Nomadix patents captive portal

by Rob Flickenger

Someone brought it to my attention that Nomadix has been granted a patent on their implementation of a captive portal for wireless hotspots. This might seem troubling for projects (such as NoCatAuth) that have been providing captive portal services for years. After all, where's the prior art? If you Google for "captive portal", you can't miss us, or about five other well-known implementations.


I think the critical phrase from the press release is: "This redirection takes place regardless of the host computer's settings and without altering the user's browser settings." Reading the full text of the patent makes it pretty clear that while they might like to make it sound like they've patented the idea of a captive portal, they've really only patented their wacky arp implementation.


Nomadix boxes look for any machine being ARP'd for that doesn't receive a reply, and issue an ARP reply using the gateway's MAC address. Combined with some DNS strangeness, this lets the gateway bring up a splash page even when the client requests a private intranet page, or even if they are using static IP address settings rather than DHCP.


NoCatAuth does none of that. We considered it for a while, but I don't think it's worth the effort, as it's really easy to confuse both the client and the gatway. I still think that software patents are detrimental to innovation, but this one doesn't seem to pose an immediate threat to our project.


Still, it is interesting that they've gone and patented it. Imagine if someone had gone and patented DHCP as "a method for automatically obtaining network configuration". Or maybe if ARP had been patented as "a method for finding the hardware serial number of a network communications device". Where exactly would that leave Nomadix?


How do you feel about software patents?