Comment by swiftcoder

> BUT I can write how to parse HTML universally

Can you though? Because even big companies rarely manage to do so - as a concrete example, neither Apple nor Mozilla apparently has sufficient resources to produce a reader mode that can reliably find the correct content elements in arbitrary HTML pages.