[
Lists Home |
Date Index |
Thread Index
]
Mike Champion wrote:
> - "XML" is more than just UnicodeWithAngleBrackets, it's a constellation of
> technologies that are inter-related in a mutually-supporting but formally
> ambiguous way. XML's success in the real world comes from their synergy, not
> any one (e.g. the syntax) in isolation.
The problem here (as I know that I have pointed out before, most recently in
the 'direct access to XML data' thread) is that document creators who assume
that consumers of their documents will be using all of the same ancillary
technologies as they themselves (e.g. WXS) create informationally impoverished
documents. If our goal is to insure the widest variety of the most useful
documents, best practice demands that the syntactic text be the whole of it. If
because you as author can delegate to a schema significant semantics which
would otherwise be expressed in attributes or other inherent document content,
you limit, if not preclude, some unexpected but valuable uses which I might
make of your documents downstream. Every additional ancillary technology which
you use effectively transforms your documents from what I understand as XML to
something which is not 'my' XML, at least until I join in the proliferation and
match your latest escalation.
This is not idle speculation. There are documents not created specifically for
me on which I utterly rely in my day job of fiduciary automation. Two years ago
I provided the profiles which the producers of those documents first used to
create them as XML. Nothing in those profiles was specifically intended for my
narrowly particular uses. In fact, the profiles were tightly drawn for the
immediate recipients whom the document creator understood that they were
creating them for. However, because *everything* was in the text I was able to
make use of those documents to automate all of the parts of the fiduciary
settlement/reporting/accounting process which the original creators of the
documents are entirely unaware of. Recently these document creators have hired
XML experts who have re-profiled those documents to be not just WXS-aware but
WXS-centric. As those experts progress in what they see as their task, their
documents are quickly becoming opaque to me. This is a market a niche where I
introduced the use of XML in order to be able to drive all of the necessary
operations off of a single set of documents which paralleled what the industry
participants were already generating as fixed-field binary records. I now
expect that by the end of this year some crucial documents will have become
again as opaque to me--and as useless for my processing--as the old binary file
formats were. At that point, management who hired those XML experts will have
to be shown that the cost of that applied expertise is that they now have to
pay to re-automate--at substantial up-front fixed cost--all of the processes
which they have been getting at no cost of infrastructure and for just the
variable cost of each execution. That is IMHO unmitigated stupidity.
> XML 1.0 explicitly states that it should be usable for a wide variety of
> applications.
I believe that I describe above the best way in which to realize that goal, as
well as the most direct means of frustrating it.
Respectfully,
Walter Perry
|
