Hi Danny,
Thanks for your revision. I mostly agree with your texts. But I put my small revision on your texts. For the last sentence, I hope we leave some spaces for any enhancement. If the network infrastructure supports On-Demand Mobility feature, the IP stack should act according to follow the application request by priority; If the application requests a specific address type, the stack should forward this request to the network. If the application does not request a specific address type, the IP stack must not request an address type and leave it to the network's default behavior to choose the type of the allocated IP address. If an IP address was already allocated to the host, the IP stack should uses this address and may not request a new ine one from the network. Regards, Seil Jeon From: Moses, Danny [mailto:danny.mo...@intel.com] Sent: Thursday, June 23, 2016 7:06 PM To: Seil Jeon; dmm@ietf.org Cc: '"???(??)"' Subject: RE: [DMM] WGLC #2 for draft-ietf-dmm-ondemand-mobility-05 Hi, Thanks for the support. I agree with your comment; the text might be misleading. The general principle is to maintain the old behavior (prior to OnDemand support) when one entity is legacy. According to this principle, if a legacy application is executed, it should experience the same behavior as if it would have been executed with a legacy IP stack and legacy network. The problem is however, that the behavior of the network in terms of IP address type allocation is not define. So I am thinking of changing the text from �C "New IP stacks must continue to support all legacy operations. If an application does not use On-Demand Mobility feature, the IP stack must respond in a legacy manner. If the network infrastructure supports On-Demand Mobility feature, the IP stack may still request specific types of source IP address transparently to legacy applications. This may be useful for environments in which both legacy and new applications are executed. The definition of what type of addresses to request and how they are assigned to legacy applications are outside of the scope of this specification." To (the red text is new, the strikethrough text will be deleted)�C " New IP stacks must continue to support all legacy operations. If an application does not use On-Demand Mobility feature, the IP stack must respond in a legacy manner. If the network infrastructure supports On-Demand Mobility feature, the IP stack should act according to the application request: If the application requests a specific address type, the stack should forward this request to the network. If the application does not request a specific address type, the IP stack must not request an address type and leave it to the network's default behavior to choose the type of the allocated IP address. If an address was already allocated to the host, the IP stack should use this address and not request a new one from the network. may still request specific types of source IP address transparently to legacy applications. This may be useful for environments in which both legacy and new applications are executed. The definition of what type of addresses to request and how they are assigned to legacy applications are outside of the scope of this specification." Is that OK? Thanks, /Danny -----Original Message----- From: dmm [mailto:dmm-boun...@ietf.org] On Behalf Of Seil Jeon Sent: Tuesday, June 21, 2016 12:52 To: dmm@ietf.org <mailto:dmm@ietf.org> Cc: '"刘大鹏(鹏成)"' Subject: Re: [DMM] WGLC #2 for draft-ietf-dmm-ondemand-mobility-05 Hi, I have read and checked the updates in -03 version. And I support this draft to move forward, if the following is clearly resolved. "4.2. IP Stack in the Mobile Host ... If the network infrastructure supports On-Demand Mobility feature, the IP stack may still request specific types of source IP address transparently to legacy applications." As this draft is based on the application-driven API selection idea, if no explicit request is given by an application, the type should be selected by default, though the default behavior could be regulated by the network operator. My comment is putting "based on a default havior" in a proper place in the sentence would be exact, leaving no misunderstanding as if the IP stack has the selection capability of source IP address type, instead of the application. One minor comment is [I-D.ietf-dmm-requirements] in the reference should be replaced by RFC7333. Regards, Seil Jeon -----Original Message----- From: dmm [mailto:dmm-boun...@ietf.org] On Behalf Of Jouni Sent: Thursday, June 16, 2016 2:16 AM To: dmm@ietf.org <mailto:dmm@ietf.org> Cc: "刘大鹏(鹏成)" Subject: [DMM] WGLC #2 for draft-ietf-dmm-ondemand-mobility-05 Folks, This email starts the WGLC #2 for draft-ietf-dmm-ondemand-mobility-05. Post your comment to the mailing list and also add your issues/correction requests/concerns etc into the Issue Tracker. WGLC #2 Starts: 6/15/2016 WGLC #2 Ends: 6/29/2016 EOB PDT Regards, Jouni & Dapeng _______________________________________________ dmm mailing list dmm@ietf.org <mailto:dmm@ietf.org> https://www.ietf.org/mailman/listinfo/dmm _______________________________________________ dmm mailing list dmm@ietf.org <mailto:dmm@ietf.org> https://www.ietf.org/mailman/listinfo/dmm
_______________________________________________ dmm mailing list dmm@ietf.org https://www.ietf.org/mailman/listinfo/dmm
