[
Lists Home |
Date Index |
Thread Index
]
Elliotte Harold wrote:
> Bob Foster wrote:
>
>> If there were to be an XML 2.0, one of the goals should be that any
>> off-the-shelf parser generator can produce a parser for it. It's
>> ridiculously hard to write a correct XML parser; it ought to be
>> ridiculously easy.
>
> There are so many more developers who use XML than who write XML parsers
> that the cost of writing a parser is negligible when spread over the
> number of users. A small increase in the usability of XML is worth even
> the cost of a major increase in the complexity of writing a parser. We
> should not be aiming for a world where everyone writes their own parser.
> No matter how easy it is to do, developers will get this wrong the first
> few times out the gate. The resulting practical subsets and supersets of
> XML would be an interoperability nightmare. We're better off with fewer
> parsers written and debugged by experts, rather than everyone writing
> their own custom one-off parser.
A reasonable pov, probably represents the majority, I just don't agree
with it. A barrier to entry is a barrier to entry. The fact that it's
hard to write an XML parser means it's hard to understand how to use
XML, too. How many people on this list published thick books explaining
the ins and outs of something that is, at heart, just tagged text? Plus,
don't think for a minute that the "experts" who wrote parsers got them
right. IIRC, Crimson never did check DTDs for ambiguity; it did
backtracking.
I'll stand by my earlier comments.
Bob
|
