Research talk:Mobile editor engagement/Editor activation
How do we know that the success rate did not just better because of a decrease in registration? For instance the graphs on this page alone would also match with a drop of registrations in both sites, or a drop on desktop only partly compensated by an increase in mobile, where only the most motivated users survive on mobile. --Nemo 22:22, 8 July 2014 (UTC)
I now have two more reasons to ask for absolute numbers, that is two doubs about the last mobile trends.
This graph shows that phone users produce a good portion of visits, but only a minuscule portion of edits; however, the portion of users who edit and read the desktop site from a phone look very similar. Does this mean that phone users would edit more if they were sent to the desktop site?
Unsurprisingly, redirecting tablets users to the mobile site, where unregistered users cannot edit, caused a drop of unique daily unregistered editors: –600 on en.wiki alone (with over –1000 edits). The graph in the next page doesn't clarify whether the increase in registered editors, caused by this restriction, was able to compensate. After paying more attention to the legend, with DarTar's help, I see the graph actually shows that registered edits and editors are stable and the total decreased dramatically. --Nemo 20:33, 1 August 2014 (UTC)
- On the first question; our definition of 'phone' is down to user agent parsing, followed by hand-coding of the common results to determine what "device" results are phones/tablets/etc. So I wouldn't take it too literally until we have a chance to map more devices to more device classes. Ironholds (talk) 20:51, 1 August 2014 (UTC)