As someone who's been slowly working up into distribution, I question the premise here. The collapse of Wizard's Attic was absolutely a disruptive, destructive event, but that was more because it hurt the individual publishers who got caught in the blast than because of its impact on distribution in general. From where I stand, distro is doing ok, or at least as well as the industry in general[1].
However, accepting the premise, then I suggest that if there's something you want to keep an eye on, it's RPGNOW's forthcoming transition to a print on demand shop. They're already an incredible resource for straight online distribution, and if they can successfully implement POD, that has the potential to create an entirely new model. It's not necessarily a model that will work for everyone, but it's going to have an impact.
It's worth noting that the effect of RPGnow POD would likely be more profound than just on RPGNow's sales. Whatever setup RPGNow settles on could become a de facto standard, and that's important as POD machines become less expensive. RPGNow's POD could create a sufficient mass of POD-ready files that it becomes reasonable for a store owner to invest in POD machine of his own, and arrange electronic distribution through RPGNow or some other vector. Such a store could produce any of these titles in a few minutes. Of course, a lot hinges on that theoretical critical mass. It would have to include enough books capable of making a decent sale to merit the upfront cost of a POD machine which, even as it goes down, will likely be substantial. Which is to say, this is far from a certain prediction.
More broadly, the role of the distributor is subject to the same scrutiny as every other member of the distribution chain from the publisher to the game store. As we move away from having only one possible model for getting games from creators to players, it becomes more and more necessary for there to be a reason for each step in the chain to get a slice of the pie (and a better reason than 'because it's always been that way').[2] Some of these benefits are simple to identify. For example, not every game designer wants to bother with maintaining a web store, handling shipping, warehousing, getting their product to cons and so on. Traditional distro may handle some of those. New-fangled distro (like IPR) handles more. But there's no one perfect product yet.
Now, as a word of caution. From a designer's perspective, it seems like the ideal solution is to establish personal relationships with many game stores and arrange the details of how you get your game on their shelves personally. The designer sometimes sees more money per unit this way, and gets to make a human connection. This is nice, but it is very difficult to scale. Not only does this get onerous for the designer beyond a certain point, it's hard for a gamestore owner to maintain many of these relationships. Chris Hanrahan, owner of Endgame (a fantastic game store in Oakland, CA) has noted that the average game store carries more SKUs (Distinct individual products) than the average Costco. That is to say, they have to keep track of a lot of stuff, and it is much easier for them to deal with an aggregator (whether traditional distro or otherwise) than maintain many individual accounts.
This is not to say there's no way to change the distro model, of course, merely a point that would need to be addressed by whatever brave new model we're awaiting.
1 - Is the distribution model flawed? That's a different and equally interesting question
2 - We're not alone in this. It's the same discussion that book printing is facing. We're just a little bit ahead of the curve.
