[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Ampersand in URLs (was: RE: </> as end tag)

• From: dgd@c... (David G. Durand)
• To: xml-dev@i...
• Date: Tue, 4 Nov 1997 13:03:33 -0500

At 8:27 PM -0500 11/2/97, Chris Lovett wrote:
>As for the ampersands, this is a real problem.  We found with our experience
>with CDF that customers just can't handle putting & inside their URL's.
>We want to comply with XML standards, but we also want XML to be successful
>in the marketplace.  One area that we didn't compromise is with case
>sensitivity.  The new parser is fully case sensitive - but with a switch
>that sets it back to case insensitive for those people that are reading XML
>that was generated before case sensitivity was decided.  You have to make
>some tough compromizes sometimes.

There was a query on the XML-SIG about HTML and the ampersand rule (XML
agrees with the HTML standard, but not all HTML implementations). I thought
that my answer fits well in this discussion as well.

Internet Explorer, ironically, already insists on the escaping of ampersand
in some circumstances. All that I've tested, actually, but I won't assume
that it follows the standard -- if they do, "some" should be changed to
"all". I am not sure about the story with whitespace, but in fact, if they
don't require & before space, it matters little to me, since space
isn't legal in a URL.

I don't see ampersand as a show stopper, especially once people realize how
useful entities can be in modularizing long URLs. And, as Paul notes, we
can fall back on the authoring tool argument. More important, since we have
"draconian" error handling in XML, simple testing of the document will
ensure that the error is detected (rather than the HTML case, where it
depends on the browser that you test with). One of the biggest problems
with HTML has been that that the standard and the implmentations differ(ed)
so widely and on so many points -- this is a primary reason that we should
be very careful to implement XML exactly. Consistent parsing will go a long
way to salve the wounds of slight differences from HTML. Divergent syntax
in any software that purports to be XML-compliant will cause real problems
from users, who may not be technical enough to read and understand the
specification to judge correctness of implementations.

We're sure to have bugs, but implementors we have a very real
responsibility to conform in every way that they can, regardless of what
design decisions they would rather have made differently. This truth is
what makes standards such an object of (seemingly pointless) passion --
because you have to take them as they _actually are_ if they are to have
the value that they promise (even when you feel that that value is
uncoscionably less than it could have been).

   -- David


_________________________________________
David Durand              dgd@c...  \  david@d...
Boston University Computer Science        \  Sr. Analyst
http://www.cs.bu.edu/students/grads/dgd/   \  Dynamic Diagrams
--------------------------------------------\  http://www.dynamicDiagrams.com/
MAPA: mapping for the WWW                    \__________________________



xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@i...
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo@i... the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo@i... the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@i...)

• References:
  • RE: </> as end tag
    • From: Chris Lovett <clovett@m...>

• Prev by Date: Re: How best to represent unrepresentable characters in NAMEtokens?
• Next by Date: RE: How best to represent unrepresentable characters in NAMEtokens?
• Previous by thread: RE: </> as end tag
• Next by thread: RE: </> as end tag
• Index(es):
  • Date
  • Thread