Talked about on the second wireless summit, it might be a good thing to have a “wireless linux” certification. There should be different levels for this and the details need to be worked out.
Let me start with a short story. Some months ago I walked in to a small computer store at my university town and asked for a WLAN card for my PCMCIA slot that would work under my Linux setup. They couldn't help me. It took them two weeks to figure out, that the very expensive Netgear card I'm using now should be suitable. And it took recompiling my kernel to make it work.
As you can tell, I'm a user. I didn't want to invest the time, browsing the internet, figuring out myself, what cards might be suited for my system. I just wanted to enjoy the comfort of WLAN with my old Linux laptop.
So what should the goal of a certification be? Making it easier for Linux users to find the right WLAN card for themselves!
Only if this is the aim, users and manufacturers will adopt a system of certification. As soon as some manufacturers got their cards certified (maybe the kick of for the process will have to be a certification without the manufacturers asking for it) users will prefer these cards because it makes it easier for them to choose. The certification implies extra information that they otherwise would time-consumingly have to search for. With a growing community of Linux users more and more people will want to have a wireless card that is compatible with they system. Manufacturers hence have an incentive to have their cards certified as “easily working with Linux”.
After having some thoughts on this and with respect to the thoughts outlined in “Why do we want to certify” I came to the conclusion that it must be clear to home users, what product they shall buy. Still I am aware of the discussion, how much of the driver sources must be open. To satisfy both aspects while laying the focus on the home user market (“I don't care, as long as it works.”) I suggest the following different seals.