Dear Log,
Thought for the night, which occurred to me as I was stuffing my face with gorda tortillas and green chile omelette: It is my impression that it is easier to implement a push parser (like HTML::Parser) than a pull parser (like HTML::TokeParser); and it seems easier to implement with a pull parser on top of a push parser, than to implement a push parser on top of a pull parser. So push parsers win. Thank you, drive thru.
