- From: Ian Graham <igraham@smaug.java.utoronto.ca>
- Date: Thu, 29 Jun 2000 11:55:00 -0400
- To: Chris Croome <chris@webarchitects.co.uk>
- cc: Christian Smith <csmith@barebones.com>, www-html@w3.org
If a browser receives a file from A Web server, and if the Web server sends a content-type header specifyinb hte character set of the data resource, then this is the charset that should be used. If the resource is of type text/html, then the browser should _ignore_ any charset value specified in a META element. If the resource is of type text/xml, then the browser should also _ignore_ the charset value specified using the encoding="xxx" portion of the XML declaration (if any). If the Web server does not specify a charset in the content-type header field, or if the file is retrieved by some mechanism (like ftp, or filesystem access) that does not provide content-type/charset information, then the browser should try and learn the charset using the mechanism appropriate to the type of data it is getting * encoding=".." for XML data * <meta http-equiv="...." > for HTML As for UTF-8, the drawback is that this is not supported by Navigator 3 (I think), and with navigator 4 you must be careful to put a META elemnent inside the document declaring the charset -- if you don,t then Navigator 4 will assume the wrong charset value. Ian On Thu, 29 Jun 2000, Chris Croome wrote: > On Thu 29-Jun-2000 at 04:30:14AM -0400, Christian Smith wrote: > > > > > > Is there any drawback to using UTF-8 for some browsers/platforms? > > > > Certainly. Some browsers may not support UTF-8 (but all modern ones do). > > Anyone have any idea which ones and which platforms? > > Chris > > -- > Chris Croome <chris@webarchitects.co.uk> > > http://www.webarchitects.co.uk/ > http://chris.croome.net/ >
Received on Thursday, 29 June 2000 11:55:15 UTC