12:25:11 RRSAgent has joined #poe 12:25:11 logging to http://www.w3.org/2016/11/15-poe-irc 12:25:13 RRSAgent, make logs public 12:25:13 Zakim has joined #poe 12:25:15 Zakim, this will be 12:25:15 I don't understand 'this will be', trackbot 12:25:16 Meeting: Permissions and Obligations Expression Working Group Teleconference 12:25:16 Date: 15 November 2016 12:25:43 Agenda: Constraints - Technical Discussion 12:25:52 RRSAgent, make logs public 12:31:07 simonstey has joined #poe 12:32:19 Brian_Ulicny has joined #poe 12:33:36 present+ 12:33:49 present+ renato 12:33:57 victor has joined #poe 12:34:17 where are you connected? do you have a webex link? 12:34:26 the usual one does not seem to work 12:34:26 benws1114 has joined #poe 12:34:30 hi ben 12:34:37 hi 12:34:41 https://mit.webex.com/mit/j.php?MTID=m68305a7186fe840c6004b820a1f4eb66 12:34:45 present+ 12:34:52 present+ 12:35:34 smyles has joined #poe 12:35:44 present+ 12:36:02 present+ 12:36:14 can't scribe this morning. Need to leave a bit early to catch train. 12:36:46 sabrina sends her regrets 12:36:53 https://docs.google.com/document/d/1qGlBEpRgi60WyNnHLCfMuXlnTV1zzpoRz8DakIiOcP0/edit?usp=sharing 12:37:05 regrets+ SabrinaKirrane 12:37:14 present+ 12:39:12 michael walking thru the document 12:39:23 q+ 12:41:42 q+ 12:41:48 ack simonstey 12:42:56 +q 12:44:43 q+ 12:44:46 q+ 12:46:30 ack smyles 12:48:37 smyles: [elaborating on the differences between constraints applied to perm/proh./duties] 12:50:52 simon: what does it mean if the constraint is not meet? 12:51:32 ack simonstey 12:51:34 ack victor 12:52:27 victor: describes XACML architecture 12:52:35 +q 12:53:13 michaelS has joined #poe 12:53:29 present+ 12:53:54 ack benws1114 12:54:02 ack benws 12:54:18 Ben: the unit makes it a complex constraint 12:54:19 odrl:permission [ a odrl:Permission ; odrl:action odrl:play ; odrl:target ; odrl:constraint [ a odrl:Constraint ; odrl:operator odrl:lteq ; odrl:dateTime "2010-12-31"^^xsd:date ] ] . 12:54:42 q- 12:55:33 About Ben's complex constraint 50 AUD, I would say this would be trivial to fix if we had only one syntax. E.g., in RDF we would speak in terms of "RDF triple patterns" and everything would be evident. 12:55:39 q+ 12:57:17 ack simonstey 12:57:28 In OASIS XACML, PDP evaluates the constraints. Whenever a policy makes reference to a dynamic variable, the PIP is queryed in real time to obtain the value of an attribute (This feature is only available since version 3.0) 12:57:36 this, indeed, assumes that we are having "variables". 12:58:00 q+ 12:58:21 ack michaelS 12:58:44 To end with, I would avoid defining an architecture or replicating what OASIS XACML has, but perhaps some simple mechamism would not harm. 12:59:01 +q 12:59:03 +1 michaelS 12:59:34 q- 12:59:54 ack renato 13:00:10 q+ 13:02:26 http://w3c.github.io/data-shapes/shacl/#MaxLengthConstraintComponent 13:04:44 Simon, the problem of leaning on SHACL is that it only works for RDF. As I said, if we had one single syntax everything would be simpler... 13:05:08 q+ 13:05:08 no 13:05:11 ack simonstey 13:06:11 victor, http://w3c.github.io/data-shapes/shacl/#functions 13:06:47 q+ 13:07:26 ack benws 13:07:45 oh! do you think an XML implementation can also use that? certainly! To compare with, these are the URN-defined functions in XACML. http://docs.oasis-open.org/xacml/3.0/xacml-3.0-core-spec-os-en.html#_Toc325047234 13:07:51 +q 13:07:58 ack michaelS 13:08:02 q+ 13:11:06 ack simonstey 13:11:15 to answer Michael: in OASIS there is a component aimed at obtaining the missing values for certain attributes (PIP). The Policy information point (PIP) acts as a source of attribute values. 13:11:19 q+ 13:11:24 ack ivan 13:11:33 my question: shall we defined components? do we all agree that we won't? 13:13:25 s/defined/define 13:14:07 I'm going to need to run, but one common constraint would be "if API key is valid" but validity of API keys is not something we will define or can define. 13:14:24 q+ 13:14:44 Status: the current value of a constraint. https://www.w3.org/ns/odrl/2/#term-status 13:15:00 agree with brian - there are always going to be things that can only be validated "outside" an ODRL system 13:16:26 q+ 13:17:06 odrl:payAmount odrl:eq odrl:var1 . odrl:var1 a ODRL:VARIABLE . odrl:var1 ODRL:ASSIGNEDBY . 13:18:48 +q 13:18:57 ack victor 13:19:26 ack renato 13:19:44 -q 13:20:51 ack smyles 13:21:32 q- 13:21:47 +q 13:23:01 q+ 13:25:06 ack michaelS 13:27:05 These API (implementing ) might be eventually parts of profiles. 13:27:20 s/parts/part 13:28:21 ack simonstey 13:28:42 q+ 13:29:41 ack ivan 13:30:19 q+ 13:30:58 Queue is closed ;-) 13:32:32 q- 13:33:53 q- 13:34:30 shall we open a page at https://www.w3.org/2016/poe/wiki to discuss/analyze term by term? 13:35:32 I volunteer to open there a table specifying application range/domain 13:36:00 (not many decisions taken so far, though!!) 13:37:00 RRSAgent, make logs public 13:41:51 phila has joined #poe 14:13:16 benws1114 has joined #poe 14:17:23 benws1115 has joined #poe 14:38:56 renato has joined #poe 14:42:42 benws1115 has joined #poe 14:46:28 benws1116 has joined #poe 15:47:41 phila has joined #poe 15:57:55 Zakim has left #poe 16:12:35 renato has joined #poe 17:03:26 ivan_ has joined #poe 18:18:56 renato has joined #poe 19:45:23 benws1116 has joined #poe 19:50:46 renato has joined #poe 20:19:54 phila has joined #poe 20:58:51 renato has joined #poe 23:01:25 renato has joined #poe