Posts
I watch 301 people in the FamilySearch.org Family Tree. Every week I get a report from FamilySearch itemizing all of the changes to all of the people I am watching. This week's report (June 19, 2019), has 38 changes to 19 people. This is the third week of my compilation of the reports. This week's report contained many new people and a few changes to the same people from the previous weeks. Here are the running totals that include both weeks.
The average of the number of changes made without any source citation or adding any new sources went up from 85.3% to 87.7%. The number of unsupported changes is climbing from 145 to 179. If I were to extrapolate this percentage out to the entire Family Tree, the number of unsupported changes would be astronomical. You can also see that the number of people with unsupported changes is now about 10% or so of the total number of people I am watching. I have to assume that every one of those unsupported changes is wrong.
I will be away from regular internet contact for a while, but I will pick up the numbers when I get back.
I've had thoughts about doing similar analysis, so thanks! Sad (but not surprised) to see so many changes without support from sources. I am curious if you are only counting changes that were associated with a source attached at about the same time or whether any source that perhaps is already there counts if it matches the information being changed? Also, one other dimension I've wondered about is the historical time frame for people being changed. I know it's more work, but could you add a column with the lifespan of the people you're watching? I'd be interested to see if there is any correlation between the amount of changes and when they lived.
ReplyDeleteMy criteria is whether or not a source has been added at or about the time of the changes. Not an exact criterion but it gives a good approximation. Going forward, I will add the birth and death dates when available, that is a good idea.
DeleteThis is where the lack of proper source tagging support for most events really, really hits FSFT very hard. If all events could be tagged with sources then it would be vastly easier to actually show what sources purportedly support a particular occurrence. This is where the unfortunate historical system architecture biases for the six "vital" events in FSFT rears its ugly head: they are the only events to which sources can actually be tagged and solving this problem is a big one in database design terms. In other words it's not a quick fix.
ReplyDeleteYes, exactly. Presently, we have to read through all of the sources to see if there happens to be one that mentions a particular fact. I realize that around the world there are places where the only family information comes from oral tradition, but even with oral tradition it is important to know where the information came from. I have always maintained that every entry should be supported by a source (even an oral history source) and it would help immeasurably if those sources were tagged to all of the relevant entries.
Delete