Adding a few thoughts:

* Nemo makes a good point that there have always been tons and tons of mirrors. In my opinion, none of them have ever been anecdotally impactful enough to warrant specific study even if many are frustrating. The degree to which the impact of these mirrors is diffuse across all of them as opposed to one major mirror also complicates our ability to study their effect on readership etc. at Wikipedia. And search engines do not like giving away raw click data so that's largely a dead-end when it comes to understanding these sites impact unfortunately.

* External re-use in general (not just mirrors but any place Wikimedia project information shows up outside of the ecosystem) is something that we are thinking about at WMF. You can see that it's been added as a metric in the medium term plan (https://meta.wikimedia.org/wiki/Wikimedia_Foundation_Medium-term_plan_2019/Worldwide_readership). I see this as a nod to understanding that much of the way that Wikipedia articles etc. are seen/heard by readers is through search engines, voice assistants, etc. The impact of these external entities is obviously a long-running debate and not one that has ready answers but certainly quite important to the future of the projects.

* Personally, I am planning to start some research in the next year around better understanding all the different places in which Wikimedia content is re-used and what impact that has on the projects themselves -- for example, the ability to attract new editors, how well the external content links back to Wikipedia. Initially, much of this will be enumerating the different ways/places in which content appears so we know what questions we should be asking. But hopefully we begin to answer questions like how to track this re-use, the value of Wikimedia projects to services like voice assistants, and what sorts of external re-use is most beneficial to the projects.

Best,

Isaac