You persist in raising the strawman of using chomp on non-record oriented input. Yes, if you've got non-line oriented files you shouldn't be trying to process it line-by-line and using chomp. And if you look you'll find that probably most of the regulars telling you you're off your chum are the first to, for instance, point someone at HTML::TokeParser when someone's attempting to treat HTML as line oriented input.

However if you persist in insisting that a sawzall is all you ever use and that everyone else is wrong in not using them as well for everything they might be doing as well (vascular surgery, engine maintenence, animal husbandry) you can't expect to be taken seriously.

Update: Changed first "non-line oriented" to the more generic "non-record oriented".