I’m subscribed to email@example.com where Mark Nottingham posted the following today: This mailing list is now the official forum of the IETF HTTPBIS Working Group. If you're subscribed to this list, you're participating in the Working Group. I don’t think RFC 2616bis will solve the problems I’m interested in solving. So far the discussions are mostly about editorial edits such as getting links to registries in place and making sure the HTTP specification is in line with later published specifications. If I make any comments on the new specification they’ll be in line with the following:
Not introducing new features to replace old features. This doesn’t solve any particular problem as the old feature still has to be implemented. It’s better to fix the old feature.
Improved compatibility with existing deployed clients. There are a number of things HTTP requires that have not been adopted by popular clients, such as web browsers. I think it makes sense to evaluate that situation in a realistic way and revise HTTP if existing clients can’t be changed.
Well defined behavior. It should be crystal clear how to implement features.