In some recent research, I began to notice a peculiar occurrence. At first I thought it was some type of anomaly, but now I'm beginning to wonder.
Rather than incorporating data as recorded by the enumerator in an original census document, the ACOM attachment/citation building process is now substituting index-generated alternate data in its place. This is a HORRIBLE development...and I don't know what to make of it.
Here's an example...http://search.ancestry.com/cgi-bin/sse.dll?db=1910USCenIndex...
The enumerator recorded "J. C. Daggs" as the name entry.
Try attaching that individual into one of your trees as a new entry and see what you get. I'm all ears here.
Updated to better link
Update (3/10 11 AM) - The problem is apparently being caused by the introduction of a new option in the new source attachment mechanism. You can now select, where provided, either the document content as recorded or alternate content (however submitted and designated) as the data to be imported into your tree.
This new ACOM sourcing/citation option is, IMHO, potentially disastrous for the integrity of ACOM data-based citations from this point forward.
At the least, where alternate data is cited, any resulting citation should CLEARLY be re-labeled as to its source.