[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: Relationships [was RE: James Clark: XML versus the Web]

• From: "Andrew S. Townley" <ast@atownley.org>
• To: "Costello, Roger L." <costello@mitre.org>
• Date: Mon, 6 Dec 2010 13:04:29 +0000

Hi Roger,

I don't know if you're familiar with it or not, but the Topic Maps family of ISO specifications[1] does a pretty good job of expressing relationships between things.  However, the issue would likely boil down to one of choosing an uncomplicated syntax to express relationships which are potentially very complex in nature.

I can't remember who mentioned it earlier in the thread, but recasting XML in terms of graphs vs. trees might be necessary.  At the very least, trying to ensure that your trees were graphs and you had a standardized mechanism for expressing relationships between nodes.  Identification of nodes within the same document could be done using XPath.  Identification of nodes outside the same document is a bit trickier, but there are certainly ways to solve this problem too.

Topic Maps also has robust identity semantics to avoid the subject vs. resource problem present in RDF so that you can expressly indicate which is the subject you're identifying [2].

I've thought more than a few times during this thread that the possibility to employ some of the Topic Maps Reference Model [3] concepts to XML might be very fruitful indeed.

I can't fully participate in the discussion right now (software release looming large at the moment), but I'm happy to try and help explain where I see some potential benefits if anyone's interested in hearing it.

Cheers,

ast

[1] http://www.isotopicmaps.org/
[2] http://www.oasis-open.org/committees/download.php/3050/pubsubj-pt1-1.02-cs.pdf
[3] http://www.isotopicmaps.org/TMRM/TMRM-7.0/tmrm7.pdf

On 6 Dec 2010, at 11:50 AM, Costello, Roger L. wrote:

> Michael Kay wrote:
> 
>> A general problem is that there are many different mechanisms 
>> for describing relationships in XML, and none of them has 
>> particularly expressive semantics.
> 
> Is there some other technology or research that has solved this problem of expressing relationships? If so, could it be mimicked or recast into a proper format for use with XML?
> 
> /Roger
> 
> 
> 
> -----Original Message-----
> From: Michael Kay [mailto:mike@saxonica.com] 
> Sent: Monday, December 06, 2010 6:33 AM
> To: xml-dev@lists.xml.org
> Subject: Re:  Relationships [was RE: James Clark: XML versus the Web]
> 
> 
>> In your messages I see the words "relationship" and "modeling" appearing repeatedly. That seems to be a key concept.
>> 
>> Consider this XML snippet:
>> 
>> <BookStore>
>>     <Book>
>>         ...
>>     </Book>
>>     <Book>
>>         ...
>>     </Book>
>> </BookStore>
>> 
>> That snippet shows a relationship between BookStore and Book; namely, BookStore consists of Books. Is that an example of the kind of relationships you are talking about?
> 
> That's one kind of relationship (what UML would call an aggregation 
> relationship, represented here by hierarchic containment). Other common 
> representations of relationships in XML are via ID/IDREF (that is, 
> primary key / foreign key), and via cross-document URIs. There are other 
> mechanisms, for example XSD and XSLT make heavy use of QNames. XSD also 
> allows relationships to be described on any data type, or on composite 
> key values, using xs:key/xs:keyref. A general problem is that there are 
> many different mechanisms for describing relationships in XML, and none 
> of them has particularly expressive semantics.
> 
> Michael Kay
> Saxonica
> 
> _______________________________________________________________________
> 
> 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@lists.xml.org
> subscribe: xml-dev-subscribe@lists.xml.org
> List archive: http://lists.xml.org/archives/xml-dev/
> List Guidelines: http://www.oasis-open.org/maillists/guidelines.php
> 
> 
> _______________________________________________________________________
> 
> 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@lists.xml.org
> subscribe: xml-dev-subscribe@lists.xml.org
> List archive: http://lists.xml.org/archives/xml-dev/
> List Guidelines: http://www.oasis-open.org/maillists/guidelines.php
> 

--
Andrew S. Townley <ast@atownley.org>
http://atownley.org

• Follow-Ups:
  • RE: Relationships [was RE: James Clark: XML versus theWeb]
    • From: "Costello, Roger L." <costello@mitre.org>

• References:
  • RE: James Clark: XML versus the Web
    • From: "Costello, Roger L." <costello@mitre.org>
  • Re: RE: James Clark: XML versus the Web
    • From: Michael Kay <mike@saxonica.com>
  • RE: RE: James Clark: XML versus the Web
    • From: "Costello, Roger L." <costello@mitre.org>
  • Re: Relationships [was RE: James Clark: XML versus theWeb]
    • From: Michael Kay <mike@saxonica.com>
  • RE: Relationships [was RE: James Clark: XML versus theWeb]
    • From: "Costello, Roger L." <costello@mitre.org>