"All buildings are predictions. All predictions are wrong." – Stewart Brand in How Buildings Learn: What Happens After They're Built (Penguin, 1994.)
Brand also wrote The Media Lab about MIT's digital media research center. In this book, though, he describes how, for reasons out of the architect's control (the stakeholders couldn't agree on who would occupy it, or what they wanted, early enough in the design process), the Media Lab's building is a futuristic debacle. The building is pretentious, ill-functioning and non-adaptable. A huge sterile atrium "uses up so much of the building that actual working office and lab space is severely limited, making growth and new programs nearly impossible and exacerbating academic turf battles from the first day. Nowhere in the whole building is there a place for casual meetings, except for a tiny, overused kitchen. Corridors are narrow and barren...."
Then there's this zinger: "The Media Lab building, I discovered, is not unusually bad. Its badness is the norm in new buildings overdesigned by architects." Well, based on some ambitous modern buildings I've known and loathed, I'd say he has a good point here.
The thrust of How Buildings Learn is that buildings are happier places when they are flexible, adaptable, and can be retooled as their future unfolds. It's better to put every imaginable future scenario on the table, and make a new building's design generic and flexible enough to accommodate most of them, than to force a consensus about what the future will be. Odds are that the future situation in general - and tomorrow's technology in particular! - could not be anticipated yesterday.
Pretentious residential architecture comes in for some of Brand's criticism. Captioning an illustration of an elaborate new McMansion, he has this to say: "Prematurely complicated, this (pricey) house attempts to look as if it has been added on to for generations. As a result, actual add-ons will be difficult, and the fussy complexity greatly increases the construction and maintenance costs of the original house." On the other end of the pretentiousness scale, countless plain, boxy houses have worked just fine for successive generations of inhabitants who subdivided rooms and added on as needed. Such houses' porches are like pseudopods – converted into an extra room as soon as the house needs one.
As an example of fruitful residential architecture, Brand singles out the basic San Francisco Victorian house. The floor plan is simple – a couple of good-sized rooms on the first floor; on one side of the house a stairway below an upstairs hallway which opens into three or four second-storey rooms. It is a practical, sturdy and adaptable kind of house, which explains why old ones are still going strong and new ones being built in San Francisco and elsewhere.
Last Sunday, I enjoyed Easter dinner with a family of four and nine other guests in a modern Victorian house. It's in the Houston Heights neighborhood, which has a goodly amount of surviving original Victorian architecture, so new Victorians fit right in. The house is just what Brand said about the Victorian floor plan. The almost windowless outer wall beside the stairs and second floor hallway is the west side of the house, its windowlessness a sensible defense against summer heat. But the bay window in the living room's front wall, and the tall windows on the east side of the house, let in plenty of daylight. The high-ceilinged living/dining room has space enough for a table for 14 people and walls painted sunny yellow with white wooden trim. There are also nice front and back porches and a pleasant narrow yard with roses. All in all a very livable home and highly congenial to guests. And isn't livable, practical, congenial space what human habitations should be all about?