ANYONE WHO COMPARES OLD AND NEW BUILD ings in a city of any antiquity will usually end up thinking: They don’t build them like that anymore. Appearances can be deceptive, of course; plenty of boring architecture from the old days has since been knocked down. Still, very few pre-World War II buildings could pass for background illustrations in a video game. And it’s hard to imagine future designers using today’s drab cereal-box architecture as a post-postmodern element, unless our capacity for irony multiplies even faster than our capacity for data processing.

All this refers to exteriors. Go inside a lovely old building and you are likely to find cramped floor plans and antiquated plumbing and heating. Flaws can also exist in less obvious areas, such as a building’s structural system. This is particularly true in a place like San Francisco, where engineers’ understanding of how to deal with earthquakes has been constantly increasing (though to be fair, most of the city’s big buildings survived the great 1906 quake more or less intact; the bulk of the damage in that disaster resulted from the subsequent fire). The recent renovation of San Francisco’s Pacific Union Building, engineered by Steven Tipping and Associates of Berkeley, provides an example of how the charm of a building’s outside can be preserved while its inside is completely revamped. (A detailed account can be found in the Fall 1998 issue of Structure magazine.)

When erected in 1907, the Pacific Union Building was an office tower, the largest one in the world made of reinforced concrete. After renovation—in a microcosm of San Francisco’s changing economy—its lower floors will be converted into retail space and its upper floors into a hotel. Fitting hotel rooms within the existing 16-foot-square column grid was not a problem, but modern retail design demands large open areas. To accommodate this mixed use, the lower four floors of concrete framing were replaced with a three-story steel structure built on a 32-by-24-foot column grid. Steel trusses replacing the old fourth floor support the upper stories.

To keep the building from falling down, the new structural system had to be put into place before the old one was removed. Of the original 86 concrete columns, 12 were strengthened and secured to deep-drilled piers, along with 15 new steel ones. Once the new frame was in place, the construction team made an iterative series of measurements and adjustments to be sure that each member was bearing the proper amount of stress. They knew they had “tuned the piano” properly when demolition contractors cut into the columns slated for removal and found that they did not settle or pinch their saw blades.

The seismic system also needed upgrading, not only because of changing standards but because the new building presented a different set of conditions from the old all-concrete structure. To protect the beautiful but fragile tile and terra cotta exterior, designers built in great stiffness against the area’s frequent small quakes. For larger seismic excitations, they designed the system to transfer stresses onto walls and braces, which can be sacrificed if necessary, and away from load-bearing members, such as columns and trusses. While computer simulations were very helpful, the designers did not neglect hands-on research. After visiting Kobe, Japan, to inspect the results of the 1995 earthquake, team members decided to reduce welding to a minimum in favor of bolted connections, which had held up better in the Kobe quake.

The same mixture of old and new pervades the entire project, even to the faux-antique name and print ads of the retail tenant, Old Navy. In a city where commuters travel to cutting-edge computer jobs on a transportation technology that was outdated a century ago, such a combination is especially appropriate. For a final touch, when the retail store opens at the end of this year, the new structural trusses will be left exposed for shoppers to walk beneath. As with the Golden Gate Bridge, the renovated Pacific Union Building will give San Franciscans a chance to appreciate how beautiful good structural engineering can be.

Computer History: Report From the Front

DURING THE RECENT SCANDALS IN WASHINGTON, partisans of both sides asserted that in the long run history would show their views to be correct. Appeals to the judgment of future historians have always been popular, based on a belief that once the passions of the moment have evaporated, a single unbiased truth will come out. Anyone who reads history knows that’s a joke. Historians are just as prejudiced as everybody else, and they make their living by disagreeing with one another. If anything approaching consensus does emerge, a self-proclaimed revisionist is sure to come along and “debunk” it, even if he or she has to invent a consensus for the purpose.

