Architecture Deskilling — Why All New Buildings Look the Same

Middlebrow Podcast (YouTube Shorts, 2026-05-16). A short argument explaining why contemporary buildings are generic, why adding any craft or complexity has become economically and legally punishing, and why handbuilt heritage structures are now irreplaceable. The argument frames the value of 601 Delaware precisely: a 1932 building built with embodied hand knowledge before the abstraction layers existed.

Transcript (cleaned)

You've actually gone on the record on other podcasts and have said that being an architect sucks ass. Yeah, this is a really important point.

Essentially, what's happened to architecture since the development of AutoCAD, especially, is that architecture has been largely deskilled in a lot of ways. So, drawing, for example, is not commonly taught in schools. Everything becomes abstracted, essentially. The computer systems are so complex now that basically, those five-over-ones, the reason they look like that is that they're essentially the built version of a spreadsheet. So, all of the materials, aluminum panels, you can just plug and play. Whereas, like if you wanted to add a curve to a building, then that takes like a huge amount of labor because you have to think about the flashing for like the rain. There's a possibility you could get sued if like something leaks. You know, then you have to add like additional construction drawings because it's more complex than just like slapping that stuff on there. And so, everyone's like, "Why are all these buildings the same?" And the answer is it's because one, you have to minimize liability at all costs for the firm, which means like if you just play it by the rules and you just play it into these sort of systems, yeah. You're not going to get sued. Second of all, like it's the developer's bottom line. And third of all is that architecture itself is an abstraction. Back in the Middle Ages, when people were building the cathedrals, they were hand building and hand cutting the stone. There weren't plans. These were developments of the Renaissance. When you take a building plan that is already an abstraction from the actual act of building the building. Yeah. You're coordinating the labor instead of doing the labor. And so, essentially, the computer is another layer on top of that. Because you are so disconnected from the material, you don't touch the material, you're just in Revit, you're just in AutoCAD. You're dragging and dropping.

It sounds like you're playing a video game.

Exactly. You don't have any sense of, you know, the materiality of the building.

Or the sense of place, too.

It's just you just plop that building down — the same in North Carolina, in Texas, in Atlanta. And so, getting an architecture degree now in a lot of the state schools, for example, it's just a technical program to learn these programs. And a lot of architects are very upset because, you know, they went to school because they wanted to make a better world a lot of the time.

The argument distilled

Three compounding forces push all new construction toward sameness:

Force	Mechanism
Liability	Every deviation from plug-and-play systems (a curve, an unusual detail) multiplies the risk of a lawsuit. Firms self-censor complexity.
Developer economics	Aluminum-panel curtain wall systems have known costs; craft details have unknown costs. Developers buy the spreadsheet.
Abstraction stack	Medieval builders touched the stone. Renaissance plans were one abstraction layer. CAD/BIM is another layer on top of that — you are coordinating pixels, not material. The building is an output of software logic.

Result: "five-over-ones" — the ubiquitous stick-frame-over-podium apartment type — are literally the built version of a spreadsheet. Same building in North Carolina, Texas, Atlanta.

Relevance to 601 Delaware

The core inversion — which is what makes this useful for 601 Delaware — is that the podcast describes a failure mode (deskilled generic buildings) and the H&H factory is the photo negative of that failure. The 1932 H&H Coffee Factory was built before all three forces fully closed:

No AutoCAD, no BIM: Morris, Nooman & Wilson drew by hand; George W. Mitchell's crew built by hand. The curve of the columns, the bond pattern of the CMU, the rail-spur setback — all were problem-solved in material, not in software.
Pre-liability culture: 1932 construction predates modern tort exposure. The building has detail that no one would commission today.
Sense of place is embedded: The Delaware Street site was chosen for the rail spur. The building's orientation, massing, and loading dock are specific responses to that particular block in that particular city. It cannot be replicated in Atlanta.

This is the core interpretation argument: 601 Delaware contains embodied knowledge that cannot be manufactured again. The coffee factory is not just old — it is materially different in kind from anything being built now. That is what makes adaptive reuse preservation rather than nostalgia.

For grant applications

The Middlebrow argument can be inverted into grant language:

The 1932 Hoffmann-Hayman factory represents pre-digital craft construction — hand-drawn, hand-built, site-specific — that has become structurally unreproducible under contemporary liability, economics, and design abstraction. Its preservation is therefore not sentimental but practical: this type of building fabric will not exist in a generation if existing examples are lost.

For museum interpretation

The deskilling argument can anchor a gallery panel or public program: