[
Lists Home |
Date Index |
Thread Index
]
At 10:41 PM +1000 7/22/02, Rick Jelliffe wrote:
>The Core WG has already set a precendent by removing the
>checks for language from the spec with an erratum, which did not
>cause a ripple. Perhaps this could be made into a general policy:
>that backwards-compatible changes which relate to tracking external
>standards will be dealt with by errata rather than by minor versions.
>
Not really. That was a very weird case. The rules were in there but
they weren't actually referenced by any other production. They
weren't reachable. Thus a very careful, fully accurate implementation
of XML 1.0 as originally published did not check these rules.
However, not all implementations were that correct and careful, and
some did check these rules. Clearly the rules were confusing, and
thus they needed to be removed since that more clearly demonstrated
what the XML 1.0 specification had always actually said, as long as
you didn't jump to the conclusion that you could use those rules.
Furthermore, the change did cause some problems because some parsers
did *incorrectly* implement those rules. This caused real problems in
interoperability tests. When the change was made, the bugs in some
parsers did need to be fixed. (Arguably they needed to be fixed
before too. It's just that no one realized that.) I'd say it made a
ripple.
--
+-----------------------+------------------------+-------------------+
| Elliotte Rusty Harold | elharo@metalab.unc.edu | Writer/Programmer |
+-----------------------+------------------------+-------------------+
| XML in a Nutshell, 2nd Edition (O'Reilly, 2002) |
| http://www.cafeconleche.org/books/xian2/ |
| http://www.amazon.com/exec/obidos/ISBN%3D0596002920/cafeaulaitA/ |
+----------------------------------+---------------------------------+
| Read Cafe au Lait for Java News: http://www.cafeaulait.org/ |
| Read Cafe con Leche for XML News: http://www.cafeconleche.org/ |
+----------------------------------+---------------------------------+
|
