You're right—it's not entirely consistent, and I can explain why that is. We evaluate special additions on a case-by-case basis, and the main criteria we use for evaluating them are these: We only add them in when a) we've heard the same feature request multiple times, b) we're positive it's a net gain as far as user usability goes, c) it fits with the ethos of the tool, d) the tool is unlikely to add the feature going forward, or if it does, adapting our API would be simple and non-destructive, and e) doesn't break workflows. One additional constraint is how widely the tool is adopted amongst Segment users. This feature (and the alternative) fails most of these criteria, and in light of the final note about adoption, makes me pretty reluctant to adopt changes we're not convinced is a long-term net benefit for all our customers.
The customer.io case is one I'm admittedly not a huge fan of personally, but a special one where the added functionality doesn't break user workflow, and helps in many use cases. This PR is different in that it breaks all existing users' workflow and adds an additional, cumbersome step to the Mixpanel-Segment setup.
We would, of course, be open to seeing how the alternate approach would work. If you'd like us to evaluate that case, feel free to put in another PR.
Thanks again!