[
Lists Home |
Date Index |
Thread Index
]
At 8:23 AM -0500 4/11/02, Bullard, Claude L (Len) wrote:
>Rick has it right. When the decision to
>make a Draconian Parse the centerpiece of XML
>to support a well-formed core, one remark made
>by the a supporter was that this would ensure the
>clean up of malformed HTML. A counter argument was
>that that approach sounded good but that practical
>programmers would still handle the errors and
>go on because that is what customers would want.
>In hindsight, the counter argument has so far
>been right with regards to HTML.
A significant reason for draconian error handling was the direct
request of vendors of some notable HTML processors, who admitted the
problems in HTML practice encouraged by their software, and asked for
draconian error handling on that account. It was a "please stop me
before I kill again" kind of argument....
Everyone in the group recognized that standards are a weak constraint
on behavior, and that the solution might be partial, or take years to
affect practice significantly.
-- David
--
-------------------------------
David Durand | 12 Bassett St.
david.durand@ingenta.com | Providence RI, 02903-4628 USA
VP, Software Architecture | 401-331-2014 x111 Cell: 401-935-5317
ingenta plc | FAX: 401-331-2015
http://www.ingenta.com/
|
