Quote:
Originally Posted by Rev. Bob
Yeah, I just heard about this odd behavior over in a Calibre thread. My main experience is on the Web, where a UTF-8 file is expected to be a UTF-8 file, BOM and everything. Deliberately breaking true UTF-8 files in this way is counterintuitive at best...but I also understand that if some of the big reading software requires broken files, that's what you need to give 'em.
|
Files without a BOM are not broken. Following the Unicode standard is not odd behaviour. This is not an opinion of mine, rather a simple fact straight from the standard. But, if the writers of the Unicode standard can't convince you that not using a BOM is perfectly valid and even recommended, I might as well give up
