RE: Survey paper

Hi Cory,

Let's wrap it up for now:

1) Recommendation to avoid XLink because of the risk of confusion (which I have described)

But it may be useful to capture all the types of requirements which xlink-based approaches may have satisfied.

2) Semantic annotations can be applied in O&M and SensorML files in different places

XML blocks which corresponds to classes (element name starting with an upper case letter)
XML blocks which corresponds to property (element name starting with an lower case letter)
special cases e.g. the definition attribute in SWE (which I interpret as a precursor of a semantic annotation)

3) RDFa is interesting but again, it can be used as a mechanism to

help to manage 'model reference to a ontological description'  = Semantic tags pointing to ontological defintions
help to manage 'composition by inclusion of remote resources'  = Semantic link to externally available Linking Open Data definitions

Here is an example of the second case where someone would want to add semantically annotated metadata to data. If you have an observation on a river weir and you know about some heavy irrigation pumping on a nearby which distorts the measure you would then want to add the information about the feature which may influence the way you deal with the observation data. To add the extra information you would use RDFa especially if the information is accessible from an externally managed LOD repository (note: the @rel attribute defined in the RDFa spec can help to link tags placed at different parts of the file). This could also be called a semantic annotation but would serve other purposes.

So it may be a good idea to check these RDFa wiki use cases: http://rdfa.info/wiki/Rdfa-use-cases are comparable to ours

and also to benefit from the experience of some of our group members colleagues on RDFa
Diego Berrueta, Luis Polo, Sergio Fernández,  Jose M. Cantera and Miguel Jiménez "Deliverable D.5.4.1 Semantic extensions for IDEAL" http://forge.morfeo-project.org/wiki_en/images/c/c1/Celtic.2008.D.5.4.1.pdf

4) Let me now try to answer to your requests to fix the paper - Have a look at this

Semantic annotations link data to more expressive ontological representations through model references [5]. As large amounts of sensor data are being made available on the web, semantic descriptions of sensors and sensor data provide a means to make such data discoverable, accessible, and queryable, and semantic annotation of sensor data provides a means of relating the data to the semantic description. Assuming sensor data is encoded in SWE format, there are currently two approaches for annotation: XLink [3] and RDFa [6].

XLink, the XML Linking Language, is an XML markup language for creating hyperlinks in XML documents. The XLink recommendation outlines methods of describing links between resources in XML documents, whether internal or external to the original document. XLink attributes can be added to SensorML and O&M documents (see Figure 2(a)) to provide semantic annotations for the sensor data [29, 38].

RDFa, Resource Description Framework-in-attributes, enables the layering of RDF information on any XHTML or XML document. RDFa provides a set of attributes that can represent semantic metadata within an XML language and a simple mapping to RDF triples. These attributes can be added to SensorML and O&M documents to provide semantic annotations for the sensor data [49, 11], but require additional syntax.

XLink is already used in SWE documents, thus, no syntactic or structural changes are required. This explains the relative success of xlink-based approaches in earlier attempts to add semantic annotations to SWE documents. Recognising which xlink attributes correspond to semantic annotations and which ones correspond to permissible SWE usages could become extermely difficult.

Approaches based on RDFa looks more promising at the level of SWE documents because it would be much easier to process the annotations independently of the rest of the file. Further work is required to check that the introduction of RDFa would not bring major changes for the implementers of the SWE standards and also how to investigate how RDFa-enabled SWE services could be further integrated in RDFa-based HTML mashups.

Finally, SWE also provides a definition attribute that provides a link to a registry definition, which may also link to an ontological description (see Figure 2(b)) [29, 38]. This is another part of the SWE standards where RDFa could also be applicable.

________________________________
From: Cory Henson [mailto:coryhenson@gmail.com]
Sent: Friday, 14 August 2009 1:28 AM
To: Simon Cox
Cc: Lefort, Laurent (ICT Centre, Acton); Compton, Michael (ICT Centre, Acton); public-xg-ssn@w3.org
Subject: Re: Survey paper

I think this is a good discussion that we should continue, but as far as the survey paper is concerned there are two questions. (1) Should this be included as an annotation technique? There seems to be several groups using xlink as a 'model reference to an ontological description,' including also MMI/OOTethis (whether this is right or wrong, and these were discussed in ssn-xg meeting).  And second, (2) does XLink have a predefined translation to RDF (in rough equivalence to RDFa)?  And be aware we only have one page for this topic in the survey.

thanks,
-Cory