Re: Doesn't the ALT field of an image support ä?

Foteos Macrides (
Fri, 26 Apr 1996 12:57:25 -0500 (EST)

Date: Fri, 26 Apr 1996 12:57:25 -0500 (EST)
From: Foteos Macrides <>
Subject: Re: Doesn't the ALT field of an image support &auml;?
Message-Id: <01I3ZXOEFF02008MM5@SCI.WFBR.EDU>

Eva <> wrote:
>>>At 10.59 1996-04-20 -0700, you wrote:
>>>The specs say:
>>>"The alt text can contain entities e.g. for accented characters or special
>>>symbols, but it can't contain markup. The latter is possible, however, with
>>>the FIG element"
>>Ummm.. what spec says that? The _expired_ march '95 HTML 3 draft?
>>Perhaps. Please cite your source.
>Cited from (Permitted
>I do understand that the draft's expired but if there is any better and more
>up-to-date source of information regarding hlml3 specs I'd be glad to know
>where I can find it.
>>>Entities in ALT don't validate.
>>I'm pretty sure you're mistaken.
>None of my pages would ever validate. I tried it last just a few minutes ago.
>SGML error at -, line 27 at """:
>       NAME = "%26;aring;lder" attribute value defaulted: invalid character
>%26;aring; is in fact &Aring; 

	Do not hex escape the ampersand if you want the &aring;
treated as a character entity ("replaceable character data").
Also, unlike entities, hex escaping does not have terminators
(%26 not %26;), and is inappropriate for ALT values.  If you
wanted the entity displayed as "&aring;", rather than translated,
you'd use "&amp;aring;"

	ALT is declared as CDATA, and the original CERN libwww
SGML parser, upon which Lynx was based, requiring RCDATA for
translation of character entities, which is why I had assumed
the problem was old Lynx code.


 Foteos Macrides            Worcester Foundation for Biomedical Research
 MACRIDES@SCI.WFBR.EDU         222 Maple Avenue, Shrewsbury, MA 01545