Re: ssn :complaint

Hi Kerry,

We are constantly monitoring sosa.ttl and it is in a stable state for a while already. Today in the phone call we have been made aware of some inconsistencies with capitalisation in the rdfs:comments, an issue that I have assigned myself an action item to. I don’t know what Wiki outline you are referring to, but several wiki pages are naturally out of date based on decisions we have made. And we are not keeping all wiki examples up to date.

Sampling was in SOSA from the very beginning, thus the name “Sensor, Observation, Sample, and Actuator (SOSA) Core Ontology”. Addressing issue-92 https://www.w3.org/2015/spatial/track/issues/92 has been on the agenda of our SSN subgroup meetings several times and finally we managed to get to it in our meeting on the 14th of March (https://www.w3.org/2017/03/14-sdwssn-minutes). We reached consensus that Sampling should be in the core (Raul changed his vote to 0) and Rob’s vote was a -1 on including it in SSN only. The vote on including it in SOSA was then accepted. It was not a rushed decision, we had a solid discussion and there are use cases in our UCR (5.16 and 5.38) that are related to Sampling. I disagree that Sampling is big (it has 3 classes and 3 properties plus their inverse) which have been on the wiki for months and I don’t know why you say it is buggy. We will need to have implementation evidence, as we will need to have for all of SOSA terms. Several people indicated that they can provide example implementations for Sampling. If we can’t it has to go to a non-normative part of the document or should be dropped anyway.

Cheers,
Armin


From: Kerry Taylor <kerry.taylor@anu.edu.au>
Date: Wednesday, 5 April 2017 at 3:35 pm
To: "public-sdw-wg@w3.org" <public-sdw-wg@w3.org>
Subject: ssn :complaint
Resent-From: <public-sdw-wg@w3.org>
Resent-Date: Wednesday, 5 April 2017 at 3:36 pm

Dear SDW,
I had a look at the sosa.ttl in github  today  with a vire to a final review and am concerned about the state of much of it, but most particularly surprised to see sampling, and to see that it has lots of problems. I started to note them, for correction, but there are so many I am certain that they will be ignored. Has anyone actually looked at  what is there, other than me? AFAICT it does not even match the wiki outline.  It also has complex consequences on ssn that have not even been begun to be considered.

I am surprised because, although I missed two  ssn meetings,  a vote to include it was never on the agenda, for either  meeting and indeed I had expressed my objection previously when it was discussed in meetings. It is a very substantial increase in scope (although we do have  maybe one use case, I think), but many other unserved use cases too at this time).  Normally, we do not make decisions on matters when people who had showed interest are not able to be present. Furthermore the minutes  records a -1 that was not mine, and a couple of zeros.  And it very clearly shows how rushed the decision was.

I am  particularly  concerned sampling is both very big, and very new, and buggy. Furthermore, in our rush to complete we have many more critical things to fix then this.

It should be removed.

-Kerry