[XSL-LIST Mailing List Archive Home] [By Thread] [By Date]

RE: Can solve the N-queens - but can't count!

---

---

I think those translate directly into user benefits. The lack of side
effects in DSSSL meant that you could build up reusable function libraries
that could be used with great confidence (since you knew they wouldn't bogus
up some global state somewhere). Also, suppose you have a gigundous XML file
in an XSL-aware editor and you change one tag which is in your relatively
small viewing window. Should you have to re-render the whole file just to
display that view window? Side-effect-freeness means (I think!) that you
don't.

James

-----Original Message-----
From: Kay Michael [mailto:Michael.Kay@xxxxxxx]
Sent: Wednesday, June 16, 1999 1:27 PM
To: 'xsl-list@xxxxxxxxxxxxxxxx'
Subject: RE: Can solve the N-queens - but can't count!


> Mike Kay wrote:
> > I think it would be useful to the debate if someone could 
> articulate the reasons why stylesheets should be side-effect-free. 
> 
To which David Rosenborg responded:

> Here are two reasons from which we have practical experience: 
> lazy evaluation and
> implicit multithreading. In general the draft does not say 
> anything about the order in which things are processed.
> 

I had a kind of feeling that the responses were going to be along the line
"if we restrict what the user can do then we can have much more fun as
implementors". Show me a user benefit!

Mike


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


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

---