@ntxdave said:
Interesting that no one else has replied but I will give you my take:
IMHO the future is not necessarily in DC's themselves but rather is in the ability to share your 3D models in an online environment that allows you to do both "walk thru" and object/component animations (i.e. movies, WebGL or one of the 3D viewers - see next paragraph). The drawback that I see (I honestly have not developed one) is that you must be in an environment that allows you to trigger the animation of the object.
I can animate the component like is done with a DC using plugins like SketchyPhysics, Keyframe Animation and SU Animate and then export to a movie. There are also plugins like CL3VER and Spread3D which also give you some of those capabilities. The key to all of these tools is that "I do not have to get me hands dirty coding".
I think the real future for this industry is the ability to view models in an interactive 3D environment. As I said above some of this already exists but I am looking forward to the day when I can build a model that has animated components, provides the ability to "walk thru" the model and stop the "walk thru" and "look around" yet at the same time interact with items in the model (something like your DCs). Again, IMHO, I think we will see these type of tools take over in the next 2 to 3 years. Oh, and without the need to do any "scripting/coding".
It will be interesting to see if any others respond 
Hi ntxdave,
Thanks for the perspective. Short answer, I agree. For DCs to become more than just a novelty that can be moved, morphed or animated, they need some of their limitation to be addressed. They need some added functionality and flexibility.
From the Sketchup user's perspective (that is the person(s) actually creating the models), I think that there is a place for DCs in that area between needing a full blown SketchUp extension to create components and having to draw one from scratch. I think if DCs capabilities were expanded a little, they could fill that niche admirably.
Finding the right component drawing on the 3D Warehouse is difficult. Have you ever tried to find a single piece of hardware, like the correct bolt or screw? This is critical for 3D printing applications, but it's a crap shoot at best. Having to do the research and reproduce a component to spec from scratch is a pain. If DCs were tweaked a little, they could be made to produce the desired results. Imagine a single DC named 'Bolts' instead of hundreds of individual models that are named and categorized in a completely haphazard fashion. See what I'm getting at?
I had envisioned a DC that through user input or web sourced table data, could replicate any bolt, standard or custom, in as many materials, copies and styles as the designer might require. Of the standard threaded bolt, there is a head of varying type (hex, Allen, knurled, slotted, Phillip's, countersunk, etc.), a shoulder of varying type (shaped, round, tined, etc), sometimes a taper between the shoulder and the shaft, a shaft of varying type (threaded [left and right twist, of varying type and pitch], shaped or cylindrical [no threads = a pin], tapered or straight, drilled or not); each portion with specific geometry (I haven't even touched on hardness and the accompanying markings). While that might sound like a lot, they could be represented by as little as four or five basic shapes. Selections could be standardized. Pick one as modify it accordingly. Custom components made from drop-down selections or user input.
I had also played with the idea of the DC creating two types of static swappable components; a low poly placeholder and a detailed, printer or render ready version.
I can't see how this could do anything but help accommodate your vision of the future, the organization of the 3D Warehouse and SketchUp modelers in general. But please keep in mind that I am a noob and don't really understand what would have to happen 'under-the-hood' to make these changes to DCs a reality. It too is just a vision.