On Fri, 2018-08-17 at 11:16 -0500, Jonathon Jongsma wrote: > On Fri, 2018-08-17 at 06:36 -0400, Frediano Ziglio wrote: > > > On Thu, 2018-08-16 at 16:08 -0500, Jonathon Jongsma wrote: > > > > > > > > The only real downside I could think of was that the stream > > > > device > > > > protocol would have to change in a non-backward-compatible way > > > > (see > > > > below). But since the streaming agent is still in early stages, > > > > that > > > > doesn't seem like a huge deal. > > > > > > I'm not sure, we made a release and no compatibility disclaimer was > > > made about the protocol compatibility (only about API/ABI > > > instability), > > > but at this stage I'd say we shouldn't worry about compatibility > > > yet, > > > probability of changes is still high. > > > > > > > "have to change in a non-backward-compatible way" is pretty false. > > I didn't see in your patch below the needs to do it. Just add > > capabilities > > and messages. > > Sure, my wording was not technically correct. But since the streaming > agent is not actually used yet, I didn't feel that it was worth adding > capabilities to the protocol yet since there's no existing > installations that need to be supported yet (right?). I agree with Jonathon here. The code is new and stable release wasn't done yet. Capabilities add complexity and maintenance burden and if we end up adding them before the stable release already, we've done a poor job. It was quite clear since the beginning this part of the protocol would need to be changed with multi-monitor too. Cheers, Lukas _______________________________________________ Spice-devel mailing list Spice-devel@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/spice-devel