RE: tidying ssn -- are you ok with?

?  Intuitively, I would vote for b (sosa/ssn:example) if this would also imply that we declare it an annotation property. Which one is your favorite? A?

I prefer A because skos:example is already known  and  maybe there are tools that work with it already. And besides it is already implemented in sosa.  Only ssn-specific terms are ever going to understand ssn:example. And I know of no other term that could be used for option(c)  -- hold that thought - I think  I recall that schema.org has something?, in which case that would be my preferred option.


> There is one more interesting issue to consider to make sure we arrive at a consistent way of documenting both ontologies, namely on the semantics of 'example'.


I totally get that issue you raise  around the unclear semantics of example  (her is another one ssn:Platform  ""Relation between a Platform and any Systems (e.g., Sensors) that are attached to the Platform."-) -- but I really don't think it is worth wasting any energy on. Whoever writes an example can look at others and make up their own mind. No amount of discussion is  going to pin down this rather inconsequential subtlety down IMHO.



-Kerry

From: Krzysztof Janowicz [mailto:janowicz@ucsb.edu]
Sent: Sunday, 5 February 2017 3:52 PM
To: Kerry Taylor <kerry.taylor@anu.edu.au>; SDW WG Public List <public-sdw-wg@w3.org>
Subject: Re: tidying ssn -- are you ok with?

Hi Kerry,


If it is to go  on the meeting agenda then some thought beforehand would be helpful.

Agreed. Some thoughts are listed below to explain why this should be on next week's agenda.


I can't recall whether sosa currently imports skos - I think that has changed a couple of times (but don't quote me on that).

Yes, I also think we changed back and forth and this is one of the reason to have a discussion next week.


I am guessing that your objection relates to skos not being in owl DL. But I am only guessing. I am also assuming that we would want to see the same treatment in ssn and sosa.

I am not super concerned about this, but I agree that it is a point worth considering. For me, the key issue is rather wether we are just using skos:example or whether we plan to use more of SKOS in the future. This is another point I would like to discuss during the meeting.


Options I can think of:
(a) use skos:example and declare it an owl  annotation property (and this will work for any other skos property too). Also don't import skos.
(b) make up our own --e.g   ssn:example
(c) find another one that people use for this purpose
(d) give up - retract everything and just bury examples inside rdfs:comment.

Frankly speaking Kerry, I think these are all good and usable approaches. All of them have advantages and disadvantages but non of them strikes me as a no-go in any sense. Intuitively, I would vote for b (sosa/ssn:example) if this would also imply that we declare it an annotation property. Which one is your favorite? A? For exactly these reasons I hope we can discuss this during our next telco. I also hope that this should be something we can all agree on within 20min.

There is one more interesting issue to consider to make sure we arrive at a consistent way of documenting both ontologies, namely on the semantics of 'example'. Consider the following example:


ssn:Platform

An Entity to which other Entities can be attached - particuarly Sensors and other Platforms. For example, a post might act as the Platform, a bouy might act as a Platform, or a fish might act as a Platform for an attached sensor.

I would assume "For example, a post might act as the Platform, a bouy might act as a Platform, or a fish might act as a Platform for an attached sensor." is part of the example and will be removed from the comment, right?

But what about "- particuarly Sensors and other Platforms". This could count as example as well.

If so, we are left with "An Entity to which other Entities can be attached."

Do you see my point?

Cheers,
Jano




On 02/04/2017 07:28 PM, Kerry Taylor wrote:
Jano,
Could you explain the problem   as you see it beforehand please? If it is to go  on the meeting agenda then some thought beforehand would be helpful.

I note that this has already been done in sosa --- and was Simon's  response to an earlier  ssn meeting discussion  about separating examples  from descriptions (rdfs: comment) in a different property. - for visibility, retrievability, and perhaps even an expectation that tools might find it helpful to treat examples differently from descriptions. I can't recall whether sosa currently imports skos - I think that has changed a couple of times (but don't quote me on that).

I am guessing that your objection relates to skos not being in owl DL. But I am only guessing. I am also assuming that we would want to see the same treatment in ssn and sosa.

Options I can think of:
(a) use skos:example and declare it an owl  annotation property (and this will work for any other skos property too). Also don't import skos.
(b) make up our own --e.g   ssn:example
(c) find another one that people use for this purpose
(d) give up - retract everything and just bury examples inside rdfs:comment.

-Kerry

From: Krzysztof Janowicz [mailto:janowicz@ucsb.edu]
Sent: Saturday, 4 February 2017 4:31 PM
To: Kerry Taylor <kerry.taylor@anu.edu.au><mailto:kerry.taylor@anu.edu.au>; SDW WG Public List <public-sdw-wg@w3.org><mailto:public-sdw-wg@w3.org>
Subject: Re: tidying ssn -- are you ok with?

Hi Kerry,

I think it would be great if we could discuss this in the group meeting next week. I would like to understand our motivation a bit better as well as some decisions that we are taking e.g., using skos:example without importing skos.

Have a nice weekend
Jano


On 02/03/2017 09:15 PM, Kerry Taylor wrote:
I'd like to follow the approach Simon used in sosa (as we discussed in a meeting last year, I think) to separate examples from descriptive comments in the ontology using skos:example.

Are you ok with me doing the same in ssn? I don't  plan to change the content substantively (although I might reword an example a little if it seems a bit too hard to follow e.g. too brief). And I'm not going to add amore examples at this point --- just move the ones already there.

I will not import skos.

Btw- I think this means specgen that we are currently using for the spec doco will no longer be able to extract the example - nor for sosa .
-Kerry





--

Krzysztof Janowicz



Geography Department, University of California, Santa Barbara

4830 Ellison Hall, Santa Barbara, CA 93106-4060



Email: jano@geog.ucsb.edu<mailto:jano@geog.ucsb.edu>

Webpage: http://geog.ucsb.edu/~jano/<http://geog.ucsb.edu/%7Ejano/>

Semantic Web Journal: http://www.semantic-web-journal.net




--

Krzysztof Janowicz



Geography Department, University of California, Santa Barbara

4830 Ellison Hall, Santa Barbara, CA 93106-4060



Email: jano@geog.ucsb.edu<mailto:jano@geog.ucsb.edu>

Webpage: http://geog.ucsb.edu/~jano/

Semantic Web Journal: http://www.semantic-web-journal.net