- From: Kazuyuki Ashimura <ashimura@w3.org>
- Date: Sat, 26 May 2018 02:07:34 +0900
- To: Public Web of Things IG <public-wot-ig@w3.org>
available at:
https://www.w3.org/2018/05/25-wot-minutes.html
also as text below.
Thanks,
Kazuyuki
---
[1]W3C
[1] http://www.w3.org/
- DRAFT -
WoT IG - TF-LD
25 May 2018
[2]Agenda
[2] https://www.w3.org/WoT/IG/wiki/IG_Linked_Data_and_Semantic_Processing_WebConf#Agenda
Attendees
Present
Kaz_Ashimura, Darko_Anicic, Michael_Koster,
Victor_Chapenay, Aparna_Thuluva
Regrets
Chair
Darko
Scribe
kaz
Contents
* [3]Topics
1. [4]Agenda
2. [5]TD template generation
* [6]Summary of Action Items
* [7]Summary of Resolutions
__________________________________________________________
<scribe> scribenick: kaz
Agenda
Darko: would like to talk about preparation for the next
plugfest
... from semantic viewpoint
... second, maybe we could also discuss iot.schema
... how to improve our presentation there, etc.
... any additional points?
Koster: alignment between iot.schema and TD
... interaction pattern
... some guidance there
... data amount restriction, etc.
... float/integer
... might be a good topic for discussion
... not totally related to semantics, though
... right now we have substantial difference property
interaction and data element property
... might be useful
Darko: ok
TD template generation
<DarkoAnicic> TD Templates generated based on iot.schema.org
specifications:
[8]http://www.w3.org/2018/03/wot-f2f/slides/2018-03-26-W3C-WoT-
F2F-Prague-iotschema-v2.pdf
[8] http://www.w3.org/2018/03/wot-f2f/slides/2018-03-26-W3C-WoT-F2F-Prague-iotschema-v2.pdf
Darko: shows slides
... how to proceed with this?
Kaz: you used this mechanism for the prague plugfest?
Darko: yes
Kaz: in that case, you should describe this mechanism within
the siemens-report.md as well
Darko: ah, ok
Koster: this is good to explain
... how you wanted to communicate to generate TD
... people are asking about how to use it
... good to explain the design flow
Darko: it was kind of idea
... describe capability
... automated generation of semantically annotated TD
... 1. select capabilities, 2. ...
... [Example: Level Capability]
... shows example
... [SHACL Shape for Level Capability]
... Shapes here
... Fujitsu air conditioner and Panasonic air conditioner
... maybe have different numbers
Koster: right
... Shapes sets actual constraints here
Darko: interoperable interaction between Fujitsu and Panasonic
... for different air conditioners
... [Generated Thing Description]
... implementation of generator
... there is one point
... in iot.schema, we don't have Shapes
... still not really confirmed
... need iot.schema to be adapted
... generated TD is new simplified one
... some of them ought to be validated
... with this tool
... validate both TD and TD annotated by iot.schema
Koster: looks like kind of design flow
... Shape language looks good
... ability to add statements for max, min, etc.
... describes pattern
... also ability for relations to classes
Darko: quite important feature
Koster: if we're to adapt Shapes to iot.schema
... Shapes available for number constraints
... use Shapes to build TD
... which capability is used for interaction
... define data constraints
... and generate TD
... Shapes has constraint for JSON Schema as well
... min/max, number, etc.
... Shapes looks like a missing piece we don't have
... but people must learn new language?
... it's not difficult, though
... how to deal with that?
... need some education/tutorial?
Darko: depends on software...
... it is a language to specify constraints
Koster: Shapes are easy to understand
... contributers of IoT need to understand how constraints work
... but we don't have to have Shapes
... which capability to use
... seems to solve the problems
... maybe we should work with Dan
... to me it would make a lot of sense
... already adapted by industry
... already a W3C REC?
Darko: yes
... SHACL is a REC
... problem of somebody's asking about how iot.schema works
... ok
... let's think about this
Koster: we could do is working on this a bit more
... we have time to work on examples
... we have some example now
... to construct an instance of TD
... this is how the tools work
... TD builder, etc.
... important thing about Shapes
... why we need the choice
... scripting language for RDF
... higher level things
... good match for RDF
Darko: btw, these slides are already online
... can think about how to extend the example
... maybe we can get comments from Ben as well
... the question is how to solve the problem
... some sort of misunderstanding
... the concept is not clear enough maybe
Koster: good to get people's impression
... just jumped in and started to see schema.org
... first thing is see browsers work properly
... real questions are how to deal with constraints
... how do I use it
... semantic annotation in general
... maybe can simply use data tag, etc.
... we need to make the discussion self-explanatory
... improving first impression and explain how it works
... what else?
Darko: that's pretty much
... for the software part, we don't know
Koster: we need commitment for iot.schema.org
... should talk with Dan again
... we're kind of ready for moving forward
... we need to immediately work on browsers
... the other question is
... organizing ontology
... direction seems ok
... we're just adding interaction for feature of interest
... how to go forward?
... now we need Shapes for adding constraints
... we need to get consensus for our direction
... this part needs to be figured out
Kaz: given this is an IG call, how about inviting Ben and/or
Dan to this call?
Koster: would be great
Darko: yes
... we should prepare some agenda to fit that discussion
... my concern is
... we need some new update for the next meeting
... having some progress
Koster: the feedback we got is...
... more definitions
... more ease of use
... hoping new features of interest
... that's one of the missing pieces
... would my favor to add that
... and ease of use also
Darko: right
Koster: could spend some time
... not really we can actually combine Shapes, though
... 6 weeks to the next meeting?
Darko: yeah
... for me, it's easy to fix the tool
Koster: yeah
... to fit with Shapes
... if we added all people needed...
... query in resource
... working with annotations
... also Fujitsu and Panasonic expressed interest in using more
semantic annotations
Darko: ok
... feature of interest should be done
Koster: for automotive use cases, we want to develop feature of
interest
... Soumya and Benjamin may want it
... Ari was asking for some guidance
... introductory explanatory materials
... for someone who doesn't really know about RDF
... would talk with Ari as well
Darko: can we contact him and see what is needed
Koster: great approach
... can send him an email CCing you
Darko: please do so
... btw, regarding the testing work
... is there any discussion about online tooling?
Koster: some discussion on F-Interop
... but seemed more related to plugfest
... also would see ETSI plugfest
... also relationship between WISHI and WoT
Darko: wondering about the online plugfest framework
Koster: could be a big improvement for plugfest preparation
... having "online plugfest" doesn't exclude ordinary plugfests
Darko: should continue to discuss it during the plugfest calls
... also iot.schema call about TD generator
[adjourned]
Summary of Action Items
Summary of Resolutions
[End of minutes]
__________________________________________________________
Minutes formatted by David Booth's [9]scribe.perl version
1.152 ([10]CVS log)
$Date: 2018/05/25 17:04:16 $
[9] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/scribedoc.htm
[10] http://dev.w3.org/cvsweb/2002/scribe/
Received on Friday, 25 May 2018 17:08:44 UTC