The FamilySearch functionality is more stable on the Windows version of Rootsmagic and that's what I am referring to here. One of the interesting Rootsmagic settings is the one to automatically add FamilySearch webhints for people. I can take an old ancestry tree with, say, 10000 people and the FamilySearch -> Automatch function might match say 2000 people. But with this other setting set as I navigate my tree, or step through the list of people, it checks FamilySearch with what is a more forgiving/less conservative criteria and slowly increases the number of matches.
At first I tried to click the down arrow repeatedly to step through and automate this process, but it became tedious.
Eventually I downloaded a windows app that records macros to press certain keystrokes, that have a hot key attached to them. So I can open my partially matched Rootsmagic tree, select the first persion in the person list, then, when I press the <cntrl><Windows>j keystroke, the app simulates me rpessing downarrow, waiting two seconds, then pressing again, in a loop with 10,000 iterations. Six hours later this navigation is complete and I have another 2500 matches (say). It's convenient, but would be great to have this functionality be . a part of the product.
Step 2 - if I am lucky, matching someone with FamilySearch.org means that I want to add the familysearch record for that person, their spouse, heir children, their parents, etc. Typically this means a lot of clicking, I'd love to be able to accept a match with a single click an dget the person, spouse, and kids.
Alternate workflow. It is possible, though it wasn't immediately obvious, to add a person and their ancestors from familysearch.orb by specifying their familysearch ID. This isn't always stable or consistent. In some cases boldly saying "50 generations of ancestors" will get me in a loop where someone has connected my family member to an FDP (FamousDeadPerson) and instead of adding 200 or 600 people I will see 30,000 people or more added over a day or longer, that goes well before 1000AD into the woods of the familysearch.org dataset. This is fascinating from the perspective of statistics, large datasets, pedigree collapse, data quality, graph traversal but I suspect it's also fantasy.
If computing power becomes ultimately free, how will we be doing this in 20 years?