The Brooklyn Rail

“Any architect sketching the layout of a parking lot these days is likely using more electronic computation than Frank Gehry did in the 1990s to design the Guggenheim Bilbao.” 
—Mario Carpo¹

Today, there are innumerable computational tools at my disposal, which seem to endlessly, perhaps even mindlessly, self-innovate. In return, as I utilize them in design, I am now forced into a routine-like, perpetual upkeep of gigabytes of digital files which I continuously produce. My architectural career now involves a ritualistic practice of “digital archeology,” a performance of maintaining an updated work archive across the plethora of drafting, modeling, and visualization software I use. These will all inevitably become outdated and fade into technological obsolescence. Instead of a growing stack of neatly organized paper sketches and submittal drawings, a future digital footprint of my practice might consist of broken links, unsupported file types, and inoperable 3D-modeling interfaces.

In this era of overwhelming digital production in architecture, it is unclear what gets discarded and what gets maintained in the ocean of virtual tools and information. For instance, MVRDV.Cloud, the Dutch architecture office’s self-curated digital data archive, has to be “live” in order for it to safeguard MVRDV’s design legacy.² With archival efforts like these, it is noticeable that a significant dissonance has emerged—a palpable, unresolvable rift where the resilient and preservable mediums of the digital (or of steel and concrete construction) work against their own upkeep and survival in the long term. Contemporary digital archives and contemporary, digitally-designed buildings both seem to continually overwrite themselves.

Historians began to grapple with this conundrum only in the early 2010s—nearly two decades after architecture’s “first digital turn.” Prompted by Mario Carpo’s seminal historical overview of the digital in architecture, the fragility of digital archiving was exposed at Archaeology of the Digital, an exhibition which opened in 2013 at the Canadian Centre for Architecture (CCA). It highlighted the CCA’s multi-year research on the impact of emerging computational technologies on architectural production across its “digital turn,” while focusing on a selection of the earliest computer-aided building designs by Peter Eisenman, Chuck Hoberman, Shoei Yoh, and Frank Gehry from 1985–95.

Physical models, photographs, and hand-drafted drawings produced thirty years previously were exhibited in pristine condition at the show. These were paired with outdated, malfunctioning digital ephemera: computer-relics as mediums of architectural production. Nearly all of the native digital files behind these first computer-aided projects were lost—fully irretrievable. Old hardware needed to open them ceased to exist and CCA’s data storage formats became obsolete, as commercial software programs were no longer compatible to retrieve them.

Greg Lynn, the curator of this exhibition, articulated this the issue as “A (Great) Loss” in his introductory writing for its catalogue:

The iterations of digital files, the native digital objects and data-sets, as well as the tools and machines used in their production are disappearing with every migration to a new operating system, every move of an office and every upgrade in hardware.

Ironically, digital mediums seem to decay significantly faster than their physical counterparts, due to this “(Great) Loss.”³

According to the Digital Preservation Coalition’s two-hundred-page report on “The Global Bit List of Endangered Materials,” the 3D digital engineering, industry-specific CAD and BIM files are in fact one of the most fragile, “digitally endangered file-species,” due to their complexity and supply chain structures.⁴ Some of the earliest records of parametric models and 3D-environments were seemingly “archived” by the CCA, although just contained on hard drives since the 1990s, lying dormant and waiting for decades to be finally restored. Eventually, as time passed without maintenance, not only did the hardware decay, but the software needed to retrieve those files became virtually “extinct.” These computational design tools—either hard or soft—have been superseded by newer and improved systems, as if they were following Darwinian theories of evolution and natural selection.

Today, the amount of data produced behind every building design grows exponentially, almost uncontrollably. Architecture and its accompanying digital technologies traffic a false notion of timelessness and innovation, superficially resisting the very entropy and planned obsolescence that is intrinsic to their operation. Similarly to our digital files on old memory sticks, buildings only appear stable over time. The same narratives of competition and supersession which dictate technological innovation dominate the management and evolution of architecture as well, from the scale of individual buildings to the scale of entire cities.

Both formerly high-tech buildings and once up-to-date digital files are now perpetually replaceable. New York City’s Gillender Building was only thirteen years old when it was turned into rubble in 1910 for a larger tower. London’s Victorian Euston Station was entirely razed to the ground in 1962, after centuries of its consistent rebuilding and upgrading. It was eventually “superseded” with a very innovative-looking modernist concrete block. That building is already dysfunctional as well, unable to meet modern commuter traffic demands, and faces a wrecking ball yet again in order to accommodate faster high-speed railway systems. Even an entire historic neighborhood in Boston, the West End, was run down, legally designated as “obsolete,” and later demolished. Then, it was entirely rebuilt and re-planned across a few decades under an Urban Renewal Initiative claiming large-scale city improvements and social benefits via newer buildings and streets.⁵

Most recently, the Union Carbide headquarters, a 1960s monolith at 270 Park Avenue designed by Gordon Bunshaft and Natalie de Blois of Skidmore, Owings & Merrill (SOM), became the tallest skyscraper to be intentionally torn down. JPMorgan Chase claimed to need a larger, upgraded, and more innovative headquarters in replacement. SOM’s tower was overwritten by a Foster + Partners monolith, which just topped out in 2025 on the exact same site in Midtown Manhattan.

The new JP Morgan skyscraper is a theatrically extravagant, overly masculine structure which boasts novel biometric scanners, nineteen different eateries, and an internal wind breeze system to flicker an American flag in its lobby amongst its innovations. These major aesthetic shifts and minor technological advancements are housed in a structure almost twice the height of the demolished one, but offering only a few more floors of office space. No old records of the old Union Carbide headquarters can be found or retrieved onsite.⁶

With such instances of lost buildings (often in premature demise) and of irretrievable, over-saved files, the current “timelessness” fallacy of both digital mediums and modern construction materializes and disperses even further thanks to our thirst for innovation. The digital and the built environment fabric is always self-regenerating and self-improving without any records of changes left behind.

A hard drive’s contents are inscribed and eventually made inaccessible, as disks decay and software support flickers. Hard-as-steel structures and rock-solid memorials are erected, overthrown, and then ultimately forgotten. As buildings and the digital files made to produce them continually supersede themselves in a vicious cycle of obsolescence, perhaps permanent archival order in their matter is an unachievable phantasm.