Interesting. I think one of the key things in this article is it is targeted less at individual productivity metrics than using as an indicator for the overall heath of the developer experience. Size of diffs, number of diffs probably does also indicate that they are making smaller more incremental changes. One of the challenges sometimes with these types of measurements is having orgs that don't have the scale or resources to throw at a problem like this for the level of experimentation they are getting back.