This isn’t about HourGlass (at least at the moment), but just a general observation/proposal. Though actually the multiple sound sources weights morph ability in HourGlass is somewhat related to this.
I can’t think of a DAW or audio editor that has any other concept of “audio transitions” than volume based crossfades. This seems rather limiting if you compare the situation what the video people have at their disposal with all the wipes and other goodies. I don’t see why the concept of a transition in the context of audio processing could not be extended to a more sophisticated level.
This would really just involve having a concept of “deconstructing” and “constructing” sounds. In the traditional volume crossfade these would correspond to fading out the “left” sound and fading in the “right” sound on the timeline. But imagine if instead those processings could be lowpass/hipass/bandpass/bandreject filters, granular processors, spectral transforms etc. Someone really should do this! (And in the context of an audio editing paradigm. Obviously these kinds of things would be doable in Max/MSP, CSound etc, but then you’d also get the problem that those apps totally suck for doing anything that should resemble editing and arranging audio on a timeline.)
Now, someone will say “just put in some plugins, route the audio as desired and automate them blahblahblah…”. But that to me just doesn’t seem like a fluid workflow as we are living in 2011. I’d really just want to be able to overlap 2 sounds on the timeline and to get a more complicated transition than just a simple crossfade between those sounds…