Agreed, template extension was the main reason we used HTMLs from the API also in our research. In general, when investigating viewing behavior in Wikipedia, looking at the HTML is also just closest to what readers actually see.

We did not use "accept_language" so far. I dont think, we need it for building sessions necessarily since we more have browser information in the user_agent_map. It could be an interesting additional feature for the analysis though (specifically if we get to a cross-edition-browsing behavior analysis).

Bengali will be by far the smallest language edition we run the survey in; even with a full survey, we will probably end up with the least responses.

I signed the document (step 2) and requested developer access (step 3).

My wikitech username as well as my preferred login name is "flemmerich"