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

[David Bird <dbird@google.com>]

On Mon, Apr 10, 2017 at 8:40 PM, Erik Kline <ek@google.com> wrote:

Doffing my co-chair hat and speaking as a general Internet citizen, I too would oppose such a thing.  We could end up with some kind of IETF Signalling System 7 where we ask the network if we're permitted to make a connection for each and every new connection like some kind of circuit setup.

I'd argue that no network can guarantee you a route to your destination. A walled garden might be an extreme case, but it could be extreme either direction; really small or large. ICMP Dest Unreach is already giving the client feedback on drops and network restrictions; route and firewall. I think the only limitations are in software :)

 

As a working group member, I do wonder how we might prevent abuse of any "solutions" we design.

I think 'abuse' in this case is a relative term, and can apply to public access networks and their users.

 

As a co-chair, I'm wondering whether, in addition to draft-nottingham-capport-problem, we need a document describing the limitations of any potential solution space.  (How does a client OS prevent malicious apps costing the user money by auto-extending a billed session? et cetera)

If we don't have an API, how would an malicious app extend the session?

 

_______________________________________________
Captive-portals mailing list
Captive-portals@ietf.org
https://www.ietf.org/mailman/listinfo/captive-portals