Review for Mining Mobile App Stores

[nenadmedvidovic]

This was a fun chapter to read, but it suffers from three minor-to-moderate problems that, if addressed, would improve the chapter in my view.

1. MODERATE: The chapter is descriptive and clear in the treatment of the problem but does not contain actionable advice. What can I, either as a developer or user of mobile apps, do (better) as a result of the described mining exercise and subsequent analysis?
2. MODERATE: The chapter does not have a thesis statement up front nor a conclusion at the end. I found the last paragraph, which listed other things one could do with app repositories, distracting and out of place. I was expecting a punchline there. Along with the lack of actionable advice (issue Added skeleton of bias chapter #1), this made it look like more support for mining “because we can”.
3. MINOR: The following sentence struck me as a tautology: “For example, a 1 to 2 star ratio of 5 for a complaint type indicates that this complaint type has 5 times as many 1 star ratings as 2 star ratings.” I think it can easily be removed without loss of understandability or content. I think it would also be interesting to see, if you have the data, to what extent developers respond to the really negative comments in subsequent releases, and whether this, in turn, improves the given app’s reviews.

[timm]

this review has been copied into #53