The Accuracy (or Inaccuracy) of Online Family Trees

We commonly accept the proposition that online family trees are inaccurate and mostly passed around copies of each other. But can this be accurately demonstrated? If the trees were merely copies, wouldn't the information in each copied tree be about the same? What's the use of copying a family tree if you don't copy all of the information correctly? So why are there so many different iterations of ancestors with lots of descendants?

MyHeritage.com gives me a perfect way to see all the duplicate family trees and the variations. I will go to my standard ancestor, my Great-grandfather Henry Martin Tanner. MyHeritage.com, with its Smart Matching software, shows me 46 family trees that have Henry as a common ancestor. Conveniently, they also have compilation of the differences from each of the other trees along side my information. So how do I know my information is correct? Let's just say that the basic facts are well beyond dispute and I presently have a long list of sources supporting each fact. Se here are the results of the comparison:

My file:

You will have to click on the image to see any detail.

OK, so here is the compiled information from 46 other family trees that share this ancestor on MyHeritage.com:

Once again, click to see any detail.

First of all and notably, the variations in his birth information go from close match to far-fetched. As a matter of note, Henry never lived and was not born in South Cottonwood, Salt Lake, Utah. He was born in 1852 but the actual date is 11 June 1852 and the place, as shown on my entries, was San Bernardino, Los Angeles, California, United States. Actually, California was state (1850), but San Bernardino was not yet a county. That might explain why some of the entries say San Bernardino County, but doesn't at all explain why some miss the state entirely. As I said above, if they were merely copying, why are there so many wrong entries?

This comparison makes me start thinking that I need to revise my opinion of online family trees. They aren't inaccurate because they are copies, they are inaccurate because they are unaware of the facts and don't even know it. Why would I say (or believe) such a thing? Well, Henry Martin Tanner is probably one of the most completely documented people you can find outside of a president of the United States or a few other famous people. It takes absolutely no effort to find correct (or essentially correct) information about his and his entire family. As a matter of note, the John Tanner book (his grandfather) and the Sidney Tanner book (his father) and the Henry Tanner book,  all have the correct information. The John Tanner book simply omits the county in the birth information for Henry Martin Tanner.

Here are the books, by the way:

• Tanner, Maurice, and George C. Tanner. Descendants of John Tanner Born August 15, 1778 at Hopkintown, R.I., Died April 15, 1850 at South Cottonwood, Salt Lake County, Utah. [S.l.]: Tanner Family Association, 1942. 
• De Brouwer, Elizabeth. Sidney Tanner, His Ancestors and Descendants: Pioneer Freighter of the West, 1809-1895. Salt Lake City, Utah (4545 S. 2760 E., Salt Lake City 84117): S. Tanner Family Organization, 1982.
• Tanner, George S. Henry Martin Tanner; Joseph City, Arizona Pioneer, Born June 11, 1852, San Bernardino, California, Died March 21, 1935, Gilbert, Arizona. 1964.

Now, you can see that Henry's Grandfather, John Tanner died in South Cottonwood, but how did it happen that all those people decided that Henry was born there? His death information is a little easier to explain. Henry had a cousin named Martin Henry Tanner who lived in Grantsville, Tooele, Utah. I guess it must be really easy to mistake, Henry Martin Tanner of Joseph City, Navajo, Arizona with Martin Henry Tanner of Grantsville, Tooele, Utah.

Now here is real issue. MyHeritage.com has made big deal out of finding source information for all its users. You don't even have to do anything and the program tells you the correct information from original sources. Doesn't this go beyond simple copying? Doesn't it get a little bit weird that these people don't correct their files? After all they are getting the same Smart Matches I am getting and they can see the same confusion. This goes way beyond mere copying.