[
Lists Home |
Date Index |
Thread Index
]
* Elliotte Harold wrote:
>Perhaps what we need, then, is not XML 2.0 but rather XML 1.0, fourth
>edition, that emphasizes readability and explicitly calls out and
>explains with examples the corner cases? Or maybe we should have an
>official W3C XML Primer instead, sort of like the W3C XML Schema Primer?
> I tend to doubt, though, that any XML 2.0 spec would be any more
>comprehensible than the current third edition XML 1.0 spec. Most W3C
>specifications are not written for end-users.
Indeed, people like to pretend that the comprehensibility of a technical
report contributes insignificantly or not at all to the success of the
technical report. To an amusing extend in fact, it is not uncommon that
suggestions for editorial improvements get rejected because the change
is either not actually purely editorial or it might not be but it's too
difficult to tell; or working group participants propose examples that
turn out to be wrong...
One can even observe systematic refusal to make editorial improvements,
for example, if you post a comment on a working draft and phrase it as a
question, you will typically (if you are lucky) just get an explanation
without any change to the specification, just as if it you are the only
one in the universe who would ask such a dumb question. Which might be
true in fact, implementers tend to implement what makes most sense when
faced with unclear aspects, waiting months, years, or even longer for a
response from the working group is typically not acceptable anyway.
Thruth is that it is difficult to find examples for non-compliance, non-
interoperability or incorrect external documentation for things that are
obvious from a specification to its entire readership. Except for HTML,
of course. But test suites will save us, won't they?
--
Björn Höhrmann · mailto:bjoern@hoehrmann.de · http://bjoern.hoehrmann.de
Weinh. Str. 22 · Telefon: +49(0)621/4309674 · http://www.bjoernsworld.de
68309 Mannheim · PGP Pub. KeyID: 0xA4357E78 · http://www.websitedev.de/
|
