txtify -- pull the main text out of any web page.
       Deterministic heuristic: strip chrome, keep the prose,
       hand it back as plaintext. No JS, so JS-only/paywalled
       pages (e.g. NYT) will come back thin -- it'll tell you.
       

SEARCH 

       
       Or go straight to the raw text from a shell:
         curl gopher://gopher.someodd.zip/0/applets/curl/txtify.lhs/example.com/some-article
       (the URL is the selector; bare host/path defaults to https)