W3C home > Mailing lists > Public > site-comments@w3.org > September 2006

Re: [Moderator Action] meta http-equiv form variants on your site

From: Ian B. Jacobs <ij@w3.org>
Date: Mon, 18 Sep 2006 16:34:15 +0000
To: Nick Levinson <nick_levinson@yahoo.com>
Cc: site-comments@w3.org
Message-Id: <1158597255.19717.58.camel@jebediah>
On Sat, 2006-09-16 at 23:48 +0000, Nick Levinson wrote:
> 
> The following, which appear on your website and seem
> to be a model for various other websites, vary:
> 
> <meta http-equiv="Content-type"
> content="text/html;charset=UTF-8" />
> <meta http-equiv="Content-Type" content="text/html;
> charset=utf-8"/>
> <meta http-equiv="Content-Type"
> content="text/html;charset=utf-8" />
> <meta http-equiv="Content-Type" content="text/html;
> charset=UTF-8">
> <meta http-equiv="Content-Type" content="text/html;
> charset=utf-8"/>
> 
> The content values differ by 1 space, the charset
> value is variously capitalized, and the http-equiv
> value is variously capitalized. Assuming we apply the
> convention that all should end in
> space-slash-closing-angle-bracket, are all of these
> correct?

Hi Nick,

The HTML 4 specification does not define a normative set of values
for the meta/http-equiv and meta/content (see the attribute definitions
[1]). Thus, the HTML 4 specification would not have any say over
case-sensitivity of values or spacing.

http-equiv is meant to provide some information to servers [2]:

 "HTTP servers use this attribute to gather information for HTTP
  response message headers."

Thus, I believe that it is up to each server to decide when and how to 
manage these hints provided by the author. I do not have any data handy
on how various servers interpret these meta elements. 

My guess is that servers would _benefit_ by being very permissive as
far as the case of attribute values and internal spacing.

Here's one example from apache [3] regarding the AddType directive:

  "The extension argument is case-insensitive, and can be specified with
or without a leading dot."

That doesn't address the question of how the <meta> element in an HTML
file gets converted into the content type in a server response, nor does
it state that media types themselves are case-insensitive, but it does
show that the server documentation may well be the place to look for
this sort of information.

 - Ian

[1] http://www.w3.org/TR/html401/struct/global.html#edef-META
[2] http://www.w3.org/TR/html401/struct/global.html#adef-http-equiv
[3] http://httpd.apache.org/docs/1.3/mod/mod_mime.html#addtype

> The first two are from
> http://www.w3.org/International/tutorials/tutorial-char-enc/#Slide0270,
> as accessed Sep. 5, 2006, in reverse order; the third,
> from http://www.w3.org/International/O-charset, as
> accessed Aug. 31, 2006; the fourth, from
> http://www.w3.org/Talks/1999/0830-tutorial-unicode-mjd/slide35-0.html,
> as accessed Aug. 31, 2006; and the last, from
> http://www.w3.org/International/tutorials/tutorial-char-enc/,
> as accessed Aug. 31, 2006, which also contained one
> identical to the second, from a different source.
> 
> I take it the following, which I had been using
> although I don't remember where I found or derived it,
> is wrongly punctuated:
> 
> <meta http-equiv="Content-Type" content="text/html"
> charset="UTF-8" />
> 
> While I program in XHTML and did in HTML, perhaps
> these are relevant to other W3C languages as well.
> 
> Thank you.
> 
> -- Nick
> 
> __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around 
> http://mail.yahoo.com 
> 
-- 
Ian Jacobs (ij@w3.org)   http://www.w3.org/People/Jacobs
Tel:                     +1 718 260-9447

Received on Monday, 18 September 2006 16:35:04 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 24 October 2012 16:21:30 GMT