That does sound interesting! I hadn't thought of the potential connection
between XQuery and MapReduce, but once you pointed it out, I could
definitely see how it could be implemented. Do you have some XQuery code
samples showing how you are doing folds?

On Tue, Dec 2, 2008 at 3:52 AM, James Fuller <james.fuller.2007@g...>wrote:

> u might want to take another look at xquery  ... I have found that
> most of the functional idioms are doable with extensions (1st class
> funcs) and XQuery lends itself readily to working with XML... u may
> even find you can port your code through time by directly using your
> existing xslt transforms from within XQuery as you figure out
> refactoring them.
>
> I have a signifcant project using XQuery and it uses a generic fold
> function as its main engine for 'doing stuff';  I think this choice in
> architecture will eventually give me the opportunity  to use a
> mapreduce algorithm to distribute processing ... well thats the
> theory. Hopefully I will make a release of this app soon.
>
> Getting the data structure right for defining rules in your app sounds
> like the crux of the problem ... grammers, parsing etc ... all old
> problems with solutions but still hard.
>
> keep us updated on your progress, Jim Fuller
>
> On Tue, Dec 2, 2008 at 11:47 AM, Christoph LANGE
> <ch.lange@j...> wrote:
> > On Tuesday 02 December 2008 11:03:04 Andrew Welch wrote:
> >> Something to consider is Scala, which contains an xml type:
> >>
> >> http://www.scala-lang.org/node/131
> >>
> >> ...so you have both a general purpose language with an inbuilt xml
> >> processing ability.
> >
> > I'm seriously considering a rewrite of my XML->RDF extraction library
> Krextor
> > (http://kwarc.info/projects/krextor/) in Scala.  The internal XSLT
> > implementation of the generic extraction algorithm that is independent
> from a
> > particular XML input language and from a particular output RDF notation
> (be it
> > text or XML) has become quite complex.  I would have liked to do parts of
> it
> > in functional style and got used to FXSL, but due to some technical
> > limitations (FXSL's implementation of currying), it didn't quite get as
> > functional as I wanted it to be.
> >
> > On the other hand, I want to keep it easy for the users of my library
> (none so
> > far, but some expected): They should be able to write easy patterns (e.g.
> > XPath) that map XML elements or attributes to URIs of ontology concepts,
> > whenever they implement a new extraction module for some XML language.
> >
> > Therefore, I need a language with XML built into the syntax.  Scala
> offers
> > this, but its XML document model is quite limited, compared to DOM.  (Of
> > course, it's more efficient instead, and DOM can be used with Scala, but
> not
> > with nice syntactical support like pattern matching.)  E.g., I sometimes
> need
> > patterns that traverse the parent or sibling axes, and for that, I'd need
> to
> > take care of passing around such arguments myself.
> >
> > Cheers,
> >
> > Christoph
> >
> > --
> > Christoph Lange, Jacobs Univ. Bremen, http://kwarc.info/clange, Skype
> duke4701
> >
> >
>
> _______________________________________________________________________
>
> XML-DEV is a publicly archived, unmoderated list hosted by OASIS
> to support XML implementation and development. To minimize
> spam in the archives, you must subscribe before posting.
>
> [Un]Subscribe/change address: http://www.oasis-open.org/mlmanage/
> Or unsubscribe: xml-dev-unsubscribe@l...
> subscribe: xml-dev-subscribe@l...
> List archive: http://lists.xml.org/archives/xml-dev/
> List Guidelines: http://www.oasis-open.org/maillists/guidelines.php
>
>


-- 
Kurt Cagle
Managing Editor, xml.com
O'Reilly
kurt@o...