30 Jun 2010


See also: IRC log


[IPcaller], Payam, laurent_lefort_cs, +1.937.775.aaaa
see, http://www.w3.org/2002/09/wbs/43337/SSN-XG-summer/results


<laurent_lefort_cs> Agenda:

<laurent_lefort_cs> Previous: http://lists.w3.org/Archives/Public/public-xg-ssn/2010Jun/0030.html

Hi Lauernt, Hi Andriy and welcome!

<Andriy> Hi!

It seems fine; I just got connected 10 minutes ago. I hope it would be stable

<Andriy> Trying to join the telco

<laurent_lefort_cs> Teleconferences tricks for Andriy http://www.w3.org/2006/07/SWD/Scribing.html

<laurent_lefort_cs> and http://www.w3.org/2005/rules/wg/wiki/TeleconEtiquette

It's fine

Welcome to Andriy

Adnriy- open University- Andriy work in smart prodcuts project

<laurent_lefort_cs> Andriy Nikolov http://people.kmi.open.ac.uk/andriy/

Andriy- the use cases from smart project could be interesting to the group

- smart objects project-

Operating and survival conditions

<laurent_lefort_cs> Payam: Looked at the model built by Michael

<laurent_lefort_cs> ... There are some properties which are not defined

<laurent_lefort_cs> ... There are more properties which can be defined

<laurent_lefort_cs> ... e.g. the role of a sensor as a router in a WSN

<laurent_lefort_cs> Laurent: commenting the refs on http://www.w3.org/2005/Incubator/ssn/wiki/Sensor_types#System_and_Product_ontologies

<cory> ?q


<laurent_lefort_cs> Discussion on things which are missing in the current model

<laurent_lefort_cs> ... in relation to the communication aspect of the role

<laurent_lefort_cs> Cory: question on clarifying how routers fits in the picture

<laurent_lefort_cs> Laurent asking Andriy if the communication aspect is important for is use case

<laurent_lefort_cs> Andriy: not really, we have use cases on sensors monitoring a kitchen

<laurent_lefort_cs> Laurent: The Phenonet example has particular device acting as gateways

Semantic Markup

<laurent_lefort_cs> scribenick: Payam

Laurent: use cases that describe different attributes of sensors;

semantic annotation allows to do more and add more metadata to these XML based description

Laurent: the question is what we can do after annotation that we couldn't do before?
... we change the description vocabluary to RDF/OWL and in overall semantic web standard descriptions
... it alos helps to discovey of resources
... if you only add annotations as they are defined now, you can describe all you want?>
... we ned to compare what is feasible with the current OGC, leaf nodes?

what are those in OGC descriptions for example in SensorML?

- some of these references are made in Laurent's presetnation on annotation (see on wiki)

Laurent: should we also recommend annotation on nodes that are not leaf nodes?

Cory: what's the benefit of annotating of those high-level concepts?

Laurent: when you use descriptions, in some cases one may need to access/anotate to high level attributes


<cory> +q

Payam: what we do is defining a common vocabluary

and then define how the annotations can be made

but what is going to be annotated in not in the scope of our work

Payam: we define vocab, common methods to define associations to our vocab; but the sources that could be annotated would be users' choice


Laurent: xlink could impose some limitations for annotations
... by adding annotations to some attributes of existing descriptions you can add more metadata and have more use cases
... in the report we should highlight how our work is different from OGC

Payam: should we also add something related to RDFa and compare it with xLink?


Laurent: we don't make recommendations; we only explore possible solutions

for annotation report

Laurent: can we have only one link to model/reference to annotate the resources?
... 1st use case; having annotation where ever we can

2nd option: having link from user definition to another ontology(?)- two step link

two steps links


Laurent: two step annotation enables to associate the value for the attributes that we have define in our ontology to domain knowledge

this means having common vocab. for domain knowledge then people can interpret instances created using our ontology

payam: this is interesting; however we have to be careful that domain knowldge specification does not fit into the (current) scope of our work

Laurent: e.g. the domain knowledge and 2nd steps of annotation for example could use Linked Data

Linked data approach


<laurent_lefort_cs> Laurent: summary of the discussion with Cory

<laurent_lefort_cs> ... Explain (report, examples) how our proposed usage of XLink extends what can currently be done

<laurent_lefort_cs> ... The motivation is then to overcome the limitations of existing standards

<laurent_lefort_cs> ... when they use definitions which are too generic

<laurent_lefort_cs> ...

<laurent_lefort_cs> ... Clarify the semantics of modelReference:

<laurent_lefort_cs> ... normal use (SAWSDL) is to link a schema element to (generally) a class

<laurent_lefort_cs> ... second use (in some examples published earlier that we need to revisit) is to

<laurent_lefort_cs> ... add links to individuals (an alternative method to include data from remote sources)

<laurent_lefort_cs> ...

<laurent_lefort_cs> ... Check the data ingestion use case defined by SAPIENCE where

<laurent_lefort_cs> ... there is a second type of link domain reference to link

<laurent_lefort_cs> ... "local" vocabulary items to "reference" ontology definitions

<laurent_lefort_cs> ...

<laurent_lefort_cs> ... Motivation for the 2nd case is different: it is to automate as best we can

<laurent_lefort_cs> ... the translation from data managed by external parties into a common format

<laurent_lefort_cs> ... and also migrate the extra information present in the original format

<laurent_lefort_cs> ... into an exploitable annotation

<laurent_lefort_cs> ....

<laurent_lefort_cs> Feedback from Payam and Cory: more work to be done to

<laurent_lefort_cs> ... check if we need different annotation mechanisms for modelReference and domainReference

<laurent_lefort_cs> ... or if this is something which is beyond our reach

<laurent_lefort_cs> ... The trade-offs for the example mimicking what Sapience tries to do are: to highlight the value of the annotation mechanism we propose without stretching the specification too far

<laurent_lefort_cs> ... Need

<laurent_lefort_cs> ... (to be continued)

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2010/06/30 22:08:22 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.135  of Date: 2009/03/02 03:52:20  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/proposed used/proposed usage/
Succeeded: s/... The trade-offs for the example mimicking what Sapience tries to do are: /... Need /
Found ScribeNick: Payam
Inferring Scribes: Payam
Default Present: [IPcaller], Payam, laurent_lefort_cs, +1.937.775.aaaa
Present: [IPcaller] Payam laurent_lefort_cs +1.937.775.aaaa
Regrets: see http://www.w3.org/2002/09/wbs/43337/SSN-XG-summer/results
Agenda: http://lists.w3.org/Archives/Public/public-xg-ssn/2010Jun/0033.html
Got date from IRC log name: 30 Jun 2010
Guessing minutes URL: http://www.w3.org/2010/06/30-ssn-minutes.html
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.

[End of scribe.perl diagnostic output]