RE: Element names guidelines

> wouldn't it be
> <car>
> <description></description>
> </car>
> (obviously there are numerous designs patterns that one might
> use but I will stick with this one for the point of this post)
>
> Instead what seems to happen with structures where xml schema
> is used is that you get naming conventions like this
>
> <Car>
> <CarDescription></CarDescription>
> </Car>

This <Car><CarDescription> approach is an unnecessary repetition of the
"Car" concept and appears to be what Jeni Tennison warned against
earlier in this thread. IMO this is usually caused by not properly
modeling your data before authoring the schema.

You might want to look at the UBL Naming and Design rules as a good
source of a comprehensive standards-based approach. In UBL we have taken
an tripartite object oriented approach (object class, property term,
representation term) to element naming based on ISO 11179 and ISO
15000-5 ebXML Core Components. However,  we don't convey the object
class as part of the element name because it is always understood from
its parent.  The object class name is used in our type definitions.  Our
design pattern (Named appropriately by Eve Maler as the Garden of Eden
approach[see:
http://www.idealliance.org/papers/xml02/dx_xml02/papers/05-01-02/05-01-0
2.html]) provides a consistent method for defining types and reusing
global elements.  

The UBL NDR has finished the first OASIS review period and last week was
submitted to OASIS for consideration as an OASIS standard.  The UBL NDR
is being adopted by a number of US government agencies and is the basis
for the UN/CEFACT XML approach.  If interested, you can get it here:
http://www.oasis-open.org/committees/download.php/9943/cd-UBL-NDR-1.0Rev
1b.pdf.

Mark R. Crawford
Vice Chair - OASIS UBL TC
Chair - UN/CEFACT XML Syntax Working Group
Editor - UN/CEFACT Core Components
LMI Government Consulting

> -----Original Message-----
> From: xmlschema-dev-request@w3.org
> [mailto:xmlschema-dev-request@w3.org] On Behalf Of Bryan Rasmussen
> Sent: Tuesday, November 23, 2004 3:29 AM
> To: 'Frans Englich'
> Cc: xmlschema-dev@w3.org
> Subject: SV: Element names guidelines
>
>
>
>
>
> You have an element that you want to match the concept car
> description how would you naturally embody that concept in xml?
>
> wouldn't it be
> <car>
> <description></description>
> </car>
> (obviously there are numerous designs patterns that one might
> use but I will stick with this one for the point of this post)
>
> Instead what seems to happen with structures where xml schema
> is used is that you get naming conventions like this
>
> <Car>
> <CarDescription></CarDescription>
> </Car>
>
> In fact this is the naming convention where I work.
>
> this naming convention seems to be related to naming
> conventions often used in certain object oriented languages
> and of course in that we want to be able to say that a
> description of a car has various limitations on it that a
> generic description might not. (Which in the case of my work
> is also related to naming and design rules that do not allow
> local declaration of elements but require all elements to be
> globally declared)
>
> I believe that this an example of the drawbacks of xml schema
> as an xml validation language, not to mention its drawbacks
> as a language in the areas of data typing, and data binding
> descriptions.
>
>
>
>               
>
> -----Oprindelig meddelelse-----
> Fra: Frans Englich [mailto:frans.englich@telia.com]
> Sendt: 18. november 2004 15:25
> Til: xmlschema-dev@w3.org
> Emne: Element names guidelines
>
>
>
>
> Hello,
>
> At the risk of starting a flamefest, I wonder: What is the
> best naming conventions for elements and attributes? A
> crucial question in modeling XML formats.
>
> Assuming the phrase "car description" should be translated to
> an element name, and the criteria for judgment are easy to
> type and readability, there exist a number of different alternatives:
>
>
> <CarDescription>
> ----------------
> In my opinion, too elaborated; The capitalized C is
> unnecessary since the phrase is automatically distinguished
> at the boundaries by the tag characters.
>
> <carDescription>
> ----------------
> I'll call this the WXS-style. I find it (visually) beautiful
> and simple.
> It's
> relatively easy to read. Perhaps I'm biased by the Qt/KDE
> API. Regarding typing ergonomics it perhaps could be better
> than capitalized letters.
>
>
> <car-description>
> -----------------
> I'll call this the XSLT-style. IMO, more heavy and dense in
> its look, but easier to type.
>
> <cardescription>
> ----------------
> The Docbook-style. Hard to read but fast to type. That this
> naming scheme was chosen, suggests that other criteria than
> readability and typing ergonomics exists..
>
>
> But which one is best, and is there other alternatives? What
> other criteria are there, rendering my ramblings simplified?
> Does it depend on usage scenario?
>
> What was the reasonings behind XSLT's and WXS's styles?
>
>
> Cheers,
>
>               Frans
>
>
>
>
>

Received on Tuesday, 23 November 2004 13:01:13 UTC