The Details of Sharing Data Across Family Tree Programs -- Part One

Many members of the The Church of Jesus Christ of Latter-day Saints now find themselves with more than one copy of their family tree. The family trees can be in a desktop-based program or online in one or more of the large online family tree programs. This proliferation of family trees has become a challenge to some and a burden to others. This series will examine both the issues involved in maintaining separate family trees on different programs and the methods, presently available, for moving data from one to another of those programs.

Why would I want more than one copy of my family tree? Aren't I just inviting trouble? How would I keep more than one family tree up-to-date? How do I move information from one tree to another? This series of posts hopes to answer all those questions and many more.

The main reason for writing this particular post series is because members of the Church have access to putting family trees on several large database programs, including presently, Ancestry.com, MyHeritage.com, Findmypast.com and possibly more in the future.

I am going to start this post with the assumption that the end product of gathering our family history will ultimately end up in the FamilySearch.org Family Tree. I suggest that this should be be goal for several very important reasons. The first of these reasons include the fact that the Family Tree program is a unified tree and that the sponsorship by The Church of Jesus Christ of Latter-day Saints adds a substantial measure of assurance that the information captured by the Family Tree will be preserved for a very long time. In fact, the present, transitional nature of the Family Tree demonstrates that FamilySearch and the Church never throw away anything even remotely of value. You might think you are deleting information from the Family Tree, but ultimately all of the changes to the program are preserved, even if not immediately recoverable. This does not mean that information cannot be lost, but the there is a rather high level of assurance that everything of value will be preserved.

For this and other reasons outside the scope of this post, I am beginning with the Family Tree. Obviously, you can add information to the Family Tree simply by keyboarding it into the program. Beginning with this simple model of the process, I will start to analyze other methods of adding information, some difficult to understand and others quite easy. Here is the first (of many) diagrams to illustrate the process:

There are some advantages to manually entering data into the Family Tree. These include the following:

• Possibility of a greater accuracy in the entries
• Greater control over the format and content of the data
• Works to add data from almost any source

There are also some definite disadvantages:

• The process of manually adding data is very slow
• There is a definite problem with typographic errors
• Entering large amounts of data will likely not happen
• Data entry is slowed by the physical limitations of those entering the data

You can probably supply the rest of the reasons. The main objection to manual data entry is that the information is probably being copied form an existing source. If the source is an individual's memory, there is probably no way to speed up the entry process. The person has to remember and enter the information. At this point there is a possible source of assistance from voice recognition software, so this adds another path to the diagram.

Voice recognition could be used to enter the data directly into the Family Tree, but the software for doing this will depend heavily on manual correction and movement on the screens. If voice recognition produces a digital document, then the resulting document is no different than any other digitized document. Here is an example of me reading a list of the names of my ancestors into a very good voice recognition program, Dragon Dictate.

• Henry Martin Tanner
• Sidney Tanner
• Thomas Parkinson
• Marianne Bryant
• over a Christian over some
• Charles Godfrey to freeze
• Margaret Jarvis
• Margaret Godfrey Jarvis

You can see that some of these voice recognition examples come out in a less than desirable format.

For the purpose of this particular series of posts, I am considering the data to have been entered initially into some written format. I mentioned voice recognition capabilities because they exist, not because I believe they are ultimately useful for entering data into family tree programs.

Manual entry of the data into the Family Tree program is mandatory unless the information has been converted from a physical format into a digital format. There is really no difference, from a data entry standpoint, and having the information on a piece of paper or in my own head. Assuming, that the information had already been typed, I could use optical character recognition software to transcribe document. Here is the diagram showing the intermediate step of transcribing the document into a digital format. Of course, the process of transforming a typed document could be done by scanning the document and then using optical character recognition software to extract the data from the scanned file, but this particular process is only slightly more accurate than my example of voice recognition software above. Many of the large online databases are experimenting with the process of optical character recognition for indexing their records. This particular technology is likely to increase in accuracy to the point where it will be a substitute for human indexing.

There is the modified diagram:

The dashed lines and question marks indicate possible, if questionable, methods of data entry into the Family Tree program. 

Even though I am presently beginning with data entry, ultimately in this series, I will be addressing the problem of moving data from one family tree program to another. Obviously, the methods that I have discussed so far could be applicable to information presently in a family tree as well as "raw data" from personal knowledge or from typed documents.

In the next installment of this series, I will address the issues of moving information from a desktop based program to the Family Tree.