I think the interesting part of that article was "distributed to 813... who had used bluesky and X". That seems a little misleading. What if you just picked

I think the interesting part of that article was "distributed to 813... who had used bluesky and X". That seems a little misleading. What if you just picked 813 at random and asked how many had even used bluesky at all?