* Our current implementation of editing sessions is tightly coupled to a page view. However, this doesn't map very well to what we think of as a single edit session: on desktop, switching between the visual editor and the wikitext editor while retaining changes causes a new page view, while on MobileFrontend, aborting an edit using the back button and then reopening the editor (which doesn't preserve your changes) all happens in one page view.
* The new ability to switch back-and-forth between the visual editor and wikitext invalidates some key assumptions (for example, we probably want to update `action.init.mechanism`)
* How should we account for "micro-editing experiences" like Flow? Should they be included in this schema at all?
* Various data-quality issues:
** Quotes around some user-agent strings?
** Sessions with many init events?
** We do not actually test whether a device is a phone, tablet, or desktop—we just assume that every client using the desktop site is a desktop, and so on.
** Events with `event_platform = other`? They all seem to be visual editor events.
** Session tokens of different lengths
** We don't use the core EventLogging code for client-side session token generation and sampling.
* Even with T124676 resolved, the table is still quite large. Consider whether to drop mostly unused fields like `page.title` or normalize the schema (T123958)