Re: [xsl] ANNOUNCE: Petition to withdraw xsl:script from XSLT 1.1

[Robin Berjon <robin@knowscape.com>]

At 12:27 02/03/2001 -0500, Charles Reitzel wrote:
>I think this argument might be falling into the category of
>giving-us-enough-rope-to-hang-ourselves.  I think that you'd like XSLT less
>if it didn't.  <xsl:script> is probably one of the less dangerous features
>of XSLT.

I do Perl at least half of the day, seven days a week so I'm usually all
for giving people enough rope to hang themselves, and then some. But if
these things are to be web broswers, then we're giving people (and browser
vendors) enough rope to hang us. I don't understand your point about
xsl:script being on of the least dangerous features of XSLT.

>You might not approve, but plenty of people code JSPs, which are
>probably an uglier embedding of Java than <xsl:script>.

I don't think that my approbation is in any way required here :) Perl
(especially mod_perl) has been having a gazillion ways of embedding code
within pages, some similar to JSP, some better, some worse. Experience
shows that this tends to bring people to put more than just presentation
logic into their pages, but also throw in business logic. It has lots of
perverse effects. What I love with XSLT is that you can't put business
logic there, so I don't have to look over someone else's shoulder to check
whether they're putting the right thing in the right place. Blogic returns
a DOM, Plogic transforms it into something that looks good.

I think xsl:script will start mixing those two things, but I don't care if
others decide to shoot themselves in the foot. I would just like to avoid
seen the web cluttered with th ehtml-script-kiddie brand of xslt. Perhaps
we can't avoid that though...

What I'd like to see is a cleaner separation, a la xbind. That would help a
lot achieve more portable stylesheets, and it's reusable within other contexts.

-- robin b.
By the time they had diminished from 50 to 8, the other dwarves began to
suspect Hungry.