Re: Proposal and RFC: DAL, the Desktop Abstraction Layer

[Ikke <eikke eikke com>]

> > I'd like to hear your thoughts on this too, as it won't be very useful
> > to write something like this if nobody is going to use it.
> 
> As Mike Hearn says, I think what you described is simply what the D-BUS
> session bus is for. There's no need that I see for an additional DAL. If
> D-BUS doesn't do this as-is we need to change D-BUS so it does. This is
> the whole point of D-BUS.
True. Revision of possible implementation here[1].

> It's useful to collect and document these kinds of interfaces, for sure.
That's what would be done now.

> But I don't think a new library or code is needed on top of D-BUS itself
> (other than the various D-BUS bindings).
I do think this is useful. If I'm an application developer, I don't want
to be bothered with DBUS, finding out what the exact syntax is of
messages I have to send,...
It's much easier to do "dal_startNewTrack(MYAPPLICATIONID,
path_to_trackfile)" or something, where libdal (working title ;)) is
just a tiny layer above DBUS which takes care of connecting to the bus,
creating valid DBusMessages, sending them, and some error handling.
Otherwise the same code (connection, acquiring service,...) would need
to be rewritten for every application that wants to be DAL compatible.
If we keep the entry level as low as possible more people will try to
comply to the standard, isn't it?

> Havoc

Ikke

[1] http://lists.freedesktop.org/archives/xdg/2005-January/005712.html