[
Lists Home |
Date Index |
Thread Index
]
David Lyon wrote:
> Nothing beats an Internal binary format for processing speed.
But it won't be an internal binary format. It will be an external binary
format.
> If you think about xml parsing, most of the time is spent searching for the
> closing tag. If these things are known internally (how far to jump to skip to
> the end tag) then processing time will obviously come down dramatically.
No, not at all. You can't just skip to the end-tag. You have to check
everything in between for well-formedness errors that may be there
irrespective of whether the end-tag is correctly placed or not.
> not to forget, xml is binary already. Everytime I do 'debug myfile.xml' I see
> all these hex codes - oh and yes, the characters are shown on the far right
> just in case.
There's no rule at all that says XML has to be binary. A document
painted on the side of a building or printed in a book is still a
well-formed XML document. Binary computers are only about 60 years old.
I expect to see non-binary computers based on qubits become a
significant factor within my lifetime, possibly even replacing binary
computers as effectively as binary replaced analog and decimal
computers. Whether that happens or not, XML will continue to be
well-defined; and data stored in textual XML will easily make the
transition.
--
Elliotte Rusty Harold elharo@metalab.unc.edu
XML in a Nutshell 3rd Edition Just Published!
http://www.cafeconleche.org/books/xian3/
http://www.amazon.com/exec/obidos/ISBN=0596007647/cafeaulaitA/ref=nosim
|
