[
Lists Home |
Date Index |
Thread Index
]
The point is not pessimistic; it is conservative.
It is based on prior experience with the W3C
specification process. Namespaces are a good
example.
First, they were just name disambiguators, hidden
system properties, then schema references, making them
part of the content. We seem to stay on the
slippery slope of minimalism and incomplete
design guidelines. That makes these processes
unreliable.
So experience says, don't believe or trust;
specify, verify, and hold
to the original agreement until a case is
made for change which adds value, not simply
specification compression. Given that namespaces are
supported as currently defined in current
specifications, and that changing that suggests
to reasonable people that DTDs must be discarded,
put into a separate spec, or redesigned, then
the impact of leaving the design as is is less
than the minimal change suggested.
One might ask, which would have the most
effect, keeping Namespaces in the current
rec or degrading the use of DTDs? My guess
is that if we include more than the XML-Devers
and the minorities of the TAG and the W3C
WGs, we may get a vote for no change.
Not everyone wants them or needs them. Those
that do have them now. Why put them in core?
That is the value added I'm waiting to hear about.
Otherwise, do what Tobin suggests and simply make the changes
which are not controversial in the two specifications.
I'd love what Joe suggests: give us back the
power of SGML DTDs if there is to be some kind
of redesign. But if we are really headed down
that path, I'd prefer to see what DSDL comes up
with. There is considerable experience backing
that up and reliable standards processes.
We should leave room for different organizations
to standardize and specify Internet systems.
The Web Is Not The Internet. The Web is as
Paul argues, RESTful. The Internet is not.
I am more convinced everyday that we need to
accept and potentially embrace a world in which
"information space" is processed by multiple
and even competing systems yet unifiable
not by addressing, but by declaration.
len
-----Original Message-----
From: Dare Obasanjo [mailto:kpako@yahoo.com]
I still don't see your point. It seems you are taking the most pessimistic
view of things but let's walk through your scenarios anyway.
|
