Development

Team Inspiration Core Goals Beta Test Sandbox

Our Development Canvas

Like watching an artist pass his early brush strokes on a canvas, we invite you to follow the development progress of this dynamic web database portal for historians and genealogists alike! Take a peek into our technical development in-progress between Director and Visionary J. Eshelman-Lee and brilliant developers Scott Williams and Geoff Lehr from its genesis 12/05 through 6/07.

 

What is genealogy, and what is identity?

What is identity? It seems like a far-fetched question for a genealogist to ponder, but the answer you come up with has a big impact on your research. For beings alive today, the question of identity isn't so burning---we feel a continuity in our bodies and spirits that implies a continuity of identity, even if we undergo surgery or change our minds and beliefs. But the question of what defines an identity that existed in the past is much more difficult. What we have is a variety of evidence surrounding the existence of events. Some of this evidence contains references to people---names, locations, or more vague references like "the children of so and so". But the people who lived these events, whose lives left us this evidence, are no longer around to sew the pieces of evidence into the whole cloth of identity. Like anthropologists digging through a jumble of ancient bones, we have the task of deciding what evidence can be linked together to give rise to a person. Looked at this way, `personhood' and `identity' acquire a nebulous, vague, tenuous kind of existence which is apt to give the modern inquirer a sense of the willies.

But by viewing genealogy this way, as the fabrication of identities from disparate evidence, has several strong points in its favor. People don't exist as first class citizens in this world---there's no way to `create' a new person in this genealogy. Rather people in the genealogy come about as a side-effect of the evidence collected to support their existence. People exist in greater or lesser definition: where a preponderance of evidence around one person gives light to his most minute detail of life, the uncertain evidence surrounding another may make her life a mystery, to the point where it is uncertain she existed at all, or was she perhaps just a fluke misspelling of someone else's name? Freed from the burden of `creating' people in a genealogy, the people the basic facts give rise to evolve as more evidence is acquired. Identities of vague persons may divide when it is discovered evidence was conflated to create one person when two or more actually existed. A multiplicity of identities may be reduced to one when key evidence links previous identities into a single cohesive person.

There is evidence giving shape to personal attributes of a person---his birth date, death date, places of residence, occupations, personal characteristics, photos. There is also evidence giving shape to his relationships---his birth certificate showing his parents, the people alongside in in photos, notes from acquaintances, clippings in papers, and mentions in journals. A far richer view of the person is the result, which characterizes him with all available evidence.

Of course, not all evidence is of equal quality. The trustworthiness of the source, the exactitude of the information contained in the evidence, and the concordance with other evidence means that some evidence is naturally more compelling than other evidence. By helping to score evidence as it is entered into Traces, the Traces Inference Engine will help you make smarter decisions about how evidence is related in forming an identity. As a side-effect, the Inference Engine can help you gauge how accurate a proposed identity would be, based on the uncertainty of the evidence chain.

That's not to say that the computer does all the work automatically. The Inference Engine functions like a giant computerized lab assistant---helping you find interesting evidence quickly and helping you keep track of your chain of reasoning. By suggesting connections between evidence and supporting them with documentation, you can propose identities from evidence that wasn't previously connected. Suppose you knew of a "Sue" from a journal, a "Susan" from a newspaper, and a "Suzanne" from a letter. Using your expert knowledge of the naming habits of the era, you might propose that these names all belong to the same identity, and with no evidence to the contrary, Traces will begin to solidify the existence of that one individual. Of course, if other evidence points to these names belonging to three different people, your contribution will serve to dilute the strength of their identities, provided your sources are strong enough.