RE: [xml-dev] Is it time for the binary XML permathread to start up aga

[noah_mendelsohn@us.ibm.com]

Alexander Philippou writes:

> People use FI for good compactness without suffering a processing 
> penalty as with gzip.

I stand corrected.  When I was more directly involved in this discussion 
in the early days, most of the focus of the FI work seemed to be on 
solving the problems of Web Services deployments that were (at the time) 
heavily CPU bound doing XML parsing and deserialization, and were 
therefore stuggling to keep a reasonable number of high speed networks 
saturated at acceptable CPU cost.  Things may have changed since I was 
following things closely.

--------------------------------------
Noah Mendelsohn 
IBM Corporation
One Rogers Street
Cambridge, MA 02142
1-617-693-4036
--------------------------------------