I'm not adamantly opposed to it, but it seems rote to me. As I've tried to explain, if there isn't a clear 'yes' then the default answer should be 'no'.

I wonder if there's a different path to getting the valuable part of it? It seems to me that instead of rushing to embalm experiments as regular features, we should just do more experiments.