Re: Andrew Layman and Don Box Analysis of XML Optimization Techni ques

Noah,

FYI:

/ noah_mendelsohn@us.ibm.com was heard to say:
[...]
| In the particular case of FastXML, my understanding is that there were two 
| flavors.
[...]
| depended on type information.  Whether, for example, it could distinguish 
| the following two instances:
|
|         <e xsi:type="xsd:integer">123</e>
|         <e xsi:type="xsd:integer">00123</e>

You are correct, there are two flavors. In it's "regular" mode, the
original document can be reconstructed by the receiver, so yes it can
distinguish between these two instances. It has another flavor that
can use an integer encoding algorithm. But the decision about which
flavor to use can be selected by the application.

| To be a true Infoset implementation usable in SOAP, for example, you must 
| be able to distinguish the above.  Note that the usual digital signatures 
| on these will be different.
|
| Are there published benchmarks of both of the above?

All of the performance numbers on the FI site were measured in
"regular" mode. In the regular mode, nothing special has to be done to
support SOAP or digital signatures, you can simply transmit the
document on the wire in FI then reconstitute the original infoset at
the other end and perform SOAP/DSig processing as normal.

                                        Be seeing you,
                                          norm

-- 
Norman.Walsh@Sun.COM / XML Standards Architect / Sun Microsystems, Inc.
NOTICE: This email message is for the sole use of the intended
recipient(s) and may contain confidential and privileged information.
Any unauthorized review, use, disclosure or distribution is prohibited.
If you are not the intended recipient, please contact the sender by
reply email and destroy all copies of the original message.