13:20:58 From Stephen : what is the difference knowledge object and information object? 13:22:13 From Dave Milham : If you need to contact us after the presentation drop me a email at dmilham@tmforum.org 13:23:54 From Michael Klotz (DTAG) : I'm also interest in that because we have several discussions about such object views in standardization groups (Klotzm@telekom.de). 13:25:21 From Jiachen Zhang(China Mobile) : Same here.(zhangjiachen@chinamobile.com) 13:26:24 From Ranganai Chaparadza : That was correct Explanation w.r.t difference between Information Object, Data Object and Knowledge Object. ETSI GS AFI 002 13:26:51 From Dave Milham : let's pick these last three points in the open discussion after the presentations. 13:41:13 From Zou Lan : Could you please clarify what does "REST and CRUD operation" mean? 13:43:40 From Chunying Tang-AsiaInfo : Please clarify why intent is an knowledge object. 13:43:50 From Dave Milham : we will pick this up after the presentation REST is Representational State Transfer ( widely used in madern HTTP based APIs, and CRUD is Create, Read, Update, Delete 13:44:20 From Raymond Forbes rwx651084 to Dave Milham(Direct Message) : What is Jorg's organisation 13:44:47 From Dave Milham : Knowledge object will be covered later 13:45:07 From Dave Milham : Joerg is with Ericsson 13:45:28 From Raymond Forbes rwx651084 to Dave Milham(Direct Message) : Thank you 13:45:45 From Raymond Forbes rwx651084 to Dave Milham(Direct Message) : it on this slide 13:45:51 From Raymond Forbes rwx651084 to Dave Milham(Direct Message) : it's 13:51:37 From Zou Lan : thanks Dave. I think REST and CRUD are not concepts on the same level, CRUD is more about stage 2（protocol neutral) term while REST represents stage3. For example, CRUD could be realized with REST. Putting "REST and CRUD operation" together causes confusion. 13:54:11 From Dave Milham to Raymond Forbes Huawei UK(Direct Message) : that's correct REST is stage 3 and CRUD is more general 13:57:07 From Dave Milham : that's correct REST is stage 3 and CRUD is more general 13:58:32 From Stephen : Could you comment on RDF/RDFS. Is it mandatory to understanding the proposed interface(s)? 14:06:23 From Cristina Badulescu (ETSI ISG NFV Vice-chair) : It would be great to schedule another session focused on the intent model in IG1253A, B. Seems many questions and interest are on the modeling aspects 14:07:07 From Ranganai Chaparadza : Enabling SDOs to "inherit" the Core Model and extend the model for applying them in their own architectural Frameworks is good idea 14:08:30 From Michael Klotz (DTAG) : Who will manage the core model and decide about the content? 14:08:40 From XU Ruiyue : RDF have mechasim for inheritence? 14:12:26 From Stephen : Achieving commonality could have been easier by starting with using the already known/widely used CRUD operations. for example, the SET operation could be achieved with CREATE. is there a chance to update the documents accordingly? 14:13:07 From Kevin McDonnell (Huawei) : Hi Xu Ruiyue, RDF doesn't support inheritance, but RDFS does (classses, properties) , OWL further extends RDFS. 14:20:42 From Jiachen Zhang(China Mobile) : If RDF doesn't support inheritance, how does the model federation deal with the relationship with the existing model of SDOs? In serveral SDOs, the modeling of intent doesn't be expected to be defined separately. 14:20:51 From Zou Lan : It would be helpful to provide an example on how federated approach could work across multiple standard groups, e.g. with indicating which group is responsible for which part and how a developer uses the stardard etc. Thanks! 14:24:21 From Michael Klotz (DTAG) : What about the legal aspects concerning the core model, etc.? 14:24:22 From Zou Lan : RH 14:26:03 From Ranganai Chaparadza : Godd discussion, thanks. The Suggestion for further discussions on Modeling, Federations and "Inheritance" aspects will be helpful 14:27:42 From Cristina Badulescu (ETSI ISG NFV Vice-chair) : The idea of a common intent management API is quite efficient, it allows to reuse in OSS/BSS the same API for different types of intents. And regardless of which SDOs extensions are federated in the models of the different intents 14:28:19 From Magnus Buhrgard : It's not only about working across standards organizations. An operator consists of different organizations and also needs to involve customer networks. A federated model would help. 14:31:39 From XU Ruiyue : Share same opinion with stephen, why not use CRUD API for intent LCM , several SDOs aleady provides the CRUD APIs. 14:38:39 From Thomas Tovinger : RH 14:40:04 From Ranganai Chaparadza : That is good Approach, to enable SDOs to extend as they like 14:42:05 From Zhengguangying : may be can use ABNF to define the meta model of intent, and map to RDF or UML, or some other languages 14:43:31 From Stephen : Critical is that the common model should in principle be as independent of any technologies, frameworks, etc as possible. 14:44:04 From Zhengguangying : right, agree with Stephen 14:45:05 From Ranganai Chaparadza : I agree with Stephen but that is what Jorg and Kevin are saying, to my understanding... the Core has to be Generic 14:47:07 From Raymond Forbes Huawei UK to Dave Milham(Direct Message) : Thank you