Friday, 12 February 2016

Evolution and Genealogy

In this post, I want to pose some thoughts concerning the evolution of genealogy, as opposed to the genealogy of evolution, and consider how my recent visit to RootsTech 2016 may have tempered my prior views.

When asked how genealogy has progressed in recent decades, many people would cite the increased availability of information through the digitisation of records and other sources. But what about the process of genealogy: the methodology, attitudes, and tools?

Figure 1 - Tree of Life, Ernst Haeckel (1834–1919).[1]

A shining light in genealogy is the Genealogical Proof Standard (GPS). Although controversially using the term “proof” in its name, this is essentially a guide to ensuring thorough research, and of reaching sound conclusions. It is not a method (as sometimes thought), and not even a step-wise recipe — hence, its five parts being described as “elements” rather than “steps”. The earliest mention of the GPS by name was by Helen F. M. Leary in "Evidence Revisited — DNA, POE and GPS," OnBoard 4 (January 1998): pp.1–2, 5, and the text of that article may be found online at Although the actual wording of this standard has undergone a number of refinements before reaching its current form, the underlying principles can be traced back to at least the 1930s, and probably the late 19th Century. It is hardly surprising, therefore, that those same elements (with minor wording changes) would apply to any field of research, including those outside of the fields of genealogy and history, and including those that do not use the term “proof” in this manner. The expression of these principles was designed to help all genealogists, not just professionals, or even budding professionals, but also the many newcomers to this increasingly popular pursuit.

Another major contribution that has been embraced by the wider community is Evidence Explained: Citing History Sources from Artifacts to Cyberspace (EE) by Elizabeth Shown Mills. Now in its third edition, this mammoth book is sometimes viewed as a recipe book for source citations, and is unfairly criticised when the one you really want is unlisted. However, the book is really about evidence analysis, and the first couple of chapters guide the reader through the things they need to know — the principles, conventions, and requirements — to be empowered to do their own analysis and cite their own sources. There is a lot to learn, here, but it’s well worth the effort. Although not the first such book, it may be argued that this particular one set the scene for the more recent works designed to help genealogists make a better job of what they love doing.

What about software genealogy? You would think that this technologically-equipped field would be contributing greatly to our advances, but I have not seen this yet. I have previously been very critical of the limited, and rather skewed, picture that software paints of genealogy. With a few notable exceptions, tools are still focused on family trees, conclusions, and form-fill data entry. Part of the problem is that such tools are produced by companies governed by commercial forces, including commercial constraints, and if not commercial then at least proprietary. In other words, genealogy has no software team of its own, and is often subjected to software written by non-genealogists. This contrasts sharply with written material such as books and online publications.


This year, I was lucky enough to attend RootsTech 2016 — my first, but hopefully not my last — and found it to be an incredible experience. Meeting all those people that I only knew by name or profile-picture would have been enough on its own, but the overwhelming message of this year’s RootsTech was of particular importance to me: stories and memories. I have long bemoaned the industry’s focus on lineage at the expense of stories and other forms of narrative, and this limitation was a major force in me developing my own software. As Steve Rockwood, the new CEO of FamilySearch, said in his Wednesday Innovator Summit keynote, it’s less about “facts of the chart” and more about “stories of the heart”. His emphasis was primarily on preserving stories for future generations but this new focus still gets my vote.

A particularly moving presentation that demonstrated the power of stories was the Friday keynote by David Isay, founder of Storycorps. The premise is for two close people to record a 40-minute question-answer session between them as though it was their last chance to say something important. David presented a number of these recordings in the keynote and they were so moving that I found myself discretely reaching for my pack of tissues, only to realise that from the folks around me there was an inordinate amount of gentle coughing, nose-blowing, and dust-in-the-eye maintenance. The amazing thing about these stories is that they told themselves; they didn’t need any hype or advertising, and that’s just how stories should be. Whether we’re telling personal stories or recounting historical ones that we may have researched then that potential to tug on the heart strings will always be there. David was obviously aware of this as his introduction was a mere 3.5 minutes before the first story. Perfect!

I strongly believe that this same emotional connection can be found in descriptions of long-passed events , even if we only know of them through research rather than personal experience. One such case took me unawares while writing a narrative report entitled Like Father, Like Son. This contained a description of a house fire in 1937 Nottingham, England, that destroyed a family and their youngest children. The newspaper reports contained eye-witness accounts of a young girl at the window of the top-most of the three levels, but they had to watch as she was consumed because of the ferocity of the flames. The firemen later recounted their grim discoveries of her body, and of her mother and another daughter on the middle level. Housing, then, was largely back-to-back terraces, and many neighbours were related or worked together — much more so than in modern cities — and everyone knew their neighbours, whether good or bad. I was writing about the out-pouring of grief, and of the thousands of people who lined the streets for the funeral procession, when I reached the part describing the coffin of the youngest girl being carried by her school friends. I was so choked that I really had to take a break. Since publishing that account, I have been contacted by direct descendants of the family who thanked me for solving a mystery because no one in the family wanted to talk about those events when they were young, and eventually there was no one left who had experienced them.