Technological history is not immune from this sort of rancor, particularly over questions of who invented what. The airplane, the electric light, the telephone, and many other inventions have multiple claimants to priority, each with their own set of supporters. Now computer history is falling into the same contentious pattern. Decades-old affronts and insults are rehashed to the point where the pioneers of computing come across like a bunch of jealous schoolgirls: Howard Aiken disses IBM, John Mauchly disses John Atanasoff (who disses him back decades later), Herman Goldstine and John von Neumann diss Mauchly and Presper Eckert, and everyone steals everyone else’s ideas. This historiographical trend is continued in ENIAC: The Triumph and Tragedies of the World’s First Computer , by Scott McCartney (Walker and Company, $23.00).

As the title suggests, McCartney is a strong partisan of Eckert and Mauchly, who built ENIAC at the University of Pennsylvania in the mid-1940s. He is hypersensitive to slights against his subjects, noting at one point that the Washington Post ran an obituary of Atanasoff but not Eckert. When the shoe is on the other foot, however, he is much more indulgent. Discussing a 1970s lawsuit, he writes, “Mauchly didn’t make a very good witness"—a polite way of saying that he didn’t tell the truth.

Despite all this, the book does provide some information that is rarely seen elsewhere. For example, it details John Atanasoff’s unsuccessful efforts to design a computer for the Naval Ordnance Laboratory after World War II. Back in 1941 Atanasoff, then a professor at Iowa State, had shown Mauchly a computer he was building, an event that has since inspired great controversy. Atanasoff soon left the university for a military job, and his postwar dalliance with computers was short-lived. Nonetheless, its very existence contradicts the statement seen in many places (including this column in our last issue) that Atanasoff had nothing to do with computers after leaving Iowa State.

McCartney also clarifies the familiar story of how Eckert and Mauchly’s UNIVAC computer predicted an Eisenhower landslide in the 1952 presidential election. As most computer histories report, the final margin in electoral votes was 442 to 89, almost identical to UNIVAC’s predicted 438 to 93. But since human pundits had forecast a close race, no one believed the computer, and UNIVAC’s projection was toned down before being reported on the air.

Arrogance? Fear of technology? Perhaps. But UNIVAC’s predicted popular vote was wildly inaccurate: 32.9 million to 19 million, or 63.4 percent to 36.6 percent. That would have made Adlai Stevenson the biggest loser ever in a two-man presidential race. The actual result was much less dramatic: 33.9 million to 27.3 million, or 55.4 percent to 44.6 percent. In light of these figures, UNIVAC’s electoral-vote prediction looks less prescient and more like a lucky guess. Shortly before his death in 1980, Mauchly made a prediction of his own by saying, “History is certainly going to change its point of view about me and Eckert and lots of other people.” Such thoughts may have comforted a dying man of modest means who had watched his successors in the industry get fabulously rich. Yet history rarely has a single point of view, on this or any other subject. And if books like McCartney’s are any guide, the fight over credit for inventing the computer will endure as long as computers exist.

The Most Intimate Machine

ANOTHER MACHINE THAT, LIKE THE COMPUTER, has benefited from advances in miniaturization and electronics is discussed in The Technology of Orgasm: “Hysteria, ” the Vibrator, and Women’s Sexual Satisfaction , by Rachel P. Maines (Johns Hopkins University Press, $22.00). According to Maines, a century ago vibrators were used for the same purpose as today, only the procedure was usually performed in a doctor’s office as treatment for a vaguely defined ailment called hysteria. Electricity allowed machinery to replace the doctor’s hand massage, with accompanying gains in speed and efficiency. Although the book is called The Technology of Orgasm , the second half of the title gets considerably more attention than the first—as you’d expect from a 1970s-bred feminist. Still, anybody interested in either topic is sure to find something rewarding. And the illustrations are fascinating, from heavy-duty office units the size of dental X-ray cameras to a home model that was advertised alongside electric fans and eggbeaters under the headline AIDS THAT EVERY WOMAN APPRECIATES .