SFAA Perspective 01

The intention of this project is to make use of the Historian Archives. I’m taking what I think is a Business Intelligence approach. I’m trying to produce results that are useful to people from the mountain of information. The ideal would be to publish a dozen new volumes of genealogy, but that’s just not practical.

As the first step, we connect the Seven-Volume Index and the Langbehn Database. That is a great outcome and within current capabilities. It will take some time, but it’s feasible.

As a second step, we connect all other submitted GEDCOM files. This is a huge gain. Before now, I just have not known what to do with submitted information. This is because I’ve been thinking in terms of creating a coherent book or some electronic equivalent.

Now, we’re thinking more in terms of a search engine. Browse around and make connections. I’m not sure where to take it from there, but that’s a great start.

This also means that we can begin actively accepting electronic submissions as GEDCOM files. Once the import process is created, it should work for any GEDCOM file. That’s because I’m taking the intermediate step of importing into the TNG software. The actual GEDCOM import becomes the TNG author’s headache. The result is in MySQL tables, and I can then mine those tables to produce the linking information needed for our User Experience data design.

Being able to accept electronic submissions is a huge gain. This does not address other submission formats, but it’s a start. We have volunteers; perhaps the option is to take non-electronic submissions and get them into Family Tree Maker (or other electronic) form.

Now that I’ve abandoned the intention to create books, this becomes simplified.

What about non-electronic submissions?

We’ve allowed for this in our User Experience data design. Pretty much any other artifact or collection gets Dwight classification, key name or names, descendant line, and repository information.

So, step one is to enter the information in our catalog, and via the name or Dwight classification, people now know the collection exists. It may be sitting in my basement in a box, but we know it exists. It’s the same situation as for those twenty-four shelf feet of file folders.

This information submission becomes a candidate for some volunteer to turn into electronic form, i.e., GEDCOM file with attached images or whatevers.

Submission Form

It would certainly pay to create and publish an Information Submission Form. Thanks to our User Experience data design, we know what information we need:

  • Descendant line or lines, or what non-descendant line
  • Submitter’s full information
  • Line of ascent/descent to Elder John Strong or whatever the known connection is. This is so that we can, on our own, figure out the Dwight classification. It might actually make good sense to model the Line of Descent to the submitter. We have this on our application form. It’s a major way we look at things.

The submission form could also be web based, i.e., give them a PDF page they can print and fill out. 

Collection of Information is a HUGE point within the SFAA.

Key Considerations

Here are some key considerations.

  • What is feasible, and in what time frame?
  • What is the opportunity cost? Are there other things which should be done instead?
  • How can other peoples’ help be incorporated? Can help be incorporated in realistic and practical fashion?

Those would be good things to discuss when we next get together. I suspect the answer is that I need to stay on top of Historian correspondence, but now have a framework for handling things like information submission and so on.

I can specifically ask for help with research inquiries, and converting submitted information to GEDCOM form.

There are some “later” activities such as mining the Manuscripts in Progress and the Walther Barnard (5,000-page) Manuscript for creating name indexes. Since we’re only mining for names and not attempting to assemble relationships between the named people, that becomes a far more feasible project.

And, by the way, mined lists of names should be allowed-for in the User Experience Design. Each manuscript being mined has a single Dwight classification, a descendant line, and so on. The mined names get added to the master names list.