[
Lists Home |
Date Index |
Thread Index
]
Yes. OTOH:
In systems that must exchange information with rendering
and behavioral fidelity, it is useful to have a sample
implementation even if non-normative. We struggled with
this issue in VRML and X3D. For that reason, X3D specifies
an object model because a pure syntax or schema standard
simply won't do the job. More natural language and formulas
didn't work. If true interoperability is the idea (many
implementations; one language), then nail this to the door
over the meeting room of standards committees:
Data is portable. Systems interoperate.
XML doesn't solve the problem of interoperability any
more than it solves the problems of truth and authority.
If interoperability is the goal, one must get beyond a
syntax/structure/name + natural language specification.
We debated a reference implementation but that has a way of
strangling innovation for qualities such as speed and ease
of extensibility. Over time, the open source implementation
became the proving ground for those who require transparency
of process and code. That works reasonably well.
It was the lack of clear open source that killed many an
early SGML project some which have well-known names. So
while it is true that ISO does not endorse them, it is a
good idea to have implementors working side by side with
the standards members for reasons most of us here understand
I'm sure.
len
From: G. Ken Holman [mailto:gkholman@CraneSoftwrights.com]
At 2005-12-04 14:48 +0000, Fraser Goffin wrote:
>In reading over a variety of articles I did come across a comment (I
>think it was from Rick) that ISO ratification does not allow for the
>creation of a reference implementation. It wasn't clear whether this
>meany 'disallow' or doesn't encourage. Can you clear that up ?
I believe Mike has it right, though I cannot cite chapter and verse
... it is not the objective of the standards committee to produce
actual running code. Interestingly the level of semantic description
for ISO/IEC 19757-2 RELAX-NG is sufficient that merely "connecting
the dots" (for those who are able to code it) leads to a conforming
implementation.
Anyone can create implementations, but it is not our role to either
endorse them or "certify" them. That would very likely be the role
of a company whose business is testing who could come up with (or use
established) testing criteria and metrics with which to establish the
level of conformance as a service.
>This surprised me somewhat given, in my experience, the ability to
>put something tangible on the table that projects can pick up an use
>is a strong motivator for encouraging adoption. I realise that it
>may not necessarily be within the remit of ISO as to whether
>specific standards get used or not, but for those (like yourself)
>who clearly labour long and hard to push them over the line, I would
>assume that this affords some satisfaction.
The act of producing an implementation isn't in our area, and yes it
is very satisfying to see standards be successful in implementations.
|
