The best outcome I see here is maybe a short run of 100 devices with the originally planned spec. They may cost a fortune, but developers and reviewers would get their hands on those and that would likely reignite enthusiasm in the project. With continued interest and some more small donations I hope there would be a second run, and it might be known by then how to lower the unit price, if newer components come on the market or some unsuccessful features are cut out. Either way it seems worthwhile to me, if only for the open documentation and designs that are being produced along the way, which are likely helpful to other or future projects.