Re: DSSSL side effect-freeness

[Pierre Mai <dent@xxxxxxxxxxxxxxx>]

>>>>> "FAC" == Frank A Christoph <christo@xxxxxxxxxxxxxxxxxx> writes:

    FAC> So, recalling that laziness does not mix well with side
    FAC> effects, you can see that there are usually two extremes for
    FAC> a functional language: either it uses CBV and has
    FAC> side-effects, or it uses Need and has no side-effects.
    FAC> Clearly both alternatives have their advantages.

    FAC> But now consider the DSSSL expression language.  In the world
    FAC> I have just drawn, it is a very strange beast indeed: it
    FAC> isn't lazy and it hasn't got any side-effects.  This puts it
    FAC> smack in the middle of two Good Things, yet possessing
    FAC> neither, like a child at Christmas that can't make up its
    FAC> mind about which present to open first.

I'd just like to point out, that there are a number of pure
functional, yet non-lazy (i.e. eager) functional languages, e.g. Opal,
developed locally at the Technical University of Berlin, see
http://uebb.cs.tu-berlin.de/~opal/.

Apart from eagerness being easier to implement, it usually leads to
language implementations that have less state to carry around with
them.  Currently, this _can_ lead to faster execution times and/or
smaller memory footprints than implementations of lazy languages
achieve, although they should theoretically be faster/more
efficient...

Especially the interaction possibilites with "current-*" in groves
might lead to much state being carried around, though this is only
speculation on my part...

On the whole I rather think, that the decission for eagerness was
probably mostly carried by the arguments for simplicity and/or
compatibility with the Scheme way of things, rather than by
theoretical considerations for speed and equivalence...

Also I wonder if it would not still be possible (even if not really
standard-conformant) to implement DSSSL in a lazy way, and probably
most Style-Sheets would continue to work unchanged...
   
    FAC> distributed, though -- and I can't imagine distributed groves
    FAC> being a typical case anyway.

Well, if we look at the corporate world, where large technical
document repositories are being maintained and deployed as IETMs for
example, I would rather think that distributed groves might be heavily
used...

Just my two centieuros...

Regs, Pierre.

-- 
Pierre Mai <dent@xxxxxxxxxxxxxxx>	http://home.pages.de/~trillian/
  "Such is life." -- Fiona in "Four Weddings and a Funeral" (UK/1994)


 DSSSList info and archive:  http://www.mulberrytech.com/dsssl/dssslist