W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > October to December 1999

Re: Is there pi?

From: Christopher R. Maden <crism@exemplary.net>
Date: Wed, 27 Oct 1999 14:43:43 -0700
Message-Id: <v01530501b43d1d1bb5ea@[209.157.134.15]>
To: w3c-wai-ig@w3.org
[Masayasu Ishikawa]
>crism@exemplary.net (Christopher R. Maden) wrote:
>
>> However, &pi; will only work in HTML 4.0 browsers; &#960; will only
>> work in some browsers if the document character encoding is defined as
>> UTF-8 or another full Unicode encoding;
>
>To clarify: numeric character references should work regardless of
>character encoding.  &#960; SHALL be interpreted as "GREEK SMALL
>LETTER PI" even if the document is encoded in US-ASCII (or whatever).
>Indeed there are some user agents which incorrectly treat NCRs
>differently depending on character encoding, but please note that
>such behavior is non-conforming.

Yes.  The behavior nonetheless exists - Netscape 4 (4.0 on Macintosh and
4.something on Solaris) is where I have encountered this.  MSIE on Windows
behaves properly; I don't know about on Macintosh.  Lynx, of course,
performs properly.

Designing a Web page based on SHALL is a fun exercise, but not very useful.
WILL is the relevant word for Joyce's question.

>Some pre-XML browsers support hexadecimal numeric character references,
>sush as Alis Technologies' Tango browser.  Tango does display &#x3c0;
>as "GREEK SMALL LETTER PI".

Tango added that feature after the XML WG developed the &#x convention.  I
didn't say that only XML browsers supported it, but that only browsers
developed after XML did.  Supporting something before it exists is
difficult.

-Chris

--
Christopher R. Maden, Solutions Architect
Exemplary Technologies
One Embarcadero Center, Ste. 2405
San Francisco, CA 94111
Received on Wednesday, 27 October 1999 17:44:49 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:13:45 GMT