On Mar 8, 2004, at 4:20 AM, Misha Wolf wrote: > You are assuming that protocols carry documents. Lots of > protocols carry small discrete snippets of text in separate > 'fields'. In such cases, there seems no point in allowing > multiple encodings. Absolutely. I think that there are some situations where a single-encoding approach makes excellent sense, and other approaches where the UTF-8/16 choice is better. I don't think one group of applications is much bigger or more important than the other, so I don't think a SHOULD in either direction is justified. -Tim
This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 12 September 2008 07:02:02 GMT