How is read_time calculated?


(Jay Pfaffman) #1

I am using Discourse as a platform for teaching my university courses. I am interested in using Discourse’s stats as a measure of “class participation.”

How is read-time calculated? I keep a discourse window open all day. I trust that doesn’t convince Discourse that I’m “reading” (and of course it can’t know whether I really am, but I can’t tell whether students in a f2f class are listening, either). Just what counts as “reading?” How easy is it for students to game the system? (The proper answer is “Use the Source, Luke,” but I don’t know where to start looking. Telling me which file to look in could be a good part of an answer).

I’d also welcome discourse about whether people think these are reasonable measures of “participation” and the degree to which they can be gamed. For example, it’s easy to give likes, even for things that you don’t read, but maybe the fact that likes are limited to X per day is enough to counter that.


(Robin Ward) #2

Read time is very much an estimate. We track how long individual posts are on screen, and report it back to the server on a regular basis.

There are some safeguards, such as whether the tab is in focus, whether the user has scrolled the page in some time, etc.


(Damian Yerrick) #3

Does the estimate take into account activity of users without JavaScript? If so, how?


(Jeff Atwood) #4

Discourse is a JavaScript app it does not function at all without JS. So the question is moot.