[
Lists Home |
Date Index |
Thread Index
]
On Sun, 5 Jan 2003 10:09:35 -0500, Simon St.Laurent
<simonstl@simonstl.com> wrote:
>
> I suspect more emphasis on educating people in the first place would be
> a more productive use of time than working on certification of any form.
FWIW, I suspect that we need time (measured in years) to figure out which
parts of the XML corpus are worth educating people about (and certifying
the status of their education) before we worry too much about this. There
are thousands of pages of "XML" specs in various states of Recomendation-
ness. Which are worth knowing, if you are a software developer not directly
involved in the XML racket? I don't think we have a clue, beyond the hard
core of XML and probably XPath/XSLT. And the subset worth knowing depends
on what you are planning to do with the XML. Rick's example of cacheing
parameter entities has a lot of relevance in that corner of the XML world
where big DTDs are relevant, but obviously not to the corner of the world
based on SOAP. Conversely, most of the raw bulk of the XQuery spec
describes the type system, but can be safely skimmed over by those in the
document world where almost everything is a String (or some type that can't
be fully described in XSDL).
Not to mention the uncertain status of various specs. Should Certified
(Certifiable?) XML Professionals be expected to know RELAX NG as well as
the minutiae of W3C XSDL? SQL-XML as well as XQuery? It would be hard to
specify "objective" criteria ... e.g. SAX has no formal status as a
"standard" but is obviously something that any programmer working with XML
should have some familiarity with. There are W3C Recommendations (XLink
comes to mind, <duck>) that could be ignored on a certification exam
without doing much damage to the world's productivity.... Even Microsoft's
blessing of XSDL as the One True Schema Standard could be rendered moot
(for the purposes of the discusssion here) if some little Office plugin
that lets the user write their schemas in RELAX NG and converts them to
XSDL becomes widely used to make "XML" palatable to end users. We shall
just have to wait and see.
I'm happy to let evolution take its course ... if knowledgeable software
developers can (as I expect they can) learn all that they really, really
need to know about XML in a few hours and pick up the other useful bits as
the need arises, so be it. I think most people learn the bare bones of
SQL, HTML, and other really really successful software technologies in a
few hours and then spend the rest of their careers plucking additional
nuggets out of the stream as required. That leads to a lot of bad
databases, webpages .... and XML systems, but Father Darwin provides a nice
elegant solution to that problem :-) And if the "bad" stuff doesn't
graciously die out, maybe something about it wasn't so bad after all.
|
