<kaz> scribe: Darko

<kaz> scribenick: DarkoAnicic

Sebastian is introducing the agenda for today

Maria: Minutes from 26th April should be corrected: I appear on the list of participants but did not attend the meeting.

<inserted> [ minutes from Apr. 26 TD model call have been updated: https://www.w3.org/2017/04/26-wot-td-minutes.html]

Open Action Items

Issue 309

<inserted> Issue 309

<Victor> (Here is the issue in the TD Github repo: https://github.com/w3c/wot-thing-description/issues/5)

Issue: Adding Semantic Annotations to JSON Schema

Victor: we introduced the issue, use case, someone already gave a proposal
... the example shows the issue when within the schema the structure of two semantically described items cannot be differentiated.
... thus we need the semantic annotation to to JSON Schema

McCool: the context in JSON cannot be added in the same fashion as in JSON-LD
... we should have a TD schema, describing the class, and the TD, describing an instance thereof

Daniel: we are talking not only about TD but also about types of elements being exchanged

McCool: talking about type systems of Things and type of data
... do we have one schema for all kind of Things?

Victor: everybody can add or extend the TD vocabulary if we rely on Semantic Web technology

Sebastian: let us keep the JSON Schema as it is. We want to add semantic annotations only.
... we define Events, Properties, Actions in TD and rely on JSON Schema for Data Types

McCool: we have to define vocabulary in advance to enable contained IoT device to handle them. But for different domains/devices we need multiple schema - this complicates things.

Victor: in addition two devices may use different vocabularies even for the same domain.
... we should not enforce one schema/one structure only, need to be flexible.

McCool: I suggest we stich to JSON as the mapping to other formats can be made.

Sebastian: this will be one of the major topics in the upcoming F2F meeting in Osaka. Please follow and contribute.

<scribe> ACTION: kaz to create an issue about two-way approach: device dependent and non-depend path [recorded in http://www.w3.org/2017/05/05-wot-td-minutes.html#action01]

<trackbot> Created ACTION-104 - Create an issue about two-way approach: device dependent and non-depend path [on Kazuyuki Ashimura - due 2017-05-12].

<kaz> [ kaz mentions possible need for 2-layered model for TD: device-dependent layer and device-independent layer ]

Issue 3

Issue 3: Namespaces used in the TD / WoT ontology

McCool: for the ns we need the year/month etc.

Sebastian: Dave already registered the ns without the date

kaz: another option could be using the shortname for TD, e.g., "wot-thing-description" => "https://www.w3.org/ns/wot-thing-description"

<MariaPoveda> It is not an exception see http://www.w3.org/ns/prov#.

McCool: the version of the vocabulary is needed to avoid errors

Maria: are you referring to the version of the vocabulary of documents that will be published
... the name space without date points always to the last (actual) vocabulary, wheres the previous versions are pointed with links with dates

<McCool> The Namespace guideline at: https://www.w3.org/2005/07/13-nsuri suggests: http://www.w3.org/ns/ssss, http://www.w3.org/YYYY/MM/ssss and http://www.w3.org/YYYY/ssss

kaz: we're already using wot-thing-description for the Git and we're planning to publish a FPWD for Thing Descirption with that shortname. in that case, we can simply reuse that shortname for the namespace. probably that would be less confusing.

Sebastian: the issue is not that urgent but we should resolve it in communication with Dave

Overview about breakout sessions in Osaka

Sebastian: topics for break out sessions focus on topics such as: JSON Schema discussion, TD model
... we have 2 versions of the TD model: one more for web developers and another, more suited for Semantic Web community. We may have both versions.
... JSON Schema is one approach. Dave proposed a proposal with plain JSON.
... please comment the issues/topics/proposals
... we should set up the remote participation for people not being able to join the Osaka meeting

Maria: if we do not join the Osaka meeting, will we discuss the output of the F2F meeting in web meetings?

Sebastian: I will take your input and present during the F2F meeting. As said we may have the both models in place

Maria: I am concerned about the implementation if we had the both models.

kaz: Maria, can you one of the sessions during the meeting if it's in Europe-friendly time?

Maria: I cannot make it on Mon or Tue.

<MariaPoveda> ok, I can't make Monday or Tuesday, ok from Wednesday

Broken links in TD draft

https://w3c.github.io/wot-thing-description/

Sebastian: asking for volunteers to look at broken link and to repair the document

Dape: I don't have write access, but have already identified the broken links

<kaz> [ kaz can help you for that ]

discussion about Script management vs Thing management

<inserted> Issue 22

Zoltan presents the topic.

Zoltan: we need the support form the TF-TD for semantic description for the Thing management

McCool: can this be discussed on the next scripting call?

Zoltan: yes, but we need people from TF-TD to be present there
... I can create an issue and describe the use case for it

<kaz> [ Scripting call is held on Monday at 1pm in Europe, 4am in US Pacific, 8pm in Japan ]

McCool: do you have pseudo code for the functionality?

Sebastian: I will try to be present on Monday. Is it planned for the Osaka meeting?

AOB

kaz: please request the time, e.g., 30mins or 1hour, for the Osaka meeting on the f2f wiki so we can plan, e.g., for Europe participants.

<kaz> [ also you can add "I'm interested in this topic." to your preferred topics for remote connection :) ]

Meeting closed

<kaz> [ adjourned ]