Re: Comment on the SOSA ontology

Hello Catherine,
I am one of the primary developers of SWEET over at the ESIP Federation
[0].
It's really encouraging to see your work and would encourage you to get in
touch with us on Github or via our mailing lists [1] if you have any
queries using SWEET.
Thank you
Lewis

[0] https://github.com/ESIPFed/sweet
[1] http://lists.esipfed.org/mailman/listinfo/esip-semanticweb

On Fri, Nov 10, 2017 at 2:43 AM, Catherine Roussey <
catherine.roussey@irstea.fr> wrote:

> Dear all
>
> Thanks to the work of Stephan Bernard
> we have created a first example of our weather archive using SOSA ontology.
> I attached some drawing that describe our usage of sosa.
>
> Has you may notice we reused directly the skos vocabulary that describe
> weather property
> NERC Vocabulary Server version 2.0 (NVS2.0).
> We also reuse the QUDT vocabulary.
>
> You will notice that we do not follow your weather example 25
> based on sample.
> The thing is that we find usefull the creation of the sample based on the
> weather station location e.g.
>
> <Air?lat=45.75&long=4.85>
>
> But this sample will be the sample of several Feature of Interest: wind,
> precipitation, atmosphere etc...
> Thus we should create
>
> <Wind?lat=45.75&long=4.85>, <Precipitation?lat=45.75&long=4.85>, <Atmosphere?lat=45.75&long=4.85>
>
> We prefere to reuse some existing vocabulary like the sweet ontology that
> propose wind, precipitation and atmosphere URI.
> Moreover we should recreate the observed property based on the sample URI.
>
> <Air?lat=45.75&long=4.85#temperature>
> Would become <Precipitation?lat=45.75&long=4.85#rainfall_amount><Atmosphere?lat=45.75&long=4.85#temperature>
>
>
>
> We obviously could create it automatically but we do not see the utility
> of the creation of these new instances.
>
> We do not describe anyThing about FoI and Sample, because we expect that
> the skos vocabulary NERC is sufficient to express the FoI.
> If you thing that our hypothesis are false we could modify it.
>
> We reuse also the inXSDDateTime Property which is now deprecated. We will
> replace it by the new property in inXSDDateTime.
>
> I give you the link to the snorql interface if you want to play with our
> weather archive
> http://ontology.irstea.fr/weather2017/snorql/.
> We already now that we have a problem with time filter. We are looking to
> solve this issue.
>
> Best Regards
> Catherine Roussey
>
> --
> Catherine ROUSSEY
> Irstea Clermont Ferrand
> Campus des Cézeaux
> 9 avenue Blaise Pascal
> CS 200 85
> 63178 Aubière
> tel: 33 (0)4 73 44 06 88
> "Imperfection is beauty, madness is genious and it's better to be absolutely ridiculous than absolutely boring" Maryline Monroe
>
>


-- 
http://home.apache.org/~lewismc/
@hectorMcSpector
http://www.linkedin.com/in/lmcgibbney