Re: Custom shell surface protocol for Plasma surfaces

Am 2020-03-05 16:46, schrieb David Edmundson:
> On Thu, Mar 5, 2020 at 3:12 PM Vlad Zahorodnii 
> <[email protected]> wrote:
>> 
>> On 3/5/20 4:51 PM, David Edmundson wrote:
>> > What I would question is if we're going to make a change, whether we
>> > should go directly to layer-shell.
>> 
>> I would say "yes," but on the other hand I don't have a list of all
>> features required to make plasmashell work on Wayland. Perhaps we need
>> to compile such a list first, then based on it decide whether
>> layer-shell suits our needs.
> 
> Sure, that's a sensible approach.
> IIRC we did many years ago at a sprint.

That was mostly driven by Pier, who honestly had not much knowledge 
about Plasma. It never really covered what Plasma did and we also 
extended it a lot since then. Which is really important in such an 
evaluation: be prepared that things change. A protocol need to support 
also future ideas.

> 
> It depends how much semantic information we're happy bundling into the
> "namespace" argument, which is a freetext string for what kwin would
> refer to as a role.
> 
> i.e set_skip_switcher could be something a client sets, or kwin could
> know that anything of the namespace "notification_popup" should skip
> the switcher.

Due to experience with X I do not want something like set_skip_switcher 
anymore. I rather prefer having things semantically defined.

Cheers
Martin