AW: I am interested in Semantics and Sensors.

Dear Peter Waher,

Dear Simon Cox,

Dear Michael Koster,

Dear Michael Compton,

 

Peter wrote:

>It's clear from these, that the Accept header should control the desired
output to the client.

 

 

What I am interested in, is the design of the solution.

I am not keen on the format of the dataflow.

Indeed I want to learn how the data flow is triggered/pushed or polled:

 

1.)    Does each sensor change it's semantic representation (in a
Triplestore)

In case a new sensor value is available by its "own initiative".
(pushing)

2.)    Or do you have a central SW-entity reading the data actively (each
minute/quarter of an hour)

And alters the semantic representation centrally
(polling)

3.)    Or are is Semantic Web information altered each time 

a request to that sensor is started.
(triggering)

 

Sincerely,

Frank Haferkorn

 

 

-----------------------------------------------------------------------------

RST Industrie Automation GmbH * Carl-Zeiss-Str. 51, D-85521 Ottobrunn 

Tel. +49-89-9616018-20 * Fax +49-89-9616018-10 * http://www.rst-automation.de
<http://www.rst-automation.de> 

 

Geschäftsführer: Dipl.-Ing.(FH) Robert Schachner 

Amtsgericht München: HRB 103 626 * ID-Nr. DE 811 466 035

----------------------------------------------------------------------------

 

Von: Peter Waher [mailto:Peter.Waher@clayster.com] 
Gesendet: Mittwoch, 27. Februar 2013 15:08
An: Simon.Cox@csiro.au; michaeljohnkoster@gmail.com
Cc: Frank Haferkorn; public-ssn-cg@w3.org
Betreff: RE: I am interested in Semantics and Sensors.

 

Hello

 

We follow the w3c recommendations on this topic. It's clear from these, that
the Accept header should control the desired output to the client. There are
various formats recommended by W3C. From semantic resources the most common
might be RDF/XML or Turtle (even though text and HTML representations might
be useful if a human user is investigating the resource). For SPARQL
endpoints there are more options, also depending on the query. It's up to
implementers to follow these recommendations. The formats you mention are not
in these recommendations.

 

Sincerely,

Peter Waher

 

From: Simon.Cox@csiro.au [mailto:Simon.Cox@csiro.au] 
Sent: den 27 februari 2013 03:56
To: michaeljohnkoster@gmail.com; Peter Waher
Cc: F.Haferkorn@RST-Automation.de; public-ssn-cg@w3.org
Subject: RE: I am interested in Semantics and Sensors.

 

Or a netCDF file, or a SQLite file, or GeoTIFF or PNG or ...

Keep it in RDF if you need to or plan to do RDF operations on the data
(primarily inferencing). 

But if not, then RDF is probably a clumsy and expensive way to handle 'raw'
data. 

 

Simon Cox | Research Scientist
CSIRO Land and Water 
PO Box 56, Highett Vic 3190, Australia 
Tel +61 3 9252 6342 <tel:%2B61%203%209252%206342>  | Mob +61 403 302 672
<tel:%2B61%20403%20302%20672> 
simon.cox@csiro.au | http://csiro.au/people/Simon.Cox
<http://www.csiro.au/people/Simon.Cox> 

 

________________________________

From: Michael Koster [michaeljohnkoster@gmail.com]
Sent: Wednesday, 27 February 2013 2:29 AM
To: Peter Waher
Cc: Frank Haferkorn; public-ssn-cg@w3.org
Subject: Re: I am interested in Semantics and Sensors.

Hello Peter, 

 

Do you store the sensor data values in RDF or do you point to a JSON or XML
object?  

 

Thanks,

 

Michael

 

On Feb 26, 2013, at 6:55 AM, Peter Waher <Peter.Waher@clayster.com> wrote:






Hello Frank

 

In our system (running on servers or plug computers, gateways, etc.),
semantic data for all connected sensors is generated on the fly (on demand).
It has a built in SPARQL engine which can process a variety of RDF triple
sources. One such type of triple source is an on demand converter of sensor
information to semantic information. As we control the SPARQL engine
ourselves, we don't have the need to create a huge triple store for all
semantic data necessary to describe the sensor network.

 

Sincerely,

Peter Waher

 

From: Frank Haferkorn [mailto:F.Haferkorn@RST-Automation.de] 
Sent: den 26 februari 2013 07:16
To: public-ssn-cg@w3.org
Cc: Peter Waher
Subject: AW: I am interested in Semantics and Sensors.

 

Dear SSN team,

 

I was able to browse

 

http://www.w3.org/2005/Incubator/ssn/ssnx/ssn
<http://www.w3.org/2005/Incubator/ssn/ssnx/ssn> 

 

and got an impression of the SSN ontology.

I learned about the description of sensor data and other interesting items
like observation etc. .

 

But I still have a big Question:

HOW IS THE REAL ACCESS of the Sensor's Data implemented.

In my Eyes there are only three possible ways:

·         Pushing - the sensor is updating the related semantics itself,

·         Polling - there is some software entity that is polling the and
creating the semantic data from the sensor's output?

·         Triggering - The semantic data is generated on the fly at the
moment of the request for the data.

 

Which option is used in real SSN systems?

 

Yours,

  Frank Haferkorn

 

 

-----------------------------------------------------------------------------

RST Industrie Automation GmbH * Carl-Zeiss-Str. 51, D-85521 Ottobrunn

Tel. +49-89-9616018-20 * Fax +49-89-9616018-10 * http://www.rst-automation.de
<http://www.rst-automation.de> 

 

Geschäftsführer: Dipl.-Ing.(FH) Robert Schachner

Amtsgericht München: HRB 103 626 * ID-Nr. DE 811 466 035

----------------------------------------------------------------------------