• From: Liam R E Quin <liam@w...>
• To: Brian Aberle <xmlboss@l...>
• Date: Sun, 23 Mar 2014 19:31:08 -0400

On Sat, 2014-03-22 at 23:40 -0600, Brian Aberle wrote:
> [...]

>  It's OVER twice as fast as the more traditional "memory copy design"
> naturally because the iterations to the temporary structure are
> eliminated, it goes beyond 2 times as fast because the tokenizer uses
> neither SAX nor DOM, but a more efficient alternative to SAX that
> avoids pushing a variable number of arguments depending on the token
> type via the SAX calls.

I'd be interested in detailed comparisons between your approach and
Efficient XML Interchange (EXI), which can transmit XML documents
without actually sending pointy brackets at all.

Although there's certainly interest in making XML faster, it has to work
e.g even if attributes are re-ordered or rewritten. In other words, you
can't break the promise than any XML processor can read any XML
document, without creating a fragmentation.

But it might be acceptable to have something as fragile as (my
understanding of) what you describe solely for the purpose of
transmission.

However, in that case, it should be compared to EXI and, if appropriate,
could perhaps be used as a basis for further work within the EXI Working
Group.

Liam


Liam
		  

-- 
Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/
Pictures from old books: http://fromoldbooks.org/
Ankh: irc.sorcery.net irc.gnome.org freenode/#xml

• References:
  • RFC for XML Object Parsing
    • From: Brian Aberle <xmlboss@l...>