{"items": [{"author": "BDan", "source_link": "https://plus.google.com/103775592027106438640", "anchor": "gp-1317393578520", "service": "gp", "text": "I disagree.  ISO-8859 and Windows-1252 are still very common, and their upper ranges do not agree with UTF-8.  If UTF-8 were really the standard for plain text, you could just set all your software to assume anything unlabeled was UTF-8, and it would always work, but that isn't the case.  Non-extended ASCII can be pretty well relied upon (nobody uses EBCDIC anymore), but that's about it -- once you have any bytes over 0x7F, you still need to know the encoding.", "timestamp": 1317393578}, {"author": "Peter", "source_link": "https://plus.google.com/103618186481362054522", "anchor": "gp-1317394901965", "service": "gp", "text": "I've run into a fair amount of UTF-16, too.", "timestamp": 1317394901}]}