W3C

- DRAFT -

WoT Security

18 Jun 2018

Agenda

Attendees

Present
Kaz_Ashimura, Michael_McCool, Michael_Koster, Kazuaki_Nimura, Tomoaki_Mizushima, Barry_Leiba
Regrets
Zoltan, Elena
Chair
McCool
Scribe
kaz

Contents


Agenda

previous minutes

Elena's PR 103

Previous minutes

previous minutes

McCool: goes through the previous minutes
... "to date" should be "to date"
... wondering about the progress on url schema

<mjkoster> URI templates are defined in RFC 6570

Koster: Matthias made some concrete proposal
... very clear about how it works
... variables would be expanded
... we're calling payload schema

McCool: ok
... minutes are OK with the small typo above
... any objections?

(none)

the minutes accepted but "to data" should be "to date"

Elena's PR 103

<McCool> https://github.com/w3c/wot-security/pull/103

Changes

McCool: goes through the changes
... bunch of statements on mitigation
... e.g., access to TD limited to some certain clients
... for privacy
... and privacy consideration referring to coo13
... typo to be fixed
... and big change here
... L2384 => L2399
... some typical things
... threat never changes or changes rarely
... fingerprinting
... persistent tracking
... and TD id changed periodically
... probably TD changes and notification happens
... it's OK with it is though some more description might be better
... would accept this PR itself and polish it later

Koster: there is another submission
... look fine to me

McCool: don't think it's perfect but ok to accept
... and clean up and polish the text later
... OK to merge this?

(no objections)

McCool: merged PR 103
... ok
... now just one PR here

Issues

* Issue 72

issue 72

McCool: adds notes to the issue
... fingerprinting risks now desicussed in text included in PR 103
... @jasonanovak, do these changes sufficiently address the issue?

* Issue 70

issue 70

McCool: adds notes
... discussed in PR 103.
... currently for various reasons the WoT TD actually requires unique id.
... however, it does not say they need to be "immutable" and they can be updated
... adds "@jasonanovak" to the notes so that Jason would notice the notes

* Issue 99

issue 99

McCool: closed

Barry: fine

* Issue 100

issue 100

McCool: adds notes
... actually, TD notifications are useful to mitigate privacy issues...

* Issue 98

issue 98

McCool: have not responded much to Matthias yet
... example of logging
... get access by credential presented every time
... exchange scheme seems to be useful
... one issue
... requiring semantics
... one way is simply to have credential
... need to look into URI template
... on my todo list

* Issue 97

issue 97

* Issue 81

issue 81

* Issue 71

issue 71

McCool: guess this is still open
... would wait for Elena's update

* Issue 64

issue 64

McCool: adds comments
... Actually, a TD update/notification can be used to mitigate certain forms of privacy risk
... for instance, theID can be updated periodically and only authorized subscribers notified

F2F agenda

McCool: is there anybody to chair the possible security sessions during the upcoming f2f in Korea?
... shows the f2f agenda

f2f wiki

McCool: quickly skims the agenda
... would add information on "session leaders" to the agenda
... there are 5 topics on security
... put McCool's name to some of the topics
... review security metadata and scripting (McCool)
... security testing and validation planning (McCool)
... plugfest security review (Elena/McCool)
... privacy threats and mitigations (Elena)
... security implementation recommendations (Elena)
... also some test topics
... specification validation tools (Ege and?)
... TD validation tools (Ege and ?)
... test sutie for scripting API (?)
... online testing/demonstration systems (McCool)

AOB

Barry: regrets for the next week. see you in Korea!

[adjourned]

Summary of Action Items

[ONGOING] ACTION: mccool to write a short proposal on what security tools to use for the next plugfest
[ONGOING] ACTION: mccool to talk with IIC Security TF and W3C Web Security IG about testing/validation timeline
[ONGOING] ACTION: mccool to work on issue 70 (Require Not Exposing Immutable Hardware Identifiers?)
[ONGOING] ACTION: mjkoster/elena to review examples in the security spec
 
[NEW] ACTION: mccool to look into URI templates (RFC6570) for issue 98
 

Summary of Resolutions

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.152 (CVS log)
$Date: 2018/06/27 05:15:37 $