Lists Home |
Date Index |
My primary goal is representing the structure of XML more efficiently,
hence in binary. The content I expect to be exactly the same encoding
as typical XML 1.1 use, however if an application needs to include
binary data, b64ing it or something similar obviously would be great to
avoid. An encoding attribute for that element can indicate that the XML
1.1 equivalent is b64, for example.
If an application needs to exchange many IEEE floats, then that is an
application-level agreement, not something that I am considering in my
main meant-for-standardization design.
Elliotte Rusty Harold wrote:
> At 10:39 AM -0400 4/11/04, Stephen D. Williams wrote:
>> Optimized XML is XML which has been tweaked to retain its existing
>> advantages while improving those aspects that need improving in
>> various circumstances. I.e. better (or no) parsing, serialization,
>> memory usage, CPU, size, avoidance of encoding (of binary image data
>> for instance), fidelity, etc.
> XML is text. There is no binary data in XML. You can encode text in
> binary, and you can encode binary in text using a variety of
> algorithms. However, if you're directly encoding binary data in your
> format that is not somehow a representation of text, then it is
> clearly not a representation of XML. It is something els, and that
> something else may be useful (I certainly don't think digital
> photographs should be encoded in XML) but it is still not XML. Not
> only that it cannot be processed with standard XML tools and APIs.
> XSLT, SAX, DOM, XPath, XQuery, XOM, have no means of representing
> arbitrary binary data. The "avoidance of encoding (of binary image
> data for instance)" means you've clearly moved beyonds the realm of
> XML. What you are then proposing is a new format, not an optimization
> of XML.
firstname.lastname@example.org http://www.hpti.com Per: email@example.com http://sdw.st
Stephen D. Williams 703-724-0118W 703-995-0407Fax 20147-4622 AIM: sdw