[
Lists Home |
Date Index |
Thread Index
]
On 10/17/05, basudeb gupta <basudeb0001@yahoo.com> wrote:
>
> How is the XML Data actually exchanged? Is it in its
> full glorious expanded form? If so, is it not a great
> loss of bandwidth?
>
> If not, I assume there are compression mechanisms
> available in the send and receive functions used by
> such programs which exchange XML data? Can someone
> please guide me on the current state of the art?
This is not an ignorant question, it's an xml-dev permathread. There
is a widespread complaint that XML is "bloated and slow" and that this
is a problem that needs to be remedied. There is an equally
widespread belief (enshrined in the XML spec) that terseness is not a
design criterion, that human readability and exploitable redundancy
trumps machine efficiency, and that the "binary XML" cure is worse
than the inefficiency disease.
The "efficient interchange" camp seems to be winning this year, at
least at W3C. There will probably be a working group formed soon to
develop an Efficient XML Interchange Recommendation that is
(supposedly) going to achieve a wide range of use cases. There is a
lot of debate about this within member-only W3C mailing lists (and I'm
on record as opposing it). Still, the proponents and the W3C have gone
to some lengths to ensure that the evidence of whatever improvements
in efficiency are achieved and whatever threats to interoperability
are uncovered are made public before the working group's spec achieves
unstoppable momentum.
Shameless plug: We've organized a panel discussion at the XML 2005
Conference (http://2005.xmlconference.org/ ) on Efficient XML in which
people on all sides of this question have been invited. Come to the
XML 2005 conference, listen to what these people have to say, and
you'll have a chance to speak your piece during the Q&A period and
lobby people for your point of view at the dinner afterwards.
|
