I live out in the sticks, and the whole reason I invested in a GPS/LTE collar for my dog is because there are broad swaths where he'd only be in range of an apple device if he had line-of-sight to it. Considering his favorite exploring path is the creek bed set 8 feet below grade, that means he'll just disappear.
I'd love to own the data from that collar, but I also looked into the BOM of doing this, and (considering what my time is worth doing the full-stack embedded electronics thing) it was absolutely cheaper to buy it and lose access to the data. It's a lot cheaper than fully fencing 2.5 acres out in the country.
My friend lives in a suburb of a city and her cat's airtag rarely went off enough to be useful. He mostly hangs out in the bushes where she was most worried about not finding him.
There's also the issue of keeping him on our land. There's added correction capabilities (first a signal noise, then a mild shock) as part of the Halo collar that we got him. And an entire training program that comes with it, so prior to enabling the feedback mechanisms, I train with him so its not just a bolt from the blue.
He's an 80lb Belgian Malinois, a breed known for their high drive, and for being pretty scary looking. People out here keep goats, horses, chickens, etc. He knows not to go after our chickens, but do the neighbors know he won't go after theirs? The last thing I want is for him to wander onto a neighbor's property and get shot because he looks like he's helping himself to their livestock. He'll get shot for that (or less), and I'd have zero recourse, even legally.
The Halo collar is the exact same price (including data plan) as the apple watch, but I trade remote feedback to him if he leaves the property (and a collar ruggedized to the rigors of an active dog), vs owning the data, but on a device that's not at all designed to be worn by a dog that likes to roll in dead fish and coyote shit.
Meanwhile, adequately fencing the property (so he won't jump over it) will cost me between $12k and $30k. Compare that to $700 + $5/month.
Are you asking me if, in the situation where my failure to build a $12k-$30k fence (for materials alone, not including labor, and the environmental impact survey that has to go along with 1200 linear feet of 6' high fencing in a wildlife corridor) gets my dog shot, will I wish that I spent tens of thousands of dollars when the ~$1000 or so I spent on equipment, and a few weeks' worth of after-work training should have done the trick?
No, I won't wish I had fenced the property. I'll wish I'd done a better job of training my dog.
I’m guessing the big downside to that is that an Apple Watch running a workout continuously has a battery life of maybe 6-8 hours at best, so you’d have to buy at least 2 of them and fiddle with them several times a day to swap them and start a workout. And also they cost $300 each but I haven’t looked at what any other option costs, admittedly.
Honestly if that's the only spot where there's an issue I'd get an old iPad LTE and solar panel and permanently mount it somewhere with solar. Then the airtag will be able to find your doggo there as well.
It's cheaper than fencing thousands of feet of perimeter, anyway.
That creek is miles long, forms one border of my property, and its all of below grade. I'd have to install numerous repeater stations to accomplish that, all on somebody else's property, all along that creekbed. Turns out the government and the telecos have made those investments for me.
I'd love to own the data from that collar, but I also looked into the BOM of doing this, and (considering what my time is worth doing the full-stack embedded electronics thing) it was absolutely cheaper to buy it and lose access to the data. It's a lot cheaper than fully fencing 2.5 acres out in the country.