Meeting minutes
Minutes review
<kaz> Mar-14
<kaz> (approved)
Testfest
PRs
<kaz> PRs
McCool: I was doing work on the report generator
<kaz> active.csv
<kaz> rrsaggent, draft minutes
McCool: is nodegen part of the node-red implementation
Toumura: no they are separate
McCool: there are also some from TUM
McCool: it would be good to have one readme but sectioned based on other implementations
McCool: we have some TMs under manual
<kaz> 2022.03.Online/TD/manual/TMs
Ege: yes these were written by hand since we do not have a TM aware tool
McCool: we need to think what it means to have a TM implementation
McCool: we need to organize where the results are taken to
Results organization
<kaz> 2022.03.Online/TD/Results
McCool: I have done it for discovery but there it is different since we do the tests elsewhere and copy to the results
McCool: we can have a "results" folder here but we should simply not look inside this for implementations
McCool: should we have two folders for TM and TD outputs
Ege: no need since for the implementation report, there will be a single table
<kaz> 2022.03.Online/Architecture
Ege: why don't we use the definitions for producer, consumer etc. that is agreed in the architecture spec?
McCool: true. Let's do this now and come back to it when architecture TF decides
###<kaz> kaz: the Architecture area has a bit different directory structure from the other spec areas
<kaz> ... will we make it similar to the others, e.g., adding active.csv and Results subdirectory?
McCool: I will create it right now
<kaz> 2022.03.Online/Profiles
McCool: we can look at the tools now
Ege: there are too many places with tools information that is getting duplicated a lot
<sebastiankaebisch> (Sebastian leaves)
<kaz> Lagally's new PR 242 - additional device, converter fixes
Lagally: we should have versions of the tools in this directory per event
Lagally: another question, is the vpn up and running?
Toumura: yes, but so far only myself is connected
McCool: sorry I didnt have time to put my devices online
<kaz> 2022.03.Online/TD/manual.csv
<McCool> wot-thing-description PR 1436 - Fixing Assertive Language
Ege: I have some PRs in the TD repository that should be merged asap
<McCool> wot-thing-description PR 1428 - Cleanup Security, Privacy, and IANA Considerations
<kaz> [Testfest part adjourned; 5-min break]
plugfest
<kaz> Oracle devices
Lagally: Do we have something from ECHONET?
McCool: Maybe from last time
Lagally: Are philips hue devices profile compliant?
Ege: My TDs are but the devices themselves are probably not
McCool: We should pick TDs that should be compliant and make copies and tag them to be profile compliant
… mlagally, please make a PR and we will discuss further
McCool: We can include the results of last event organizations that haven't submitted yet, but I will not copy them over to this event yet
… we should ask to use ECHONET and NHK from last year
Kaz: This is a testfest topic, so we should clarify which part to be checked automatically for the Testfest and which part to be checked by some kind of emulator for the Plugfest part. Maybe we can ask NHK about their emulator software as well.
McCool: Will put it on todo-list, someone can talk to them and ask them
Lagally: Toumura-san, could you ask them?
Plugfest senarios
McCool: I want to have a simple door switch, temp sensor and RGB light with a simple http stack and Node-red integration
David: The Connexxus presentation deadline would be end of April
McCool: It's simple but we need to finish assertions
… I will need to make an MQTT TD but I never did that before
… what we want to do is in conflict with industrial interests
… because there is little incentive to provide interconnected devices
David: There are also the security considerations
McCool: I think for now we should not bother with security and do that in the second round
… so first round LAN with no sec
… M5Stack emulators
… Second round would have PSK security and secure hub for access but node-red could be a problem with PSK
… In the third round: Industrial devices
Lagally: So devices in the first round are telemetry?
Ege: not all
Lagally: so which devices are battery-powered?
McCool: door switches and maybe temp sensors
Lagally: I ask because for telemetry devices this information is important
McCool: devices that can sleep have either a way for waking them up or collecting missed data from somewhere like in MQTT
… so my goal is to do a basic demo that also has discovery
… we want to improve our internal implementations but also discover what industrial devices do
Ege: Most of smart home devices have no security but the gateway is implements the security because you can only access the devices through the gateway
McCool: I tried to buy an industrial iot remote temperature sensor with sleep mode etc.
… shelly is something in between industrial and home devices
McCool: Philips Hue is good for RGB lights
Ege: Philips Hue is HTTP through the gateway but devices use Zigbee so you can communicate directly
<kaz> 2022.03.Online/scenarios.md
Lagally: Are there any other active devices that are outside the safety net of security?
McCool: Maybe next time, maybe you can setup your digital twin instances
… tomorrow testfest/plugfest call will be an hour later
… will fix invites
<kaz> [adjourned]