W3C home > Mailing lists > Public > public-sdw-wg@w3.org > May 2017

Re: Question about SSN [for wide review]

From: Maxime Lefrançois <maxime.lefrancois@emse.fr>
Date: Fri, 05 May 2017 16:52:29 +0000
Message-ID: <CALsPASWHH4giDy-kq-2Lna1Emjj6VB-aayCnTcEFCbQcnR7=eQ@mail.gmail.com>
To: janowicz@ucsb.edu, SDW WG Public List <public-sdw-wg@w3.org>, Armin Haller <armin.haller@anu.edu.au>

I see two comments in this mail.

The first one is clear: How does ssn:Property map to QUDT ?

[...] He and I are having a disagreement about ssn:Property mapping to
> QUDT. The rdfs:comment says that ssn:Property is intrinsic to an entity,
> which would suggest that it is most equivalent to qudt:QuantityKind.

I don't agree with Dr. Jack Hodges here, because ssn:Property is "intrinsic
to an entity", then my optinion is that it is aligned to qudt:Quantity (and
not qudt:QuantityKind, which is generic).

Depending on the final decision we make about the status of ssn:Property,
I'll mention this alignment in the note I have to write for ACTION-318 -

The second comment is about the specification of ssn:Condition, which
becomes misleading in the new context where SOSA/SSN are made independent
of DUL.

But if I look at the rdfs:comment for ssn:Condition (which subclasses
> ssn:Property), and published examples, then I see that ssn:Condition
> relates a property, a value, and a unit, which seems more like a
> qudt:QuantityValue. A subclass that can represent value would suggest
> that the parent can represent value, so I am confused. We can go either
> way, but if the intention of the ssn:Property definition is one way or the
> other, then of course we would want to honor that intention.

Please first note that the definition of ssn:Condition has been simplified,
and is a bit more fuzzy. It is now: "Used to specify ranges for qualities
that act as conditions on a system/sensor's operation. For example, wind
speed of 10-60m/s may be used as the condition on a SystemProperty, for
example, to state that a sensor has a particular accuracy in that
condition."@en .

I totally agree with the comment here: the definition of ssn:Condition is
incompatible with it being a subClassOf ssn:Property (may ssn:Property be
intrinsic to a feature of interest, or generic and applicable to multiple
feature of interest).

I propose we take this comment into account by deleting the triple

ssn:Condition rdfs:subClassOf ssn:Property .

On the other hand, this doesn't help in understanding how ssn:Condition
should be mapped to QUDT. I am soon to write examples of use for
ssn:Condition (see https://github.com/w3c/sdw/pull/730 ), and I still don't
really know what I am going to propose. Any suggestion for exemplifying
"wind speed of 10-60m/s may be used as the condition on a SystemProperty"
is welcome.

Maxime Lefrançois

> Comments?
> Best regards,
> Dr. Jack Hodges
> [...]
Received on Friday, 5 May 2017 16:53:15 UTC

This archive was generated by hypermail 2.3.1 : Friday, 5 May 2017 16:53:16 UTC