RE: [xsl] Haskell programmer's rant about xslt

[Norm Birkett <Norm.Birkett@xxxxxxxxx>]

> -----Original Message-----
> From: Wendell Piez [mailto:wapiez@xxxxxxxxxxxxxxxx]
...
> I agree. I was puzzled by this criticism. Surely it wasn't about the
> encapsulation of logic in templates, subroutines, or (gasp) functions.
> So is he unhappy about data-driven transformations? (Does he have a
> better idea for this class of problem?)
...

I suspect that a very effective cure for XSLT-phobia would be to require a
person to use the high-level, general-purpose programming language of his
choice, along with an XML API like MSXML or System.XML, to transform a
smallish, not too complicated XML document from one schema to another. Round
about unit testing time, I bet he'd be ready to take another look at XSLT,
with a little more openness to a language that looks a bit different from what
he's used to, but is making more sense every hour.

When I was teaching math, I found that almost no students are capable of
grasping the value of a method until they have first been subjected to having
to use an inferior method--and the more inferior the better (pedagogically
speaking).

Norm