Groves, the next big thing (Re: ANN: XML and Databases article)

[schen at falconwing.com]

Hi John, everyone,

On Fri, 10 Sep 1999, John Cowan wrote:

> schen at falconwing.com scripsit:
> 
[ ... ]
> > Just as
> > XML is taking off as a simpler version of SGML, how about a simplified
> > version of groves?
> 
> That is just what the XML Infoset WD is all about.

[ ... ]

> AFAIK the Infoset meets the grove object model: there are objecdts with
> properties, some string-, integer-, or boolean-valued; other properties
> are either ordered or unordered lists of other objects.

I read the Infoset standard yesterday, but it seems to me that it's not
enough.  Among objections raised on this list is that too many things are
optional.  What I like about the grove system is that at least everything
is present but the application developer can specify using a grove plan
what information can be omitted (or rather, what info he/she wants).

Also it seems to me that groves go a step further in defining a canonical
representation using SGML markup and also a mental model using a tree-like
data structure.

I do appreciate the difficulty of defining the Infoset what with XML not
mandating that all processors validate documents.  But it should
be possible to define a complete InfoSet first as a grove, then define
that non-validating processors should present a grove plan that specifies the
InfoSet subset they do provide.  Other XML standards like XPath and the 
DOM can also provide a grove plan for their processing model.

One concrete advantage would be that developers who are doing particular
XML processing can specify a grove plan saying what they need out of the
InfoSet, then match it with XML processors and technologies like XSLT.

> Does the grove model capture the difference between ordered and
> unordered lists directly, or does that require extra properties?

That I'll leave for the grove experts to answer =)  From a brief scanning,
it looks like everything is ordered though.

. . . Sean.


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)