer (1727 - 1799) wrote the first handbook for genealogy. By the 19th century of the Gregorian Calendar, the idea that genealogies should be based on sources was well-accepted among serious practitioners of genealogy. Yet, the idea that you should cite your sources still needs advocating today.

The citation-source-repository model popularised through GEDCOM is the first widely used digital citation standard for genealogy.

digital genealogical citations

Genealogical magazines, started in the 19th and and 20th century, set their own genealogy publication and citation formats. Those citation formats were, naturally, largely based on those used in other disciplines.
Cite Your Sources by Richard S. Lackey was published in 1985. Its subtitle A Manual for Documenting Family Histories and Genealogical Records makes it clear that is an entire book dedicated to how to cite sources for genealogy.

GEDCOM

It was around that time that some of the earliest genealogy applications were published. PAF 1.0, a BASIC program, was released in 1984. PAF 2.0, released in 1986, was a complete rewrite in the C programming language, and the first application to support GEDCOM 2.0, an early version of GEDCOM.

FamilySearch's citation-source-repository model popularised through GEDCOM is the first widely used, digital citation standard for genealogy. It is fairly simple and reasonably flexible, and certainly allows citing most sources in a clear and consistent way, but the associated FamilySearch terminology is so confused that it is confusing.

citation templates

It seems that Roots was the first genealogy application to support citation templates. Howard Nurse had these templates created by Elisabeth Shown Mills, who went on to write Evidence! Citation & Analysis for the Family Historian, often abbreviated to Evidence!.
That book was influential in getting other vendors to support citation templates, including The Master Genealogist and RootsMagic; the SourceWizard in RootsMagic 1.0 is based on Evidence!.

Evidence Explained: Citing History Sources from Artifacts to Cyberspace, often abbreviated to Evidence Explained, published in 2007, is an expansion of the earlier work.
Following the publication of Evidence Explained, several major vendors of genealogy software started to offer Evidence Explained-style templates. Legacy 7 was first, followed by RootsMagic 4 and Family Tree Maker 2009 Service Pack 1.

I'm only half-joking when I say that citation templates make citing sources easy, but that you now need a Wizard to help you find the right template.

template issues

Many genealogists agree that Mills' citation templates result in better citations than FamilySearch's GEDCOM model, but that does not mean that switching to these templates is an improvement in every way.

First of all, the current thinking on citation templates is not definitive. Evidence! was followed by Evidence Explained, and it was followed by second edition.
A common criticism of even the latest edition is that it is awfully Americentric.
It certainly seems fair to say that citation templates are a moving target.

Secondly, the number of templates keeps increasing. I'm only half-joking when I say that citation templates make citing sources easy, but that you now need a Wizard to help you find the right template. Finding the right template is not made easier by the fact that it is not guaranteed to exist.
The growing number of templates presents a considerable implementation burden to software vendors wanting to compete in this arena. RootsMagic had the right idea when they decided to define their templates within a template system, but the fact that every user can create their own templates introduces problems of its own.

compatibility

A fundamental issue with the various citation template implementations is that Mills wrote her books for genealogists; for genealogy software users, not genealogy software developers. Her texts provide guidance, not unambiguous feature specifications. She consulted on the creation of templates on most if not all of the aforementioned products, but that does not imply she fully endorses the final product. It does not imply that the different template implementations are compatible with each other. In fact, genealogists who've used them have become aware that the existing implementations of citations templates are not compatible with each other.

comparison

In 2009, Mark Tucker posted a YouTube video that demoed how citing online sources should work; different sites should use all use the same standard citation format, one which would enable desktop applications to automatically receive all the pertinent data from the online record collection. This was followed by a series of blog posts about online citations, with the first two posts discussing Mills' QuickCheck models and the GEDCOM output of EE-style supporting genealogy applications respectively. The output of the three applications is different, and incompatible.

Today, choosing to use any vendor's citation template system is choosing vendor lock-in.

conversion

I left Mark Tucker a comment noting we’d need some kind of GEDCOM standard for EE-style citations to have them transfer well from one application to another, and that there is none yet. I also suggested that he look into how well each application was able to read their own citations back in, and how well older pre-template applications imported these GEDCOMs.
Early this year, Randy Seaver did a series looking into how well templatised citations transfer from one application to another. His 2011 Feb 7 post looked into how well each application reads back its own GEDCOM file. Both Legacy and Family Tree Maker failed this basic test. Only RootsMagic succeeded in reading its own GEDCOM back in.
That does not mean that RootsMagic is perfect, it does mean that Legacy and Family Tree Maker need to be improved, but the most important observation remains that the distinct GEDCOM dialects that the different applications use are not compatible with each other. Once you to choose to use these templates, your citations will not transfer to another application. Today, choosing to use any vendor's citation template system is choosing vendor lock-in.
No surprise then, that at the end of his blog series, Randy Seaver concluded that he would not be using the templates, but free-form source citations; the free-from citations not only provide direct control over the formatting, and survive conversion from one application to another.

The harsh reality is that there is no digital standard for citations templates.

state of citation

FamilySearch has done nothing to standardise support for EE-style citations in GEDCOM. In fact, they have not done anything about GEDCOM for more than decade.
The various limitations of GEDCOM are such that many GEDCOM alternatives have been suggested. Today, both OpenGen and BetterGEDCOM aim to produce an alternative, and there is much discussion on the BetterGEDCOM Wiki about supporting EE-style citations, and on the larger subject of evidence and conclusion.
Neither OpenGen nor BetterGEDCOM is close to releasing a specification. Moreover, none of the vendors of any of the aforementioned products is a regular participant in either project.

This then, is the state of citation in genealogy today. There is so much interest in the topic that entire books are dedicated to this aspect. Users clamour for the ability to cite by the book, and vendors spend significant resources implementing citation templates. Genealogists eager to cite correctly, vendors aiming to please them, it sounds wonderful, but it isn't good enough; something is still missing.

The harsh reality is that there is no digital standard for citations templates. There is a citation style, but no digital citation standard. Each vendor is doing their own thing, and the various implementations do not interoperate with each other.
Because of that, one user who publicly evaluated his options to make an informed decision came to the conclusion that his best choice is to ignore the shiny new template systems in favour of old-fashioned free-form citations.

updates

2011-09-07 Yates templates

I forgot to mention that John H. Yates released free open source EE-style templates back in February of 2010. The Yates templates have been out there for 1½ years now, yet, ass far as I know, not one genealogy application developer has taken advantage of it.

links

Mark Tucker: ThinkGenealogy

Randy Seaver: Genea-Musings

sources, citations, standards