Well, it took several decades, but standardized html parsing is now actually fairly robust. So you may not be wrong, but anything that breaks is probably heavi

Well, it took several decades, but standardized html parsing is now actually fairly robust. So you may not be wrong, but anything that breaks is probably heavily outdated and probably not worth much to anybody (because it's parsing html different from browsers today, and it's not like optional tags are a new thing...). I wouldn't get too worried about crawlers and parsers.