RFC: Attributes and XML-RPC

[Tyler Baker]

Marcus Carr wrote:

> Dave Winer wrote:
>
> > Attributes are for wispy little very optional things, if they are for
> > anything at all.
>
> Wispy little things like security levels? How would you represent the following pseudo-defence
> data:
>
> <chapter security="u">
>    <title>Wheeled Armoured Toys</title>
>    <para0 security="u">
>       <para security="u">WATs are your friend...</para>
>    </para0>
> </chapter>
>
> The two options that come to mind are either:
>
> <chapter>
>    <security>u</security>
>    <title>Wheeled Armoured Toys</title>
>    <para0>
>       <security>u</security>
>       <para>WATs are your friend...</para>
>          <security>u</security>
>    </para0>
> </chapter>
>
> ... which requires the evaluation of a child element before the application is able to
> understand how to deal with an elements such as para0 or para, (presumably the values are
> inherited) or something like:
>
> <u>
>    <chapter>
>       <title>Wheeled Armoured Toys</title>
>       <u>
>          <para0>
>             <u>
>                <para>WATs are your friend...</para>
>             </u>
>          </para0>
>       </u>
>    </chapter>
> </u>
>
> ... which obviously requires a ridiculously loose content model for element u and any other
> value that may have been assigned to the security level.
>
> I find attributes to be cleaner to look at and more logical to apply than elements, though I
> admit that the nature of the data may blur this distinction. Perhaps consideration of the
> diverse nature of data might have lead you to word your (copied) opening statement less
> emphatically?

I agree. In object to element mappings in an application, I have found it useful to think of
attributes as arguments passed to the constructor of the object you are creating and the child
elements of the containing element to be the properties of the containing element. But approach
number 2 I think is much more extensible since attributes can ony contain primitive type values
unless of course the attribute is just used as a pointer to some other element.

Tyler


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)