I can't find the post, but I remember reading that they'd rather not expose full raw data because of privacy concerns — there could be a way to mine this data to de-anonymize users.
I'd love to play with the raw data too, but I understand their concern. So far many "anonymous" large data dumps ended up exposing too much (e.g. recently NY taxi data).
Oh I know this might be very sensitive privacy-wise. People should just be aware if they're contributing to a proprietary database (like Google Maps) or an open one (like OSM) when using Mozilla Stumbler.