IT MAY BE HARD FOR TODAY’S GENERATION TO BELIEVE , but America’s space program was once the very emblem of high technology. Nowadays the era is rarely invoked except in smirky phrases like “spaceage bachelor-pad music,” and with spacecraft having come to resemble sport-utility vehicles, 23-year-old graphic designers use clip-art images of rockets to convey retro campiness. Even the space suit has an antique air, as science fiction characters now mostly dress as if they belonged in a Shakespeare play.

With the moon long since conquered, and manned travel anywhere else manifestly implausible, the country’s space program has lost its adventurous aspect and become nothing more than a scientific research project, and science, as every American knows, is boring. That’s why space flight is as passé as a fondue set.

All of which renders it less than surprising that the Historic American Engineering Record (HAER), an archival body that normally makes drawings of abandoned textile mills and wooden bridges, has in the last few years begun documenting Alabama’s Marshall Space Flight Center and its Redstone rocket test stand. These projects are just a few of the scores that are mentioned by Eric DeLony, the chief of HAER, in a recent retrospective article covering the 30 years since HAER’s founding. (The article appears in a recent issue of IA , the journal of the Society for Industrial Archeology.)

HAER was established in January 1969, in the dying days of President Lyndon Johnson’s administration. The timing was fortunate, for as DeLony points out, “had the HAER appropria- tion been postponed one year, it may not have passed because of the mounting expense of the Vietnam War.” At the same time, the prevailing knock-down-and-replace mindset of urban renewal (another 1960s idea that failed, though much less tragically than the Vietnam War) was giving way to a new revitalization concept based on rehabilitation and reuse of historic structures. The shift opened many new opportunities for historic preservation.

The recession of the mid-1970s, however traumatic it may have been for the country, was fruitful for HAER. While economic trends were leaving many industrial sites idle, few of them would actually be demolished until con- ditions improved. Even better, the poor employment market made for a bountiful supply of applicants for summer jobs to take measurements, make architectural and engineering drawings, and sift through records.

The late 1970s and early 1980s were bad HAER days. The trouble began in 1977, when the federal government instituted a new policy called “rehab-action.” Under this scheme HAER went beyond its usual historic documentation to “identify adaptive reuses for industrial buildings and strategies for revitalizing depressed industrial towns.” Like most Carter administration initiatives, rehabaction was well meant but ineffectual. As DeLony notes, “Some critics questioned the validity of rehab-action studies that were conducted by graduate planning students in 12 weeks to resolve social and urban ills that had baffled professional planners and sociologists for decades.” Carter also decentralized HAER’s organizational structure and merged it with several other agencies in an umbrella body called the Heritage Conservation and Recreation Service (HCRS).

Then came Ronald Reagan, elected President on a smallgovernment platform. His Secretary of the Interior, the diversity-minded James Watt, abolished HCRS and its local offices and sent everything back to the National Park Service in Washington. Like most Reagan administration initiatives, the cutback was well meant but draconian. In the ensuing confusion many staff members left, and the organization did not completely regain its bearings until 1984.

Things have improved considerably since then. Beginning in the mid-1980s, a wave of infrastructure replacement projects has given HAER plenty of business recording publicworks installations before they are demolished. (The results of many old and new HAER surveys can be found on the Library of Congress’s Web site at www.loc.gov. ) Now, with the end of the Cold War, many defense sites are also being decommissioned. It seems inevitable that a decade or two from now, HAER will have its hands full documenting the coffee machines, warrens of cubicles, and bulletin boards filled with Dilbert cartoons of today’s Silicon Valley companies.

AS THIS ISSUE’S “POSTFIX” points out, trains need smooth, reliable rails to run on or disaster will result. The New York Central’s record-setting jet-propelled run in 1966 would never have been attempted without a long, straight stretch of nearperfect track. Even at ordinary speeds a single rail that cracks or splits can cause a horrific derailment. One of the hardest challenges facing the steel industry from its inception was mass-producing steel that could stand up to decades of pounding by tons of hurtling machinery. Ever since the birth of railroads, in fact, the fragility of rails has limited how fast and how heavy trains could get.

In the April 1999 issue of Technology and Culture , the journal of the Society for the History of Technology, Mark Aldrich of Smith College explores a period early in this century when increasing train loads and defective rails combined to cause a distressing number of wrecks. As one might expect, the first reaction of the steel and railroad industries to the rash of accidents was to blame each other. Railroads attributed the accidents to low-quality rails, while steelmakers cited oversize trains, ineffective counterbalancing, flattened wheels, poor roadbed maintenance, extreme temperatures, and other circumstances beyond their control. They suggested the adoption of heavier rails, which would use more steel.

At first the railroads carried the day. Investigation showed that “A” rails—those made from the top portion of a steel ingot—had the highest rate of defects. This happened because impurities tended to accumulate in greater quantities at the top and because improper cooling sometimes left a cavity there. Some railroads rejected all the A rails in a shipment if tests showed an unacceptable number of defects, or they simply demanded that producers discard the top 25 percent of each ingot. Others switched from U.S. Steel and its Bessemer process to smaller producers that used the open-hearth process, which yielded a steel that was lower in brittleness-inducing phosphorus. Adding titanium, silicon, or aluminum also helped reduce fractures.

These changes caused a dramatic drop in rail failures—75 percent between 1908 and 1912. Yet as trains got heavier and traveled faster, rail-induced accidents continued to plague the industry. The chief culprits were “transverse” fissures, which shattered a rail into segments, sometimes a dozen or more. The fissures tended to start on the wear side of the rail (the side that made contact with the wheel flange), and inspection of rails that had failed in this way showed no obvious flaws. Railroad men found it hard to blame the steelmakers when they didn’t know what the problem was.

In the end, improvements in analysis, testing, and production combined to greatly reduce the problem of transverse failure. In the early 1920s deep etching with acid and microscopic inspection revealed the existence of tiny “shatter cracks,” similar to the “snowflakes” that had weakened gun barrels during World War I. A few years later a British engineer named C. Peter Sandberg found a way to decrease the formation of the cracks by cooling rails in an oven that kept their insides and outsides at the same temperature. (The Canadian engineer I. C. Mackie came up with a similar process in 1931.)

Sandberg had thought that his method worked by avoiding the thermal stresses that were caused by differential cooling. Later investigation showed that his slower cooling actually prevented dissolved hydrogen, which was known to make steel brittle, from coming out of solution. Whatever the reason, it worked. At the same time, Elmer Sperry, known for his gyroscopic inventions, came up with a device that detected incipient rail fissures electromagnetically. It could be mounted on a test car and driven across a stretch of track to find problem rails. In one dramatic case the Sperry device indicated a fissure, and when the rail was removed, a few taps with a ball-peen hammer were enough to break it.

By the early 1940s rail shattering was rare on America’s railroads. The author cites this episode as an example of productive failure. Equally important, it shows how two antagonistic industries can cooperate in the face of a common threat. By itself, carnage on the railways was not enough to bring carriers and steelmakers together; each side’s chief concern was to avoid blame and make the other side pay for finding a solution. But as Aldrich writes, “Both the carriers and the steelmakers perceived rail failures as a political threat, for each understood that a well-publicized record of wrecks could lead to increased government intervention.” Perhaps the most important lesson is that government intervention in technology often works best when the government does not actually intervene.