Re: [xml-dev] Re: Reflecting on a decade of XML

[Evan Lenz <evan@evanlenz.net>]

On 12/28/10 2:46 AM, Jean-Marie Gouarné wrote:
> --
> More basically, the lack of preliminary conceptual data modeling was, up to now, one of the major causes of defects.
>
> The (probably erroneous) discrimination between structured and so-called "unstructured" data, and the more or less conscious filing of XML in the unstructured side, installed a disastrous cultural break between the XML ecosystem and the old-style, database-oriented data modeling techniques, that is more familiar with the distinction between the "what" and the "how".
>
> As a consequence, a lot of XML-related issues used (and use) to be discussed without  preliminary shared premises on the basis of, say, documented UML class diagrams or so, and the markup takes precedence over the business meaning.
I assure you that a lot of people are using XML and even creating 
applications that use XML without thinking in such terms (conceptual 
data modeling, UML, etc.). Rather than a liability, I believe that is 
XML's strength. Agreement on syntax while being agnostic about how you 
want to use the syntax is what has enabled so many different approaches 
to using XML. The common syntax means we can make those different 
applications talk to each other.

I think the "cultural break between the XML ecosystem and the old-style, 
database-oriented data modeling techniques" is overall a good thing and 
far from disastrous.

My two cents. :-)

Evan