Peter Kirk wrote on 09/26/2003 02:21:59 AM: > >Unlike Jame's cup of wine, this really is a good analogy. Suppose the > >document is stored on the server in ISO 8859-1 and the browser requesting > >the page understands only EBCDIC. The server must convert it -- if it > >doesn't, it will appear on the client as complete garbage. As Jon > >mentioned, the server is the last one to touch it, and this illustrates
> >why it is appropriate for the server to touch it. > Is server software actually obliged to perform such conversions on > request? Surely, rather, browsers should be expected to support a > certain minimum set of encodings... Folks, feel free to spend your time bantering on about whether something should or shouldn't do this or that. But while you're at it, if you want to know whether the http encoding declaration is supposed to have precedence over the encoding declaration inside the HTML doc, go read the specs to find the definitive answer. Peter
