[PATCH] Refine compositor grabs behavior

[Daniel Stone]

Hi,

On 21 November 2015 at 01:13, Bill Spitzak <spitzak at gmail.com> wrote:
> You know I have been trying to point out that the equivalent bug exists on
> wl_pointer enter/leave events for a couple years now, but all I get is
> hostility (witness recent emails). If you do manage to fix this problem,
> perhaps you could look into getting it fixed there, too?

The hostility is nothing to do with you pointing out bugs, it's the
singularly unconstructive way you insist on doing it.

> Using a frame is a method of fixing it, but I think sending the enter event
> first, or not sending the exit at all, would also work.

Yes, I know you think that, because you've said several million times
that you think that.

Every single time, someone has pointed out that the protocol as it is,
requires a fixed ordering with leave coming before enter. Changing
this will break real clients. We cannot break this protocol. That's
what 'stable' means.

Grouping the events together in frames is a clever and useful solution
to the problem, which allows new clients to work better without
breaking existing clients. Not only is this a productive suggestion
from Peter, but it's also backed up by real patches. In contrast, all
you do is decimate the signal:noise ratio by copying and pasting the
exact same thing over and over; people give you the same explanation
as to how it would break real clients, and then you go silent until
the next thread where you do the exact same thing.

If you want to solve the problem, take review in its true spirit: read
and understand the criticism, respond to it, and produce something
better. Something you will have seen from literally everyone else on
this list. Instead, you use it as a platform to grandstand and
pontificate the same broken suggestions over and over. And you wonder
why you aren't taken seriously? I can't take you seriously, because
you refuse to engage and improve your mostly poor suggestions. Instead
of repeating yourself to an audience who don't care, why not be
productive and refine your suggestions to the point where they're
useful? What mileage do you actually get out of lectures so useless
that your good suggestions are dismissed out of hand, on the grounds
that 90% of what you say is rehashed, debunked, and would be
detrimental if implemented?

Going on past history, I don't expect you to take any of this on
board, so it will just have to stand as a marker that I can point you
at, for next time you try to derail a thread by derailing a productive
conversation into part 73 of a tiresome monologue on how you would've
implemented things if given a totally clean slate. But I'd love you to
prove me wrong by becoming a constructive contributor, rather than
that guy who makes constructive contributors avoid the list where
possible.

Cheers,
Daniel