Indeed, UTF-16. I'm behind the times :) s/UTF-8/UTF16/ in my concern 4. Please correct me if I'm wrong, but you can't embed UTF-16 inside a serialization that's otherwise UTF-8 or another encoding. So my concern 3 is that even if it's specified, it's almost certainly *wrong* as which ever system created the serialization that the client is seeing at that point, must have encoded the content somehow. What happens when there's multiple characterEncodings, one for UTF-8, one for UTF-16 and one for some archaic windows code page? Rob On Tue, Oct 14, 2014 at 9:58 AM, Doug Schepers <schepers@w3.org> wrote: > Hi, Rob– > > Small comment... > > On 10/14/14 12:08 PM, Robert Sanderson wrote: > >> >> And my concerns with characterEncoding: >> >> 4. The encoding should be UTF-8 regardless. >> > > UTF-8 is not always sufficient. For some Asian scripts, you sometimes need > UTF-16. > > I'm not sure where that fits into your conceptual model, but we need to > take it into account. > > Regards- > -Doug > -- Rob Sanderson Technology Collaboration Facilitator Digital Library Systems and Services Stanford, CA 94305
Received on Tuesday, 14 October 2014 17:03:15 UTC