Water drainage is still going to be a big problem, as is damage/scratching caused by shoes, bicycle kickstands, trolleys, anything else with wheels that is pushed/pulled, you name it. Plus the simple fact that most sidewalks (at least here in the city) are often in the shadows of buildings.
It only makes sense to start thinking about sidewalks once you've covered all the roofs, which are way better in every sense.
It's not that simple. Looking at the damaged roadway, I see lots of punctate damage. That puzzled me until I read your comment. I'm pretty sure that it's gravel under truck tires. So the same force-concentration dynamic as high heels.
Not only that, a common problem round here (UK) is that trucks park partially on the pavement. They routinely crack two-inch/50mm thick concrete paving slabs. I'd like to see a solar panel that can withstand that.
Surprised this is getting down voted, this is a common example in physics 101 describing the concept of point pressure. Architects have to consider the point pressure of heels when designing certain walkways.
Common physics example, but it disregards how women actually walk and stand in heels. They walk on their forefoot, rather than balancing their weight on the heel.
Shows less than 30 N/cm^2 (40 PSI) for women in heels - and higher heels result in lower heel pressure. Forefoot pressure is higher, despite the larger surface area.
Sidewalks are still flat on the ground, which is not great in much of the world (like Normandy) where sunlight comes in at an angle. You would still get some generation, but would it be worth it compared to cells mounted elsewhere at a better angle, with less need for hardening, and less accumulation of dirt? Doing this on the sidewalks would add a lot of costs and complications.