Yes exactly -- we weight common feedback from multiple Discourse instances much more heavily than individual opinions from folks on meta.discourse, for example.

Yes exactly -- we weight common feedback from multiple Discourse instances much more heavily than individual opinions from folks on meta.discourse, for example. (There is also the issue of observing what people do versus what they say they do. At least early on I find the problems are severe and obvious enough that this subtlety isn't a big deal yet.)