> My question: What is the correct for special characters in texinfo's PO > files - to use texinfo's syntax, or to use UTF-8 (or whatever charset)?
.po files long predate Unicode and UTF-8, and I don't think gettext supported UTF-8 well for some years after it was invented. Therefore it does not surprise me that some po files use Texinfo syntax to represent some characters. Both are correct, as far as I know. I'm not aware that either is preferred. FWIW, I concur with Gavin. Also FWIW, I suggested/invented @U not for anything about .po files, but for the case when a document contains only a few non-ASCII characters. Such as almost all the manuals written in English. In that case, it is far more portable to everything to keep the source in 7-bit ASCII instead of throwing in a couple of "binary" UTF-8 bytes. (Although many manuals don't follow this, and just blindly use UTF-8. Whatever. I've given up trying to talk to people about it.) For cases when a document is actually written in another language, or otherwise needs to have lots of "special" characters, then of course it makes sense to use UTF-8 directly. Good luck, Karl