If it's an OpenType font, the encoding and naming of ligature glyphs are largely irrelevant. It's all in the OpenType layout code. Basically if the ligature forms automatically, it *will* spell-check correctly in the original app. There's no way to defeat that as far as I know. I've never seen a shipping OpenType font that had "bad ligatures."

There are various complications and exceptions (notably with underlying text representation in Acrobat and PDFs made in ways other than exporting from an Adobe app), but that's the gist of it.

This all ties in to the "character/glyph model" which I will write about at more length.

Cheers,

T