I recently summarised the status of my STEMMA research project in a three-part article on my blog, and the second part presented a view of narrative genealogy that embraced story telling, narrative reports, proof arguments, and transcription (of both old and new material). I believe that this seamless inclusion is necessary for useful genealogy, and for micro-history in general. So how did this view tally with the views and products I encountered at RootsTech?

I talked to one high-profile software developer who was very keen on the concept of semantic mark-up (as discussed in my article), but his working constraints currently made it difficult to justify any associated development. Looking around the booths in the Expo Hall was informative; many of the new vendors openly admitted that they were not genealogists, and the designs for sharing memories appeared to demonstrate more than a little influence from existing social media. Selling the idea of sharing memories by alluding to products such as Instagram is smart, but this is different from designing a product using a similar paradigm. For those memories to become essential parts of our family history, and to make them shareable, searchable, and navigable on the Web, then someone needs to look at the nuts-and-bolts of how the data is be stored, and how it's going to be found by search engines such as Google.

As I write this, findmypast have announced a partnership deal with Twile, a Web site that organises your stories and memories according to a timeline, and it will be interesting to see how they view potential integration. Twile was placed third in the RootsTech Innovator Showdown, but was also the people’s choice. So what are my arguments against simply publishing videos, recordings, and images on some “social history” Web site? Well, not all stories are going to be private; sometimes you want them to be found by the families of other people who were there, whether they’re related to you or not, and irrespective of how they were referenced in your story. Not all stories will relate to one particular event that can be neatly placed on a timeline, or hung off someone’s tree. Essentially, the data model by which the stories should be organised is not trivial, and it requires the bigger picture to be considered.

When asked about their data model, those representatives that were aware of the term talked about how the resources (videos, images, etc) were organised on a Web server. None had organised the information according to real-world relationships, as opposed to operational ones, and none had any integration with other genealogical or family history information, such as lineage, places, events, sources, and so on. To me, given both my background and my personal requirements, this approach is several keys short of a piano, but to recognise the requirement itself is a huge step in the right direction. I just hope that the commercial momentum, and the desire to get there first with a killer product, isn’t at the expense of a more-considered design that gives the end-user greater scope and potential.

Most of the representatives expressed some interest in the use of mark-up, and in any guidance, best practices, or data standards that might emerge. Of particular note were Legacyscribes, Pass it down, and kindex, all of whom wanted to know more. In fact, Legacyscribes already had a basic semantic mark-up scheme in place, and I was thrilled to see it demonstrated.

On the subject of transcription, things were a little quieter. Most listened as I explained how finding information in a recorded story needed more than some simple tags applied externally to the associated digital file, but it was still viewed very much as a service rather than an essential cohesive feature. There was clearly a lack of knowledge about subtleties of transcription, and how it would apply to historical documents as well as to modern ones, and to digital recordings. Only two were entirely dismissive, both of the opinion that it just required someone with a word-processor. Although I didn’t ask at the time, I would wager that none had heard of TEI: Text Encoding Initiative.


To a lesser extent, I asked about the process of collaboration in genealogy. I have also criticised the current concept of a unified family tree at What to Share, and How, and presented ideas for crowd-sourced types of collaboration at Collaboration Without Tears, and for tree-based approaches that embrace alternative conclusions, authorial control, and automatic attribution at What to Share, and How – Part II.

Just before I left for RootsTech, a new development project emerged called Trepo, and this has since grabbed the attention of several software developers. It claims to support alternative views in a unified tree, and although it’s unclear how complex this might appear to an average end-user, it’s great that someone is thinking about this.

One of the booths where I questioned the practicality of a unified tree was the one for Geni. In 2012 Geni was acquired by and their World Family Tree contrasts with the user-owned trees of MyHeritage. The representative vehemently denied that there were any conflicts or differences of opinion that could not be resolved by a moderator. To ignore the very real “wars” that exist on such trees, and the losses incurred when someone else takes down more than they should, is rather unprofessional, but to believe that evidence analysis is a mechanical process, and that there will always be one unambiguous winner that can be decided by some third-party moderator is just na├»ve; the same evidence may support more than one view of the truth. I could see that this conversation was doomed. I wanted to point out that when recounting our first-hand personal experiences, or presenting information given directly to us by a family member, then no one else has a right to change it, and so his suggestion of “giving up control” was ill-founded. The stories and memories would be our recollections, expressed in our words, and this was a fundamental message of the conference. However, I was shouted down so I went and got a coffee instead.

In summary, it is high time that genealogy came down from the trees, and walked upright alongside history, but it will not be today.

[1] "Tree of Life", by German biologist and naturalist Ernst Heinrich Philipp August Haeckel (1834–1919). Image in the public domain. Copy obtained from Wikimedia Commons ( : accessed 11 Feb 2016).