Re: [xsl] XSLT Hello World - outreach

[Graydon <graydon@xxxxxxxxx>]

On Thu, Mar 27, 2014 at 06:24:06PM +0100, David Rudel scripsit:
> On Thu, Mar 27, 2014 at 2:15 AM, Liam R E Quin <liam@xxxxxx> wrote:
> > Ideas welcome.
> 
> Would it make sense to start with considering what prevents people
> from using XSLT for projects that yearn for it? I have a very limited
> view on what this is, but from what I can tell, there are basically 3
> things:
[snip David Rudel's experiences for brevity, I'm by no means disagreeing
with them!]

What prevents people from using Lisp?

XML doesn't work like anything else; XML is graph theory, XSLT is an
imperative tree-transformation language specifically for XML, and
nothing else works that way.  People with solid coding skills from other
domains try to use XSLT and wind up in the special hell that is trying
to make XSLT do anything imperative.  It builds up a reservoir of
loathing.

Since the analysts and the project-managers don't like it either -- XML
data is mostly illegible to them (doesn't go into Excel, you can't query
it[1], etc.) -- this makes advocating for XSLT difficult.  Why should we
go with rare (and more expensive) skills when we can find a way to do
this with SQL?

I've never got anybody to believe XSLT was easy; I have got people to
believe it was powerful, and I suspect the best way to evangelize for
XSLT is along that axis. "Here are things we can do reliably that are
much harder any other way".


-- Graydon


[1] of course you can query it, but that gets into more rare skills and
things the project manager and the analyst have never heard of, which
means they're against it because they have to say "I can't do that,
you'll have to get someone else."