<scribe> scribenick: mjk_
McCool: is there a volunteer to organize
things around scenarios and goals?
... any other agenda items?
McCool: softether vpn was a lot of work to set up, do we need it or should we work on something else?
Ege: depends on each company, whether
they need a LAN
... we should ask every participant
McCool: we could set up a proxy for those
that need a LAN
... local discovery would require a LAN
... any objection to publish?
<scribe> ...done
<McCool> doodle
<kaz> doodle
McCool: likely time slot is same as this
meeting, 0900 US Eastern
... but Kaz has a conflict, maybe an hour earlier
... we need the doodle poll result
<inserted> March 2021 Plugfest page
McCool: we can cooperate on editing the
md document to brainstorm scenarios and goals
... interest groups could work together
... TUM tutorials could be useful but need to be worked through
Ege: it can be done anytime, we could focus the plugfest on devices and things that need to be present
McCool: any other major themes?
Ege: geolocation has issues that need
to be worked out
... we could get more people to use geolocation
McCool: we would need to define the
information model in advance
... what applications, automatic integration could be worked
on
... also we could work on discovery filters, but that is further
along
... we could try some json-path filter constructions
... a proposed spec is needed
<Ege> neighbour is drilling, may not speak :/
McCool: what about the node-RED discovery implementation?
Toumura: will try to work on the node-RED discovery
McCool: will talk with Farshid to make
sure we have a running directory service
... we can integrate node-RED discovery with the directory
APIs
... need some work examples, maybe as a tutorial
... collect examples into a document
... same for geospatial queries
... integrate keyword/semantic + geospatial filters and populate node-RED with the discovery result
... thoughts about general goals?
... refine and test new features
Ege: thingmodel integration from
multiple sources
... two independent implementations
McCool: thingmodel in the context of
discovery?
... test mqtt binding - using node-RED
Koster: we could make OPC UA a topic and ask the experts to bring their integration cases
McCool: any other goals?
... address gaps in the specification
Kaz: do we want to distinguish
between thingmodel and partial TD/fragment?
... we may need to identify concrete algorithms for handling these
cases
McCool: for example, how do we validate a
partial TD?
... maybe a set of shapes
Kaz: maybe we can define a set of
defaults or examples
... can we provide a draft feature list for automatic testing for
the Plugfest in March?
McCool: we need to do pre-work for the
implementation report associated with our next CR
... new specs like Discovery will need test cases
... assertions for profiles
... create a tools directory
McCool: no meeting next week
... aob?
... adjourn