Re: [Captive-portals] time-based walled gardens

[David Bird <dbird@google.com>]

On Tue, Apr 11, 2017 at 12:03 AM, Mikael Abrahamsson <swmike@swm.pp.se> wrote:

On Tue, 11 Apr 2017, Mark Nottingham wrote:

AFAICT addressing these use cases would require re-chartering, and that's something I would argue vigorously against. I'd like to hear a clear statement from the Chairs about what they think the scope of work is here.

I am not talking about control, I am talking about information. PVD is about information, ALTO is also about information.

Information can be used to control. 

 

For instance, I don't see a reason for CAPPORT to have its own RA option when a more generic one could be used for both CAPPORT and MIF PVD information. Unofrtunately I was "asleep" when the RA option was sent to 6man for review. Both (RA/DHCP) are about providing information to devices on the network. ALTO is also about providing information.

I think you can view all DHCP options as providing information about the network. Just because two options use a URL, doesn't mean the information is the same or even relevant for the use-case. Overloading existing mechanisms would make security considerations far more difficult. 

We already have RFC7710. I wasn't involved in that, and I agree it could have probably used more review. As it is now, it adds no value. However, it provides a starting point, and combined with ICMP it starts completing the picture.

So I guess what I'm saying is that I think it's a mistake to just make the work here, looking only at the CAPPORT use-case, and looking nowhere else. A more generic option, if not allowed by the charter to be done here, can be done in INT-AREA or 6man or somewhere else.

I don't like captive portals, I want them to die. But there are other use-cases for providing information to devices and I'd hate to have siloed solutions for every problem-space. I'd love to hear from implementors what their view is, do they want more specialised ways of doing things, or fewer more generic ones.

Seriously? What implementor of captive portals would want to comment in this (anti-)capport WG? Sure, you can probably find a couple venues that were forced to use a captive portal, but didn't want to (I can't think of an example, but maybe you can). We could design solutions for just them, but is that a sizable market for vendors?