It's Sunday, and one of those days I set aside for checking my biweekly research progress. Only today, we will see that all the progress I've made in the past two weeks will mostly be invisible. This is for a variety of reasons, chief of which is my work on speculative family trees which, due to my doubts about what I'm finding, I've chosen to hide behind a shield of privacy. There is, however, another reason my biweekly count will seem off.
While I don't often mention this aspect of my genealogical progress, not only do I count how many individuals I've added to my four family trees, but I also keep track of how many matches have been added to the count at each of four different DNA companies for my test and for my husband's test. In the past two tallies, however, something has been off. For our DNA tests at 23andMe, while the count generally inches up by two to six matches for each two week period, suddenly our numbers jumped up to 1,500. And stayed there. Why?
For instance, on September 6, I counted 1,451 DNA matches at 23andMe, an increase of five over the previous period. Strangely, on September 20, my DNA match at that company was exactly the same: 1,451. And yet, on October 4, after weeks of single-digit-only advances, I suddenly had a DNA match count of 1,500. Exactly.
My husband's case was even more curious. His, on September 6, had dropped by one, giving him a DNA match list comprised of 1,379 matches. He did gain one match back by the next biweekly report—and then suddenly, in the space of the next two week period, rocketed up to 1,500 matches. Exactly—and more, incidentally, than any increase he had ever had in any biweekly change since first testing at 23andMe.
Curious turn of events, I'd say—until I noticed a post by genetic genealogy blogger Roberta Estes who, in her October 8 post at DNAeXplained, mentioned an abrupt change at 23andMe: to a "subscription model" (currently by invitation only). A concurrent change is the capping of DNA matches at 1,500.
For Roberta, that meant a loss of over two hundred of her DNA matches at 23andMe. For me, well, invisible progress. Somehow, I now have 1,500 matches, too, but the forty nine I apparently "gained" since October 4 are likely ghosts to round out my numbers to that 1,500 cut-off point. Somehow, I didn't quite measure up, while others lost some real numbers.
Roberta and I are not the only ones witnessing those funny numbers. Blogger Margaret O'Brien of Data Mining DNA noted many customers complaining about their losses, as well, in what she called a "surprise purge."
As aggravating as the data loss may be, that is not the only cause of my invisible progress these past two weeks. Work on those private, unsearchable trees I have hidden at Ancestry.com, while useful as I muddle through this Falvey family DNA puzzle, has kept me from making any progress on my four public Ancestry family trees.
For instance, the best I did in the past two weeks was the work on my mother's tree. Now at 23,548 individuals, that tree only gained forty five new entries in the past two weeks. That, however, was much more reassuring than the puny three names I added to my mother-in-law's tree—which still hovers at 19,257. And it certainly was more encouraging than the two flat-lined trees for my dad and father-in-law, stuck at 731 and 1,813, respectively.
That secret, hidden Falvey tree may have gained quite a few new entries, but since I haven't been tracking my progress there, I can't say how much they've grown. Besides, I hardly can vouch for some of the entries. While each branch grows in step-by-step research-based accuracy, the premise which attaches each branch to the main Falvey tree is tenuously held there by nothing more than reasoned arguments. I may—or then again, may not—be correct in my theories.
We'll take a bit longer to explore the last of the DNA possibilities for connections to the Falvey line down in New Zealand, but the time will soon arrive to set aside that exploration for a later time when more documentation may become available to those of us researchers who ply our tasks half a world away from the proof we require.