I would love to be able to use surfraw as a generic "information
extractor". For instance, with elvis like leodict, rfc, rhyme,
thesaurus, foldoc, etym, currency, etc., I don't really want/need it
to open a browser. Instead, all I really want is to see the results
right there in the terminal.

The way I envision this is that elvis can set $SURFRAW_CROP_head and
$SURFRAW_CROP_tail to regular expressions, and surfraw then strips
the matches accordingly.

Alternatively, $SURFRAW_CROP_match may be used to just select a part
of the HTML returned.

Surfraw would then wrap that HTML with simple boilerplate and
display it with the browser of choice.