[
Lists Home |
Date Index |
Thread Index
]
> -----Original Message-----
> From: mgushee@havenrock.com [mailto:mgushee@havenrock.com]
> Sent: Tuesday, September 10, 2002 19:52
> To: xml-dev@lists.xml.org
> Subject: Re: [xml-dev] attributes vs. elements
>
>
> On Tue, Sep 10, 2002 at 01:21:37PM -0400, Maciejewski, Thomas wrote:
> > This is sort of a newbie question but one that I never
> fully understood or
> > maybe just don't know how to explain it to others.
> >
> > When should one use an element and when should one use an attribute?
>
> That's a very common and very important question, and it's probably in
> a FAQ somewhere, but I have no idea which FAQ. In any case,
> there is no
> hard and fast answer to that question.
>
> The traditional distinction between elements and attributes
> is that the
> former are for content (information of interest to the end user or the
> consuming application) and the latter for metadata (information about
> information), and there is no clearcut, universal distinction between
> those two. Generally speaking, the text in a paragraph or the
> data in a
> database record would be considered content; the date that content was
> last modified will probably be metadata, but it's not hard to imagine
> scenarios where you might want to treat it as content. Ultimately,
> deciding whether something is content or metadata is up to
> the creators
> and users of the information.
>
> But back to elements and attributes. Some considerations that
> may affect
> your choice:
>
> * Elements may not have more than one attribute with the same name,
> but child element names can be repeated indefinitely (unless
> forbidden by a schema, of course).
>
> * XML parsers must preserve the order of elements, but not
> attributes.
>
> [ given the above two statements, we can say that in programming
> terms, the attributes of any given element are analogous to a
> hash (AKA mapping, dictionary, etc.), while the child elements
> are analogous to an array ]
>
> * Attributes can't contain elements--though of course they can
> reference them.
>
> * White space in elements can be preserved, whereas white space in
> attributes is normalized (leading and trailing spaces are
> stripped,
> each extent of internal white space is collapsed to a single space
> character).
Matt,
I don't follow you. If the schema is specified in XML Schema and the
datatype of an attribute is "string" (whose "whitespace" facet is
"preserve" by default), I don't believe white space is normalized. Are
you saying that an attribute of type "string" behaves differently from
an element of type "string" as to white space handling?
Alessandro Triglia
>
> * Attributes are more compact, since their names only occur once per
> instance and they have fewer 'overhead' characters: =""
> vs. <> </>.
>
> * Many people find elements more readable.
>
> I think there is a tendency for document-oriented
> applications to prefer
> elements, and messaging and automated data-exchange applications to
> prefer attributes--the rationale being that bandwidth is more
> important
> and human-readability less so. However, in cases where
> human-readability
> is unimportant, you might want to ask whether XML is an appropriate
> solution at all.
>
> That's my $.02.
>
> --
> Matt Gushee
> Englewood, Colorado, USA
> mgushee@havenrock.com
> http://www.havenrock.com/
>
> -----------------------------------------------------------------
> The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
> initiative of OASIS <http://www.oasis-open.org>
>
> The list archives are at http://lists.xml.org/archives/xml-dev/
>
> To subscribe or unsubscribe from this list use the subscription
> manager: <http://lists.xml.org/ob/adm.pl>
>
>
|
