[openbox] Re: wine and focusing

[Clay Barnes]

On 18:38 Wed 27 Jun     , Dana Jansens wrote:
> On 6/27/07, Clay Barnes <clay at hci-matters.com> wrote:
> >On 12:18 Wed 27 Jun     , Dana Jansens wrote:
> >> Given the extreme inconsistencies between apps in how they use the
> >> standard protocols, this is really hard.  Adding any assumptions about
> >> the standards inevitably ends up breaking some applications that use
> >> them differently.
> >>
> >> You could look at the client_activate function if you want to prevent
> >> ever giving focus to applications except when you do it yourself (when
> >> 'user'==false).. but I don't see this going into a release, as it
> >> would only break lots of other situations.
> >>
> >> Focus handing is just a real big mess in X and we're doing the best we
> >> can with it, to get applications to work the way their developers
> >> wanted.  But it really limits what we can do beyond that.
> >>
> >
> >I really like the idea of keeping a default setup that does the least
> >harm with the most applications (especially those that are
> >standards-compliant) and having per-app focus rules in the
> ><application> section.  We should also have an easy to find section on
> >the wiki for maintaining a list of application fixes as they are
> >developed.
> 
> Yeah.. but I absolutely hate the idea of per-application work arounds
> and bug fixes.  Really really hate.  Broken apps should be fixed, not
> catered to so they remain broken.  Openbox would eventually turn into
> a giant prolog program or something.  So don't count on that either.

I fully agree---it's not our job to fix broken apps.  However, as
we're clamoring for those developers to fix standards compliance, we
could still offer the ability to deal with the very worst, most common
problem(s), like the focus issue, at least.  Perhaps?

-- 
Clay Barnes