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

Re: A multi-step approach on defining object-orientednatureof

• To: xml-dev@l...
• Subject: Re: A multi-step approach on defining object-orientednatureof DOM
• From: Joe English <jenglish@f...>
• Date: Wed, 21 Aug 2002 13:38:33 -0700
• In-reply-to: <ZW2ZLTOIC06RO5PN86LK76F0W1Y4Y.3d63e01b@MChamp>
• References: <ZW2ZLTOIC06RO5PN86LK76F0W1Y4Y.3d63e01b@MChamp>

Mike Champion wrote:

> Can the current [namespaces] spec be "profiled" to suggest sane best
> practice here [...] ?

I think so.  See <URL: http://www.flightlab.com/~joe/sgml/sanity.txt >,
which proposes a best-practice profile (called, in fact, "sanity").

The main points are:

    * Do not bind a single prefix to different URIs at different places
    * Do not bind a single URI to different prefixes at different places
    * If possible, put all namespace declarations on the root element

To that I would further add:

    * Do not use default namespace bindings
    * If possible, do not use QNames-in-content

Documents that follow these rules avoid most of the problems
involved with namespace scoping.  If documents can be
guaranteed to follow these rules, applications can be
considerably simplified (basically, it means they can 
track the namespace environment on a per-document basis 
instead of a per-node basis.)


--Joe English

  jenglish@f...

• Follow-Ups:
  • Re: A multi-step approach on defining object-orientednatureof DOM
    • From: John Cowan <jcowan@r...>

• References:
  • Re: A multi-step approach on definingobject-orientednatureof DOM
    • From: Mike Champion <mc@x...>

• Prev by Date: Re: ANN: XML Schema of measurement units
• Next by Date: Re: A multi-step approach on defining object-orientednatureof DOM
• Previous by thread: Re: A multi-step approach on definingobject-orientednatureof DOM
• Next by thread: Re: A multi-step approach on defining object-orientednatureof DOM
• Index(es):
  • Date
  • Thread