W3C

WoT Security

14 Sep 2020

Agenda

Attendees

Present
Kaz_Ashimura, Clerley_Silveira, Elena_Reshetova, Michael_McCool, Oliver_Pfaff, Tomoaki_Mizushima, Cristiano_Aguzzi, David_Ezell
Regrets
Chair
McCool
Scribe
elena

Contents


<kaz> scribenick: elena

Review the minutes from last meeting

<kaz> Sep-7

McCool: any objections publishing the minutes?

no objections, minutes approved

Agenda updates?

McCool: any updates from anyone?
... we might need two producers and two consumers for implementation to be approved. This can be a problem for Oauth implementations.
... does anyone know about wot-node and oauth?

Cristiano: difference in implementations between producers and consumers can be very minimal for node-wot

McCool: need to bring it up with node-wot, could Cristiano create an issue about this and test cases for node-wot?
... let me do issue creation now

McCool creates a new issue in wot-testing

Cristiano: I am afraid that LinkSmart wont implement consumer side

McCool: we need then another consumer
... node-gen or node-RED might be an option for that
... we need to have two tests per flow

Cristiano: code is not implemented in node-wot, might be a problem

McCool: node-wot also assumes that security configuration is the same, another thing that needs review

<McCool> https://github.com/w3c/wot-testing/issues/51

McCool: we need to review security implementation of node-wot

McCool creates a new issue under wot-security on this

https://github.com/w3c/wot-security/issues/184

McCool: Cristiano, could you walk us through node-wot implementation since you know it well?

Cristiano agrees

McCool: we should also dig into node-gen also
... are we doing something special for plugfest? I have not seen any security focus there
... does anyone have any thoughts on this?
... oauth is something we should do but we dont have enough time for this plugfest. Maybe next plugfest that is in February/March?
... if we want to be safe to get things done in time, we need to finalize test cases by the end of the year

Kaz: a bit off topic but I attended the Singapore Geospatial Week's Smart Cities session this afternoon and some of the presenters mentioned end-to-end security would be important for IoT purposes. so I'm wondering how to deal with end-to-end security in wot.

Oliver: that depends on definition of the ends

McCool: should we have security schemes for object security?

Oliver: we have to double check first how to express object security in order not to redo this in TD

McCool: we don't have any existing issues about object security and how to deal with it
... we need to decide how we support object security

McCool creates a new issue for this

https://github.com/w3c/wot-security/issues/185

Kaz: this issue 185 could include a definition of end-to-end security. right?

McCool: we need to make a list of object security alternatives

McCool adds some initial options to the issue 185

Oliver proposes more schemes that McCool adds to the issue 185

McCool: next let's look into issue tracker

Issue 183

McCool looks into issue https://github.com/w3c/wot-security/issues/183

<kaz> Issue 183

McCool: should we also add monitoring into this issue?

elena: IMO it should go into separate issue

McCool creates a new issue https://github.com/w3c/wot-security/issues/186 on monitoring

<kaz> related issue on IETF MUD

Issue 180

McCool: next issue https://github.com/w3c/wot-security/issues/180

McCool adds some todos to the issue

McCool: should we also be looking into mozilla hub or other hubs?
... what about open Hab?

McCool creates a new issue on OpenHab https://github.com/w3c/wot-security/issues/187

<criis> https://github.com/iobridge/thingspeak

McCool creates another issue on mozilla WebThings gateway https://github.com/w3c/wot-security/issues/188

McCool creates an issue on ThingSpeak https://github.com/w3c/wot-security/issues/189

McCool: we don't have wot integrated in projects like the above
... we need to talk to these groups
... and we need to look into their security architecture to make sure we are compatible

Issue 170

McCool: let's look into issue https://github.com/w3c/wot-security/issues/170
... last time we created issues for follow up work, should we close this issue?
... or do we still have some missing actions?

elena: i don't see anything else from my side

McCool: let's create an issue about trust levels of actors and then we can close the issue 170

McCool creates a new issue https://github.com/w3c/wot-security/issues/190 on this

McCool: any objections to close 170?

no objections, closed

<kaz> [adjourned]

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version (CVS log)
$Date: 2020/09/21 14:12:56 $