Monday, November 12, 2018

Good News and Bad News On Ancestry Member Tree Searches

1)  There is some GOOD NEWS about Ancestry Member Trees being indexed!!!  Hooray!  It appears that sometime in October the Ancestry Member Trees were indexed after about one year of the Index not being updated.

How do I know?  I last wrote about them not having been indexed since October 2017 in Has Ancestry.com Indexed the Ancestry Member Trees Yet? (15 August 2018).  Marcia Philbrick has been writing about this also, and yesterday posted Finally Indexed! But? and today posted Issues Searching Trees.  I totally agree with Marcia's statements and conclusions, based on her own search examples.  Please read her blog posts.

2)  Using my search example of my 2nd great-grandfather Isaac Seaver (1823-1901), I did the search today for my "Randy Seaver's Family Tree - November 2017" Ancestry Member Tree.  I had attached four Ancestry.com records to Isaac in October 2017.  Here is his profile (screen captured today):

Note that there are four "Ancestry Sources."

Here are the results:

a)  I searched for Isaac Seaver born 1823 and Lived in Massachusetts - all exact.  The search for all records resulted in:



There were some results for Family Trees.  I clicked on the "See more like this" link near the top right of the screen, and saw only the "All Family Trees" result:


The title at the top says "Results 1 of 1."  However, there are 4 matches on the list, including my "Randy Seaver's Family Tree - November 2017" tree.  Great!

b)  Based on previous experience, I knew that if I clicked on the "Public Member Trees" link on the left-hand side of the screen above, I might get a different result.  Here is what I found (two screens shown):



My November 2017 tree is not on the list of 17 results.  However, my "Seaver-Leland Family" tree is listed on this search result (but wasn't on the first set of results above!).  

How can this be?  Unfortunately, this is not new.  In May 2018, I and several others noted that the search results were wonky - they didn't find all of the results and the counts were often wrong (as shown above!).  In May 2018, Ancestry.com published a statement - I blogged about it in Ancestry.com Says They Will Work on Technical Problems.  They hoped to have the problem solved by May 31 (2018?).  They haven't mentioned it since to my knowledge.

So the Isaac Seaver in my updated tree finally was indexed recently.  But why wasn't the "Seaver-Leland Family" tree not found in the first search (from the general search results with 4 trees listed) and why was my "November 2017" tree not listed in the "Public Member Trees" search?  

The number of results should match for both types of searches, and the same results - the same 17 trees (or are there more?) should appear on both lists.  My experience is that the same thing happens when the user searches for records.  The general search gives a number of results for a specific database, and a search in the specific database provides a different number of results.

Has Ancestry.com fixed all of their Search results issues?  Apparently not.  How can anyone trust anything from an Ancestry search if the results differ when the user does a search in different ways?  This is BAD NEWS.
  
3)  It is apparent that Ancestry.com searches in Ancestry Member Trees provide results for profiles in a tree containing only an "Ancestry Source."  This means that a record (e.g., a census record, a death record, another Ancestry tree, etc.) has been attached to the profile from within the Ancestry.com record attachment system.

Unfortunately, almost all of my source citations in my Ancestry Member Trees are not "Ancestry Sources" but they are sources crafted in RootsMagic genealogy software from records found on Ancestry.com and other online, home and book/periodical resources.  Those sources are listed on the Ancestry Member Tree as "Other sources."  They don't "count" as Sources for the purpose of indexing a Member Tree profile.  

My current "Randy Seaver's Family Tree - November 2017" tree has over 51,000 persons and over 98,000 source citations attached to it.  But only a handful of the profiles are indexed by Ancestry.com because I have attached only a few records that count as "Ancestry Sources."

I don't have the time left in my life to attach at least one record to every person in my Ancestry Member Tree just so it could be indexed and found by another researcher.  If I did one attachment a minute eight hours a day it would take me 204 days to add all of my sources to my tree.  If I did one attachment a minute for only one hour a day, it would take me 1,633 days (almost 4.5 years) to complete.  Frankly, that is wasted research, education, society and writing time.  And it would be very boring.  I already have source citations for my tree profiles, Ancestry chooses to ignore them.  Essentially, they hide my tree profiles from the rest of their customers.

My opinion is that Ancestry.com should index every person in every tree whether they have "Ancestry Sources" or not,  so that their customers can find information about their ancestors that other researchers have found.  Yes, I understand it's a big computer challenge - but they're a big company with many servers, and charge customers money to search all of their databases.  This is more BAD NEWS!

4)  "Approving" an Ancestry WebHint from within RootsMagic does not create an "Ancestry Source" for the person profile.  Here's an example:

My RootsMagic screen for the profile of Frank Walton Seaver (1852-1922) shows that I have "Accepted" ten WebHints (green check mark at the top left of the screen below):



However, the profile for this person in my Ancestry Member Tree shows only one "Ancestry Source:"



As you can see, there is only one "Ancestry Source" but there are 38 "Other Sources" on the profile.  

By the way - Frank Walton Seaver, even though he has an "Ancestry Source," wasn't found in a general search or a specific Public Member Tree search.

5)  My bottom line conclusion is still:  Ancestry Searches do not consistently provide ALL of the results that should be found for a given set of search criteria.  A search using different techniques (from a general search, from a tree person search, from a specific database search, etc.) should provide consistent results.  They don't in my humble opinion.

6)  I would write more about the "quality" of "Ancestry Sources" but that's a rant for another time.

7)  Does anyone else care about these issues?  I hope so.  What can customers do about it?  Ancestry.com should care about it too as the putative biggest fish in the genealogy ocean.  


                                 ==============================================

Disclosure:  I have always had a fully paid Ancestry.com subscription since 2000.  Ancestry.com has provided material considerations for travel expenses to meetings, and has hosted events and meals that I have attended in Salt Lake City, in past years.

Copyright (c) 2018, Randall J. Seaver

Please comment on this post on the website by clicking the URL above and then the "Comments" link at the bottom of each post.  Share it on Twitter, Facebook, Google+ or Pinterest using the icons below.  Or contact me by email at randy.seaver@gmail.com.

2 comments:

Rick Crume said...

I asked Ancestry what percentage of profiles in its Member Trees don't show up in search results because they are not linked to sources on Ancestry. Ancestry replied that nearly all of the profiles are linked to sources on Ancestry, so almost all of them show up in results. My contact said that even most profiles in free Member Trees are linked to free record collections, so they show up in results, too. I'm skeptical of these claims. Most of the profiles in my own Member Trees are not linked to Ancestry records and don't show up in search results.

However, a few profiles in Ancestry Member Trees show up in searches, even though they don’t have sources from Ancestry. For example, a profile of Johann Georg Schaubhut, born in 1734 in the Mueller, Ratz Tree shows up in searches, even though it is labeled “Unsourced.” I wonder why that profile is indexed, even though it’s not linked to sources on Ancestry.

David said...

This is a kind of reverse tu quoque fallacy. The correct response to the "most" profiles claim is: so what? "Most" in this circumstance is simply not good enough.