I’m thinking along the lines of older spouse dies, younger spouse marries someone younger and becomes the older spouse. Then older spouse dies again and repeat. Has anything like this happened in a long enough chain to be significant? Is it so mundane no one cares?
The data is mostly already there and publicly maintained. Ancestry/familysearch/etc should get us something interesting at least, data is a little bit light outside the us but someone would just need to go through it.
This sounds like a job for a weird spaghetti code contraption consisting of selenium, perl, DBI, postgres, and shitloads of caffeine. I’ll give it a look tomorrow, hoping that the captcha I assume is there can be circumvented or automated one way or another.
I’m not really interested in the people, as I’m not from the US and am unlikely to be related to any of them. I’m just curious about the dataset itself, especially in relation to OPs question.
And yes, I wrote perl and not something newer. Suck my camel hump, it’s how I roll.
With ancestry, yeah, that’s going to suck and it’s the bigger database, but with familysearch, you’ve got an API:
https://www.familysearch.org/developers/docs/api/resources
Not sure what your limits are.