Re: Re: XSLT Date

Francis Norton <francis@xxxxxxxxxxx> wrote:
[rearranged]
> Basically I'm trying to say that I think there is a reasonably
> implementable spec that is worth doing, moans and all. Do you think it
> would be worth doing this subset or that only something with full
> localisation should be done? Or that full localisation is impossible,
> and we shouldn't attempt it all? 
The serialized lexical format should be ISO derived, as set forth by the
XSchema spec. Localization is unwise, any serious attempt would end up
as voluminous as the Unicode standard (including the character tables).

The problem with the moans is that computers can solve partial differential
equations, which most humans are bad at, but have trouble parsing randomly
formatted dates, something trivial for most people. Nevertheless, it's time
to adapt people to the computer's point of view instead the other way around.
This greatly simplifies software without inconveniencing people too much.
After some time people will get over it. And with XML buried under GUI
tools, only a few IT-geeks will have to deal with it anyway. 

> Yes, but if we describe it as "XPath 2.0 accepts understands dates in a
> system format" then we run less risk of the moans.
No, no, no, stick to the XSchema format and wait for the moans to fade.
Functions for formatting and perhaps parsing localized dates may be provided
later as add-on module. Don't put it into the core. This way implementors
have a choice, and there should be enough time to get the spec right.

Regards
J.Pietschmann

 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list