OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help



   Re: [xml-dev] semantics in schema (xsd)

[ Lists Home | Date Index | Thread Index ]

On Fri, Feb 27, 2004 at 10:50:02AM -0500, Irene Polikoff wrote:
> What you are bringing up is that in some countries a husband may be also
> a male. One way to handle it is to create a subclass of homosexual males

Actually it's nothing about whether one is heterosexual or
homosexual -- a homosexual man may marry a woman in every country
I'm aware of, and in some are encouraged to do so.  So can a bisexual.

The point I was trying to make, though, was just that it's an
example of overspecification in software.

Another (perhaps less controversial) example is the time I could
not book an airline ticket from a particular US airline... their
Web page had a "Canadian orders" page, but they checked that the
city in the billing address of your credit card matched the
"zip code" of your address... and the form didn't let you put
anything except a US-format zip code... my postal code starts
with an "M", and they didn't allow that... so the credit card
test filaed, and I couldn't order the ticket.

They've sinced fixed it, so I won't name them :-)

Other examples include requiring telephone digits to have
a particular number of digits or not allowing "extension"
or "ask for Susan" in text there...

These are all examples of the software applying too rigid a test
and failing with real-world data.

> This is, however, going quite beyond the scope of the original example. 
Agreed :-)

But it can be useful to think about sometimes, for people who work
with schemas that constrain data.


Liam Quin, W3C XML Activity Lead, http://www.w3.org/People/Quin/


News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS