I've just received an update from Brad (thank you) which mentions that client/server modes can't be run together, or even switched between.
That's unfortunate for all the work that's been put into getting the project this far, so I really hope it's just a temporary drawback?
Every application I have in mind for the shield needs the ability to "switch" between server/client modes. It's important for me that if an action is being requested via a URL string, then hardware activity is fed-back to another server, especially in the event the request fails to execute... for one reason or another.
So I'm wondering (Brad/Ben) if this is an issue that can or will be resolved, or are the shields capabilities at the limit?
Can this problem be overcome by adding a dedicated 328 to the shield (next time you do a production run) just to handle the stack and libraries, and then have the shield connect to A4 and A5 of the board which handles user-specific client/server applications over i2c?
Adding a dedicated micro to the shield for a few extra dollars seems like a cheap solution (to me) for the problems it would solve.
What are your thoughts?
Have you guys seen this?http://news.bbc.co.uk/1/hi/technology/8063975.stm
Sounds like you might have 60-70 buyers wanting WiFi modules for Arduino to "tweet" at the push of a button = )