I've had decent luck with getting some weight out front but think that for my body, a truly satisfactory solution would remove or substantially reduce contact with the front of the body by carrying the weight of the front load on a structure that is hard-tied to the back load.
The front-loading project is worthwhile in my estimation, not only to reduce the fore-aft balancing effort, but also to allow lower-effort vertical axis rotation enabled by the reduced polar moment of inertia created by spreading the load around more of the body surface (thereby keeping more of it near the vertical axis) rather than cantilevering most of it farther in the rearward direction. All else equal, a longer cantilever should require more massive structure, and as Glen alludes, the structure needed in his circumstances for a full water load is wasted when the water load diminishes.
But I think there are some limitations for my body with soft front-loading arrangements. Reversed fanny packs and the speeder belt arrangement were of some use but I found that pressure near the front of the hips tended to induce ilopsoas spasms. Among other approaches, I tried twisted windbreakers loaded with various items, and a water bag held by straps sewn into a nylon casing. Body temperature regulation was also affected, sometimes for the better, but since there was a net reduction in flexibility of temperature regulation if the load was left in place, it was hard to count that as a positive. Heat input to the load was also altered. Finally, there was some sense of wasted effort as the front of the legs applied force to the load with each step.
I think the gym and commute applications for audio have some greater potential, but that it will take some time for that potential to be realized. It seems to me that we're in the very early days of providing a base of useful audio that can be informedly and selectively deployed on occasions when other information delivery channels are not available. Personally, the inability to get much utility from audio has led me to fairly strictly minimize transportation time (even at substantial direct career cost), and to long ago switch my cardio time at the gym to a recumbent stationary bicycle, which readily allows reading, at least up to a certain cardio output level.
I think it's probably helpful to keep the idea of illustration vs. alternative delivery mechanism in mind when preparing content, so as to produce both the best text-centric content and the best standalone multimedia content.
Bill, it's not so much that I want more (in the sense of overall volume) from online information, but that I want to preserve effective random-access capabilities so as to support effective research and time management. Better random-access capabilities do, however, let me gorge more overall effective volume :)