Oppy's current predicament got me thinking about this again. It seems to me someone somewhere (in JPL) must be deciding what the site location is going to be called. I can't believe the naming is a random process and therefore it seems logical to tie it to distance.
Let's separate those two keywords because they actually behave differently.
IMO, a change in site id is decided *before* the move itself; it's a sort of planning decision like "drive from A to B in driving mode X" and then "reset to a new site id".
Changes in drive id are (imo) determined automatically by the rover (call those changes "steps") but I don't see those steps related to distance.
QUOTE
I'm now wondering if the site drive location is more related to commanded driving distances rather than actual drive distances achieved. This would make some sense, as they probably have to decide on the site number for file naming before the drive takes place, images are taken and then sent to Earth.
The "steps" are almost certainly related to commanded drive and not actual drive, but not necessarly to distance. Instead of using the current situation as an example, we could go back to Purgatory I. On that time, what happened (have a look to the daily change in drive ids *and* the MER status reports) was that sometimes a step was equivalent to 0.2m, some other ones to 2.0m, and iirc they also had 12m steps executed.
And here we are not talking of autonomous driving modes; in those cases who knows (ok, rover drivers will for sure) how to translate "steps" in meters?