Attributes vs. text content (Was Re: RFC: Attributes and XML-RPC)

[Joshua E. Smith]

In my XML-conformant programming language (Nimble, mentioned here a couple
days ago at http://www.kaon.com/SDK ), I did what seemed to me a pretty
neat thing using attributes and elements together.

In many cases, an object (represented by an Element) needs to reference
another object.  I allow the Nimble programmer to do this either as:

<Image name='splash_screen' etc... />
<Application image='splash_screen' etc... />

-or-

<Application etc...>
  <Image etc.../>
</Application>

-or even-

<Application etc...>
  <Image name='splash_screen' etc... />
</Application>
<SomethingElse image='splash_screen' etc... />

The first approach is generally only useful when a machine is generating
the program (export from a 3D modeling tool, in my case).  Or if the Nimble
programmer is name-happy.

The second is just like XML-RPC.  Simple, elegant.

The last approach is particularly powerful, since it allows me to create
graphs  in what would otherwise be just a tree language.

The simple ID and IDREF DTD constructs (along with an #IMPLIED) then allow
validating XML editors to make sure you use defined names.

Doing this without attributes would be a real trick, and not nearly as
elegant.

So while I agree that sticking to just elements or just attributes can be
elegant in some contexts, neither rule is going to be the most beautiful in
every case.


-Joshua Smith


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)