Wednesday, 25 March 2015
Improving Captive Portals
Yesterday at IETF92 in Dallas, we had a “Bar BoF” (i.e., informal meeting) about improving the behaviour and handling of Captive Portals — those login pages that you have to click through to get onto networks in hotels, airports, and many other places.
Captive Portals cause a lot of interoperability problems and user frustration. In part, this is because they span across the entire stack, from WiFi and Ethernet to DNS and HTTP.
We hoped that getting some of the stakeholders together might give us some more information about the issues, and even ideas about how to improve things. One of the items discussed was a draft currently in the OPS area that defines a new DHCP option to make it easier for clients to know that they’re on a captive network, and we were able to get several Operating System vendors interested in it as a result - a promising start.
It’s early days, but we have a wiki page to collect what we know, and now a mailing list to continue work on. If you work on a captive portal implementation, or do captive portal detection, please join!