[mythtv-users] Programming remote button bindings (WAS: What major features are planned for 0.27?)

[jedi]

On Fri, Nov 30, 2012 at 09:23:17PM +0000, Mark Greenwood wrote:
> On Friday 30 Nov 2012 12:18:57 Michael T. Dean wrote:
> > On 11/29/2012 07:32 AM, Mark Greenwood wrote:
> > > I was actually trying to help, since usability is my area of preofessional expertise. But I'm getting a real impression that the myth team aren't interested in making the user interface better, they only want more features and more stuff to fiddle with. So I give up, I'll go and try to help out with xbmc instead.
> > 
> > No, MythTV developers are working extremely hard to try to make things 
> > easier for user and to make the user interface better.  However, we have 
> > a plan for how it will be done--such that it doesn't just make things 
> > easier for you (or other people who have a remote like yours) and at the 
> > same time make things harder for other users (with remotes with very 
> > different layouts).
> 
> At the risk of going over the same old ground again, the kernel mappings of "the play button" to "KEY_PLAY" etc, make the remote layout irrelevant.
> 
> Yes, it would be *much* better if we had userspace tools to choose a remote 
> layout the same way we choose a keyboard layout, (that's all the abstraction 
> you need IF *applications* start supporting multimedia keys) but LIRC is not 
> the answer to that - Keyboard layouts are global to a partiuclar system, 
> remote layouts should be too. (IMHO, if that's necessary). LIRC makes the remote 
> layout *application* dependant, which is too confusing for 99% of the population. 
> Me included :)

No it doesn't. Remote layouts are global.

The only thing that is application centric is what those keys decode to. It's
a single configuration file for mapping the system defined remote keys to 
functions in multiple applications. It could be spread to the four winds in
the various configuration files for the n+1 apps you might be interested in.
Or it might be completely ignored.

MythTV already does this. It has it's own "keymap" file.

Furthermore, when I see complaints from Windows users about the remote I've 
happily used for years, I am in no rush to "fix" anything.



> > must go into the key editor and /first/ remove the current binding for 
> > KEY_PLAY (from all contexts which would conflict) and then add the new 
> > binding.  This makes mapping those keys much more difficult than just 
> > "starting fresh."
> 
> Yes, I agree. It's a hugely complex problem, binding keys in mythtv. I 
> understand your point of view. My point of view is that there are simply 
> too many keybindings. The best interfaces can be operated with the minimum 
> number of buttons. We've both denigrated Apple remotes, but hey, 
> they work and they only have 5 buttons. Can the myth interface be operated 
> with 5 buttons? Probably not. Is that a limitation of the remote, or

No. Apple remotes don't "work". They require "chording" for basic features.

Apple products just take mundane features and ignores them. You can use the
equivalent of 4 cursor keys and enter not because it's a well designed interface
but because they kept removing features to the point where that's enough to
handle what the features the product still supports.

[deletia]