Still I am impressed about the movement PoGo has created. (And I like the Russians conspiration theories about it is CIA running PoGO to get free spies mapping up/filming areas. It would be cute).
Well here's the interesting bit about that:
- Niantic Labs wrote Pokémon Go (aside from the through-the-camera Pokémon capturing and the actual Pokémon IP - names and artwork - the game is a pretty clear follow-on to Niantic's previous game, Ingress).
- Niantic Labs was spun off from Google a while back, where they'd developed Ingress.
- The people who became Niantic, inside of Google, came from Google's acquisition of Keyhole Inc. (per Wikipedia, "Keyhole's marquee application suite, Earth Viewer, emerged as the highly successful Google Earth application in 2005").
- Keyhole Inc. was, per Wikipedia, "a pioneering software development company specializing in geospatial data visualization applications", who got "additional capital ... from [various sources including] the CIA's venture capital arm In-Q-Tel, with the majority of In-Q-Tel' funds coming from the National Geospatial-Intelligence Agency, ..."
So, yes, there is a (historical) CIA tie-in. Though I'd guess it's more on the order of sharing common interests than spying specifically.
But mapping and filming? The Russian conspiracy theorists
completely missed that boat (it sailed two and a half years ago), if they're worried about Pokémon Go doing that: every single PokéStop and Gym in Pokémon Go is an Ingress portal, the overwhelming majority of which were selected, photographed, named and described, by Ingress players, who submitted them (complete with GPS coordinates, in the Ingress app) to improve the game (and to improve the players ability to play in their local area, and you could also get in-game badges if enough of your portal submissions got approved). (By the way, the PokéStops and Gyms are only a modest subset of Ingress portals - Niantic has many more locations they could light up.)
So Ingress already got all the locations, landmarks, descriptions, coordinates, and pictures a couple years ago. No need to worry about Pokémon Go players spying in that capacity.
Now,
if you want to put on your tinfoil hat... Ingress is all about getting within close proximity of each Ingress Portal (just the way you have to get near a Gym to interact with it - in Ingress, you take over a portal from the rival team, then link portals together, connect-the-dots fashion; make closed triangles of linked portals and they fill in as fields in your team's color; the ultimate goal is to make the entire earth your team's color, while, of course, the other team is trying to switch it all back to their color)...
What Pokémon Go does that's different is, it spawns valuable Pokémons, randomly,
all over the map, not just at portals. They've already got data in their maps for what areas are parks, water, lakes, etc. Say, however, that your algorithms had identified a park, and you (in the form of your ever-curious software) wanted to know if there was a path across that park, or if some obstruction (a fence, say) blocked the way: no problem, simply arrange a few dozen times (over the course of weeks) to spawn various desirable Pokémon on one side of the park when you know there are players are playing on the other side of the park, and watch, via GPS, what path they take: if they seemingly ignore the Pokémon in every case, then the players likely see an impassable fence or other obstruction in between and don't even try to go; if they all seem to walk partway and then give up and turn around, there's likely a fence or some other obstruction that isn't obvious from a distance. If they all walk partway and then detour 50 meters to one side before continuing, you gradually get an idea of the shape and extent of the obstruction. If some of the players go straight across, with a pause, and others go around, now perhaps you know the obstruction is something like a small stream or low fence that some more enthusiastic players elect to cross. A single experiment could give misleading data, but repeated
many times you can statistically develop a picture (a theory, really) of what is there. You can also vary the test by time of day, to find, say, gates that are locked at night (though it could simply be places players don't feel safe traveling at night, so there is much interpretation to be done on the data). Repeated hundreds of millions of times, you can refine your original basic image-based map data (shot from satellites long ago, then enhanced with things like, identifying all the roads), now adding "how difficult is it to get from point A to very nearby point B" data for, literally, everywhere.
Now, for your local park, this is interesting, but not a big deal. No military agency is likely to invade your park any time soon. But, again, they could be effectively collecting this data worldwide, for every square inch of ground. Map data (remember, their roots are in collecting and visualizing map data) that has been enhanced with detailed data about the crossability of each small bit of terrain, complete with most common paths taken, could be quite valuable. Imagine a decade from now, a squad of troops (or SWAT officers, etc) making their way through a newly-war-torn (or terrorism-torn) urban area, and the squad leader having their heads-up-display (or more low-tech, a commanding officer half a kilometer back with a laptop and a radio) suddenly tell them, "now 100 meters ahead, you need to detour 50 meters to the right - there's a tall fence you can't see from here that would slow you down and leave you exposed." That, could be very useful, and thus
very valuable, data. And an original bit of funding did come from the "
National Geospatial-Intelligence Agency". Sound like a good fit?
Like I said, tinfoil hat territory, but interesting.