W3C home > Mailing lists > Public > www-validator@w3.org > September 1999

Re: Problems with & in URIs

From: Liam Quinn <liam@htmlhelp.com>
Date: Sun, 19 Sep 1999 20:01:01 -0400
Message-ID: <37E5793D.F5C00F0B@htmlhelp.com>
To: "L. David Baron" <dbaron@fas.harvard.edu>
CC: www-validator@w3.org
"L. David Baron" wrote:
> <a href="http://foo.com/search?keyword=food&amp;time=today">Today's food</a>
> Every browser I've tested this on (probably something like NN 4.x, IE
> 5.0, and Opera 3.6) handled it correctly.  I'm not sure how far you
> must go back in history to find one that doesn't.

I've tested all the major Netscape releases since 1.0N and all properly
handle &amp; in attribute values.

In testing around 30 browsers, I've only found two which fail to
interpret &amp; as & in attribute values:

- Amaya (http://www.w3.org/Amaya/), including the current version 2.1. 
Amaya is a rarely used test-bed browser.
- Gzilla (http://www.gzilla.com/), including the current "stable"
version 0.2.2.  Gzilla isn't yet a usable browser.

Netscape 3.x has many, many more users than Amaya and Gzilla combined,
and Netscape 3.x gets confused in some cases when the & is *not* written
as &amp;.  For example, Netscape 3.x will fail to follow the invalid
link given by href="foo.cgi?chapter=1&section=2" since it assumes that
&sect is the entity for the section sign.  So it seems that valid HTML
is safer in this case.

Liam Quinn
A Real Validator for Windows, http://arealvalidator.com
Web Design Group, http://www.htmlhelp.com
Received on Sunday, 19 September 1999 19:59:57 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 14:17:25 UTC