I used the profile of my great-grand uncle as an example in this thread about nicknames and name order. After Rhonda's follow-up, I revisited the profile LHJS-P9Y, to make sure there was nothing else I needed to do. (Thanks again, @Rhonda Budvarson)

Alexander died in Montgomery, Montgomery, Alabama, United States. Alabama death certificates are restricted to AL or FSC access, and I retrieved a copy 6 years ago.

The PQS called out a conflict in the place of death. I had entered "Montgomery, Montgomery, Alabama, United States" on Alexander's profile. The conflict detailed by the PQS:

Alex J. Burgess, "Alabama Deaths, 1908-1974" has a place of Election Precinct 16 Greenwood, Calhoun, Alabama, United States for death, which is different from Montgomery, Montgomery, Alabama, United States.

I pulled out the certificate, compared it to the profile, and then I saw the cause of the so-called conflict.

The certificate, neatly printed, shows Montgomery, Montgomery, Alabama as the place of death:

But what happened to the index, when the place name algorithm took hold?

Be sure to click the down arrow to the right of the Event Place. https://www.familysearch.org/ark:/61903/1:1:J6MS-9L6

In this instance, I have a copy of the certificate, and I was comfortable dismissing the conflict with the comment "place name algorithm error." Often, however, we can't easily have a copy of the certificate. That place name as shown in the index may be all someone has. It needs to be as correct as it can be, but that algorithm has rendered many indexes nearly useless.

The PQS certainly makes the problem more visible. Can we hope visibility can lead to eventual correction?

Thank you.