[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: XML For The Average Monkey

• From: Arjun Ray <arjun.ray@verizon.net>
• To: "xml-dev@lists.xml.org" <xml-dev@lists.xml.org>
• Date: Tue, 25 Mar 2014 22:29:40 -0400

On Tue, 25 Mar 2014 09:48:33 -0600, Brian Aberle <xmlboss@live.com>
wrote:

| In December of 1998 the largest HMO committed to using XML for all data 
| transport in the implementation of a database that is shared between 
| hospitals. 

Interesting.  Was this discussed on the (old?) HL7-SGML mailing list?
(There used to an archive at www.mcis.duke.edu, but that has gone 404
for a while now, and the new www.hl7.org site is a labyrinth.)

| The only parser I can remember from back then was Xpat. 

You probably mean James Clark's expat.  Good choice.

  http://en.wikipedia.org/wiki/Expat_%28library%29


| It was the fastest tokenizer in the performance tests we ran.  We 
| wanted to make so many enhancements [...]

Out of curiosity, such as?

| In 2 weeks we had a home grown functioning tokenizer in a layer that 
| we could easily swap out with Xpat. It just got better over time [...]

Have you tried benchmarking against other alternatives?

	http://pugixml.org/benchmark/ 
 
| Just a little historic note about how all this came to be. 

A new parsing technique would be interesting.  Format tweaks in XML
aren't.  Really.

• References:
  • RE: XML For The Average Monkey
    • From: cbullard@hiwaay.net
  • RE: XML For The Average Monkey
    • From: Brian Aberle <xmlboss@live.com>