-- Removing the check-in feature from the Foursquare app and putting into a separate app called Swarm.
-- Making Foursquare 100% about exploring and recommendations, a la Yelp.
This seems like a bizarre move. Unless I'm mistaken, all of their data comes FROM check-ins. People write tips, answer questions ("Is this place good for kids?") and generally contribute their info to the web of data that Foursquare uses to intelligently recommend venues. Couple that with the network effect of having access to friends' check-ins, which to me is probably the strongest signal of a particular venue's quality.
So all that being said, it seems like receiving a high volume of check-ins is critical to Foursquare's recommendation success. They don't have the broad review base of Yelp or an editorial team, so check-ins are all they have. By splitting this functionality into a separate app (outside of the Foursquare app itself), they are explicitly going to lower check-in volume, because SOME percentage of people simply won't download "Swarm". And then if they lose some of their user base, it decreases the utility for all Foursquare users as (a) the recommendation data won't be as solid, and (b) users will lose valuable friend check-in data.
Maybe I'm overplaying the slippery-slope angle, but when your business largely relies on a particular user action (check-ins, in this case), it seems unwise to split that functionality outside of your core product in a move that will unquestionably make taking that action more difficult.