A friend on Facebook pointed out the Skytran to me recently. It's quite a fascinating concept transit system, and the Skytran home page is worth your time to check out since I really don't think I could describe it with words that do it justice.
My initial thoughts on this concept:
First, it's definitely a much better option than traditional rail mass transit. As Randal O'Toole has argued (here, for example) rail mass transit is terribly inefficient(in cost per passenger-mile) and inflexible when compared to buses, for example, and also that it primarily exists because of perverse incentives in federal mass transit funding.
Second, slightly more negative, is that NASA is the primary researcher mentioned. As much as they've come up with cool technology I'm skeptical of their ability to either meet a consumer demand or develop and implement anything as cost-effectively as a private sector R&D organization.
Third, to be a serious alternative to the private auto, the Skytran would have to be flexible in its routing (i.e. each vehicle could go to any boarding station on the metropolitan Skytran network without the rider having to transfer from one vehicle to another).
Fourth, as simple as erecting utility poles is, it's formidable and expensive to replace them all across a metro area. You'd also have to plan on making major upgrades to the local power grid to be able to supply sufficient electricity for the Skytran system. Additionally the cost of sufficient vehicles to meet demand might add a lot, though certainly not as much as the tracks themselves.
Fifth, making the vehicles available when and where they're wanted is a challenge as well. Demand for highways, parking spaces, mass transit seats, taxis, etc. - modes of transport that already exist - is an order of magnitude higher at some times of day and days of the week than others, and is also affected by predictable anomalies like sporting events and concerts. Skytran systems would have to make some tradeoff between purchasing vehicles that sit unused much of the time, or increasing waits during high demand times.
Related to that, and I think this is potentially positive, the Skytran rider could much more easily be charged based on their point of origin and time of day than a highway user or bus rider. Until point 6 takes effect...
Sixth, federal and especially local politicians would love to impose their preferences on such a system. It's not farfetched to imagine generous handouts to transit unions, arbitrary route and fare mandates/restrictions, and various costly add-ons that wouldn't be chosen in a free market.