[
Lists Home |
Date Index |
Thread Index
]
Bullard, Claude L (Len) wrote:
> We are assuming that the W3C validator at
> http://www.w3.org/2001/03/webdata/xsv
> is the gold standard (not quite but really the reference implementation):
Assuming that any WXS validator is a "gold standard" would be a serious
error. I am quite sure that Henry Thompson or the other authors have
never claimed anything of the sort.
The following is based solely on anecdotes and personal observation, but
I believe it to be true. Every WXS validator has significant errors, and
XSV is no exception. The serious WXS user has little recourse but to
validate with several fairly reliable validating parsers and appeal to
the text of the standard if they report differently. I would choose the
XSV, XercesJ, MSV and MSXML parsers. I can't be sure that at least one
of this set will validate all documents correctly, but I am pretty sure
that all of the set will fail on at least one test case.
Bob Foster
|
