5 ms·
For internal use, OK, fine. But what part of the following is not "just text"? "Don’t be naïve. £100 is a lot more than €100—and Hélène knows you know that."
by gjm11 11y ago
For internal use, OK, fine. But what part of the following is not "just text"? "Don’t be naïve. £100 is a lot more than €100—and Hélène knows you know that."
- TazeTSchnitzel 11y agoWhile you can't represent that as plain ASCII, I think Windows-1252, an 8-bit single-byte encoding, can handle that quite well.
- pekk 11y agoI guess we can get some arbitrary text, then check whether all its characters can be encoded as Windows-1252, store it with a note that it is in Windows-1252, and then check that note before decoding it. Clearly so much better than UTF-8.
- TazeTSchnitzel 11y agoThey tried: https://en.wikipedia.org/wiki/ISO/IEC_2022 https://en.wikipedia.org/wiki/ISO/IEC_2022 It was... horrible.
- masklinn 11y agoSure I mean who wants to visit Госуда́рственный Эрмита́ж or eat 삼겹살? Moonspeak's not really text right?
- jquast 11y agoThere is an art in discovering which encoding the text vs. represented by application, by making sense of nonsense. https://jeffquast.com/soviet-encodings.jpg https://jeffquast.com/soviet-encodings.jpg
- kstrauser 11y agoThat's an excellent point. Let's generalize that Windows-centric idea of a code page to something that can handle most/all known languages: a Universal Coded character set, if you will.