OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: More binary XML



On Tue, 10 Apr 2001, Stefan Zier wrote:

> A generic binary format would take away the pain of creating specialized
> binary representations for different applications - WBXML is actually almost
> generic. I think the critical question here is: Can we come up with a
> one-size-fits-all solution that fits most relevant applications well enough
> to be useful?

That's the spirit :-)

I'm aiming for something quite simple to describe, since I feel that will
appeal to the kinds of markets XML appears in.

Dilemma:

Do I use 32 bit lengths for everything, meaning that no CDATA or element
name or attribute name or namespace URI or PI body or target can be more
than 4Gb in length? (Don't laugh, it may be an issue). Or do I encode the
length scale in the top two bits of each byte-long tag, so we can use 8
bit lengths for most stuff, 16 bit lengths for large spans of unbroken
CDATA text, and 32 bit lengths for embedded images and pathological
cases? This would be a slight increase in complexity, but it'd be more
compact in the common case of strings being quite short.

...those are the kinds of issues I'm mainly worried about.

ABS

-- 
                               Alaric B. Snell
 http://www.alaric-snell.com/  http://RFC.net/  http://www.warhead.org.uk/
   Any sufficiently advanced technology can be emulated in software