Kinetic scroll in libinput Xorg driver

[Peter Hutterer]

On 25/10/2016 18:48 , Alexis BRENON @Wayland wrote:
> Hi Peter,
>
> Thanks for the feedback.
>
> I really understand the rationales behind this choice nevertheless, it
> comes with some drawbacks...
>
> You say that the kinetic scroll must be implemented by toolkits and
> applications, this implies to choose at which level? Desktop
> Environment, Window Manager, End-User Application.

in the wayland protocol, the axis sources are implemented in a similar 
way, enabling the end-user application (or rather, the toolkit) to do 
kinetic scrolling. In the end it's the only one that can do it, nothing 
else has the semantic context required.

>   * In the case of DE, what about users not using one?

ok, this will be unpopular with the internets, but the times of "I don't 
want to have a DE" are over. yes, you can still do it. you can run 
scripts in xinit to set up things, but the vast majority of users will 
need to/want to have a DE running that takes care of things dynamically. 
Hardware has moved on since the 80s, so has the functionality we've come 
to expect. It's not enough anymore to throw a few pieces of metal on the 
ground and say "here's your car". you can still do it, but you're now 
talking about a tiny subset of users that want things that way.

so the case of no DE is niche in X, and effectively non-existent in 
wayland (because you *need* a compositor).

>   * In the case of End-User Application, the user will have to define
>     his friction factor in each one, each using a different
>     implementation leading to inconsistent behavior... Moreover, it will
>     be a feature not directly related to the app, and I fear that some
>     (if not most) will not implement it at all...

then complain to the app. just because we could force it onto apps in 
libinput doesn't mean we should. especially given all the other drawbacks.

>   * Finally, my window manager is not ready to change to Wayland
>     (http://www.mail-archive.com/awesome@naquadah.org/msg05058.html),
>     and so is far from implementing features lacking from Wayland drivers.

again, a rather unpopular opinion but: change window managers. or stick 
with synaptics, or maintain a patchset on top of 
libinput/xf86-input-libinput (as you pointed out below). I understand 
why awesome can't switch (manpower is the main issue, I guess) but the 
same applies for us having to maintain features we think are a bad 
design decision.

sorry, I don't have any happier solutions for you there. In case of 
emergency, I've heard there are kitten videos on youtube though ;)

> Well, that said, I'm a real newbie about Wayland, libinput and Weston
> (that I just discovered yesterday). So I think that this question was
> already debated. If this is the case I'm really interested to see the
> thread (I didn't find it in the archive).

it's been coming up repeatedly, but I'm not sure there's a single thread 
that I can point you to, sorry.

Cheers,
   Peter



>
> Finally, I will probably implement the kinetic scrolling for myself. If
> someone else is interested, let me know.
>
> Kind,
> Alexis
>
>
> Le mar. 25 oct. 2016 02:31, Peter Hutterer <peter.hutterer at who-t.net
> <mailto:peter.hutterer at who-t.net>> a écrit :
>
>     On Mon, Oct 24, 2016 at 06:42:31PM +0000, Alexis BRENON @Wayland wrote:
>     > Hello everyone,
>     >
>     > I would like to implement kinetic scroll in the libinput driver
>     for Xorg.
>     >
>     > I know that it's probably not the intended use of libinput ; as
>     explained
>     > in the documentation, it's the client that have to manage that.
>     >
>     > However, as an Xorg user not happy with the synaptics driver, I
>     would like
>     > to add a similar feature (fixing small disagreements encountered with
>     > synaptics) to libinput, allowing Xorg users to easily move to libinput
>     > without losing this feature.
>     >
>     > My first idea is to implement the kinetic scroll using a thread
>     that sends
>     > axis events as long as there is no button event, key event or
>     motion event
>     > higher than a threshold.
>     >
>     > It makes some time since the last time I developed in C, and maybe
>     it's not
>     > the better way to do it. I would be happy to hear your advices.
>     >
>     > One thing I'm thinking of is then to add some options in the Xorg
>     > configuration file to enable/disable this feature, choose the events
>     > stopping the kinetic scroll and change some thresholds. This will
>     allow to
>     > easily disable this feature in the future in case the clients
>     manage the
>     > kinetic scroll on their own.
>     >
>     > What do you think of this? Is there someone already working on it?
>     Is my
>     > proposition a good way to implement it?
>
>     simple answer - "no". sorry :)
>
>     there are two reasons we didn't implement kinetic scrolling in
>     libinput. one
>     is the series of bugs we have in synaptics with stopping scrolling
>     at the
>     right time (leading to inadvertent zooms, etc.). that's easier in
>     libinput
>     because we know about all devices but the fundamental problem
>     remains - a
>     driver cannot know when kinetic scrolling is appropriate. e.g. sending
>     scroll events when the document is already at the bottom? What's the
>     threshold going to be? A single pixel movement may land you in another
>     window that you don't want to send scroll events to. A two-pixel
>     movement
>     may be considered acceptable to not stop scroll events.
>     A shift+scroll may be acceptable in one application but not in another.
>
>     Pushing this to the toolkits and applications is the only sane long-term
>     solution. Anything else gives us a short-term solution that in two-three
>     years time we'll just waste time having to working around.
>
>     Cheers,
>        Peter
>