ONG-RANGE prediction is a singularly no-win venture: before being proved right you are likely to be dead, but being wrong enough can make you immortal. So it is that we remember the pioneers of aviation writing who predicted at the turn of the 20th century that buildings of the future would have flattened roofs to provide takeoff and landing facilities for ubiquitous personal aircraft, and that rich people would remain in flight for as much of the time as possible because things would be less congested and the air would be cleaner up there. No less a futurist than H. G. Wells stated positively that mankind's dream of powered flight would be realized ''long before the year A.D. 2000, and very probably before 1950.'' He made this prediction in 1901, two years before the Wright Brothers' first flight at Kill Devil Hills.

In doing so, Wells defied Arthur C. Clarke's observation that while people tend to overestimate what can be achieved in a short time, they underestimate what is possible in the long term. This is a mistake that James Fallows has avoided as well. In his new book, ''Free Flight: From Airline Hell to a New Age of Travel,'' Fallows argues with grace, passion and thorough reporting that we are at the dawn of a new era in aviation, one that could finally usher in the personal airplane (if not flat-roofed houses) and, more important, a ''nationwide air taxi fleet'' of new five-seat jets that will liberate us from the airlines' despotic hub-and-spoke system and all its attendant evils. Fallows brings to the subject his ardor as a professed aviation enthusiast and his long experience in journalism, as national correspondent for The Atlantic Monthly and as former editor of U.S. News and World Report.

The idea that airline travel needs improvement is of course not a hard case to make: every passenger knows that virtually all the hub airports are saturated, causing record-setting delays; that seats are oversold; that bags disappear; and that under the cockamamie tyranny of hubs a storm in Denver can wash out a flight from New York to Orlando. To suggest just how many heels are cooling in our nation's airports these days, Fallows reports that 27 new restaurants opened at San Francisco International Airport just last year. As air travel has exponentially increased over the past 20 years, the commercial airlines have actually slowed down -- just in time for the heyday of warp-speed multitasking. In an era of wireless this and broadband that, when you can do your banking or order a potted plant online at three o'clock in the morning, Fallows argues persuasively that the airlines' maddeningly roundabout routings and steep premiums for last-minute bookings are ''distinctly out of phase with the evolution of the modern economy as a whole,'' which is being driven by companies that meet the schedules and needs of their customers, not their own.

Fallows sees the cure in thousands of underused airports that could be used for direct flights between small to medium-size cities -- Lancaster, Pa., to Ithaca, N.Y., say, a flight that requires five hours the airlines' way and a half-hour by private jet -- and in the development of new airplanes. He tells in detail the story of two of these planes: the four-passenger, single-piston-engine SR series from the Cirrus Design Corporation of Duluth, Minn., with 21st century electronics, new safety features and a price of less than $200,000; and a revolutionary five-passenger jet from Albuquerque's Eclipse Corporation that will cruise at 41,000 feet, fly more than 400 miles per hour, and sell for less than a million dollars. (Behind the Eclipse 500 is a truly astonishing innovation from Sam Williams, inventor of the engine that made long-range, low-altitude cruise missiles possible: jet engines that weigh only 80 pounds but deliver 700 pounds of thrust, for an unheard of 9-to-1 thrust-to-weight ratio.) ''Eventually,'' Fallows writes, these airplanes ''should make it possible for many people, much of the time, to travel the way a few very rich people do now: in greater comfort, without fighting their way to and from the crowded hubs, leaving from the small airport that's closest to their home or office and flying direct to the small airport closest to where they really want to go . . . at their own schedule, at a cost no higher than today's coach fares.''

To realize that vision, Eclipse estimates the need 10 years from now for 35,000 of their new small jets, which would make 30 million trips a year. Since the airlines have 7,000 airplanes in service today, that would represent a 500 percent increase in the commercial fleet (and a nice return for Eclipse's investors, including Bill Gates and Paul Allen). According to Bruce Holmes, who has directed NASA's important support of small-plane evolution, not a single new runway would have to be built to support such an increase in traffic except at the most crowded hub airports. Fallows quotes one of Holmes's stump speeches: ''In the late 90's, general aviation'' -- meaning everyone but the airlines and the military -- conducted a total of 37 million 'operations' (takeoffs and landings) per year. If landing systems and control services were improved at the 5,400 largest public airports, operations could go to more than 500 million a year.''

Hold on just a minute. My home airport is a concrete strip in Longmont, Colo., which must be among the smallest of those ''underused'' airports. But as small as it is, with our sky divers and kit plane builders and aerobatics pilots and antique restorers and student fliers and small-plane owners doing touch-and-goes, the traffic at Longmont's Vance Brand Airport takes on the aspect of bumper cars on nice days. I would not think of flying out of Boulder's tiny airport, which is much closer to my home than Longmont, because I can see all of its converging traffic -- sailplanes, warbirds, doctor-driven twins, business jets -- from the discomfort of my porch. Divide Holmes's 500 million takeoffs or landings a year by 365 days and then by 5,400 airports, and that would be 200 more flights a day into and out of our little home-away-from-home. Even assuming a future in which there are 800 people (four passengers per plane) each day who need a ride -- right now! -- to or from Longmont, Colo., where will we get 35,000 more well-trained pilots, unless a massive air war breaks out every few years? (Or would that be 70,000? Right now very few airlines fly without a second pilot, even if the airplane is certified for single-pilot operation.)

Without offering a solution, Fallows does acknowledge the need for ''faster, surer, more natural ways of training pilots.'' He also addresses directly the widespread impression among nonpilots (99.75 percent of the population) that flying in small planes is about as safe as diving off tall buildings. The planes of the future will be much safer, he explains, both because of new navigation technologies -- highway in the sky'' guidance systems with moving maps, for example -- and because of innovations like the Cirrus's parachute, an emergency device attached to the airplane that would allow it to sink rather than plummet to the ground. (It would also remove any chance of a controlled landing. Cirrus acknowledges this concern among pilots but believes the chute will make passengers feel much safer.)

If all goes according to plan, the Eclipse 500's will be, as Cirrus's airplanes already are, very welcome newcomers in a general aviation fleet that is for the most part very old and very tired. But I can't see anyone hailing an aerocab in my lifetime, or a lot more people becoming pilots just so they can spend less time in airports. This idea seems purely fanciful to me: Eclipse, Cirrus, Fallows, NASA, they're all dreaming. I believe I can safely predict that if personal airplanes or fleets of air taxis are ever to be, their time will not come before 2050 at the earliest, and very probably not before A.D. 3000.