Good question. It could work, but a few of issues come to mind off the bat: scalability, quality, and durability. Especially for exterior content, it’s important to have coverage (e.g. a whole city). Users expect their area to be covered, if you say you have “Boston.” So, if you can pretty much blanket the whole city, and make sure gps is good, intersections are good, and it *works*, then great job to you.

Second issue is quality. Exposure controls, seamless stitching, synching, etc. need to be solved. They are absolutely solvable.

Finally, if you can have a bunch of webcams to work on top of a car while driving around at ~40mph, then cool.

In the end, I suggest first thing to think about is the requirements, then think about how to solve (assuming there’s a market for it, that is). “just because” isn’t a good enough answer for using webcams. (Don’t think you’re asking that, but for clarification.)