OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

 


 

   Re: [xml-dev] The general XML processing problem

[ Lists Home | Date Index | Thread Index ]

Bill,

Bill de hÓra wrote:

>
>>-----Original Message-----
>>From: Patrick Durusau [mailto:pdurusau@emory.edu] 
>>
<snip>

>>
>
>Are we really done then? There's a lot of work done after Earley, ie
>Marcus, (and of which Schematron is possibly a special case, I'm not
>sure yet) that suggest to me we're not done without resorting to syntax.
>The ideal at the XML level would be not to resort to context-free
>parsers that require probabilities (PCFGs). And there's still markup to
>deal with, even when it's called punctuation ;)
>
Certainly not done with syntax!

Point being that we have artificially limited the range of options that 
we have with syntax by the presumptions (unnecessary in my opinion) that 
underlie it.

Would not necessarily even require context-free parsers. Imagine (you 
can do this with SAX) that you specify rules that consume markup by 
namespaces (well, at least namespaces as in classic concur). Certainly 
could validate an XML document that is not "well-formed" through the 
various namespaces.

Imagine using a context-free parser where you specify the resolution 
rules for ambiguity. Resolution rules specified in RELAXNG? Suddenly 
your markup syntax becomes a much more powerful tool since you impose 
presumptions on the markup at the time of processing. Does this mean 
that current markup syntax needs further work? Yes and the development 
of that syntax will not be easy, particularly for those of us (I number 
myself in this group) who are habituated to thinking in terms of classic 
tree formulations.

There has certainly been a lot of work done after Earley and I am sure I 
have only scratched the surface of that vast body of literature. My main 
point being that we should examine the fundamental assumptions of markup 
syntax in light of advances in parsing theory and practice. Particularly 
when there are a whole range of problems that current markup practices 
deal with poorly, if at all.

Simon: Not ignoring your post. I need to pack to leave for Berlin later 
today and a proper response will take longer than I have at the moment. 
Will response by later this week.

Patrick

-- 
Patrick Durusau
Director of Research and Development
Society of Biblical Literature
pdurusau@emory.edu







 

News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 2001 XML.org. This site is hosted by OASIS