Fair point, but I don't have enough knowledge to suggest the best way to split the task up.
I guess the (end)goal is to make it possible to compile a compatible client without needing the whole WIAB server code (so what is currently box/webclient could be a separate project in eclipse, referencing some library that abstracts the process of updating the wave state on the server). I am not sure what intermediate steps would be required to reach this though. I guess we would also want to separate out the data being transported from the method used to wrap it up....I vaguely remember this talked about recently but cant find the emails. On 17 January 2012 19:11, Yuri Z <[email protected]> wrote: > In any way, c/s is too vague. For any practical reason you should split it > into several items. > > On Tue, Jan 17, 2012 at 8:06 PM, Ali Lown <[email protected]> wrote: > >> On 17 January 2012 18:02, Thomas Wrobel <[email protected]> wrote: >> > (as I am pretty good at making webapps, guis etc, but >> > cant for the life of me get my head around the WIAB code base). >> >> I agree that the codebase does lack a clean separation between c/s, >> and the split of client code between >> src/org/waveprotocol/box/webclient and >> src/org/waveprotocol/wave/client does feel a bit awkward. >>
