How buildings learn: some reading as we prepare to discuss common spaces and other building-related matters

Adam Greenfield is an acquaintance who is also a sort of famous urbanist. When I told him I am involved in building a cohousing, he recommended I buy multiple copies of Steward Brand’s book How Buildings Learn and distribute them out to fellow Reeflings before we talk to architects, as it contains information that might avoid us many mistakes.

The main idea proposed by the book is that buildings adapt, and that adaptation defines the success of a building.

WHICH MEANS THAT

You should really, really try to get hold of an adaptable building. We had the right idea here, because after all we were going to adapt an old factory into a home. But we failed, and so now we need should really, really try to build a new building for adaptability.

WHICH MEANS THAT

We should try to build something with good bones (insulation, plumbing and services, light), but other than that as flexible as possible. We plan to live in The Reef for a long, long time. People will come and go; be born, grow up, age and die. Units might be converted into guest rooms, then offices, then back into units. The common area is almost certain to shapeshift as the Reefling’s changing interests feed changing activities, and the latter require new arrangements.

WHICH MEANS THAT

As @joannes said at the last plenary, every decision that can be postponed should be postponed. In a way, the ideal situation would be to build a common area as cheap and nondescript as possible: an undivided rectangle, polished concrete as a floor, minimal furniture (ideally scrounged about, or made out of Europallets); then live in it for a couple of years, and during this period experiment with different setups: “let’s move the living area over here, and the kitchen there (at this stage it’s just a camping stove, so it can be moved in ten minutes at zero cost, no remodelling needed), and that will free up this nice semiprotected corner for reading a book in the afternoon light”. That way the design of the common spaces would emerge from what we do in them. Brand:

“What makes the building learn is physical involvement from the people inside it.”

Apart from the design of the common spaces, the book also has implications for the building in general:

  • Design it not for the aesthetic pleasure of the passer-by, but for the pleasure of the Reefling using it. In fairness, I think that Stekke and Fraas are already on this wavelength.
  • “Flat roofs always leak”, and water is the enemy, so do not make the roof flat.
  • Choose materials that age gracefully, and will look good 40-50 years from now, not materials that will look good in the first five years of live of the building.

There are many more, and many inspiring examples of successful buildings, and others that were not successful at all, and from all of these we can learn, I am about halfway now, and am absolutely inspired. Maybe @joannes and I could propose a discussion in the runup to the workshop? But when, given that the workshop is coming up very soon (and I will be out of the country on the date that was selected…)?

Meanwhile, I have bought it in e-book format, but also in paper, and will bring the paper version to the next full members meeting for you to borrow and read. The book was also made into a TV series in six episodes, which Brand has helpfully loaded onto YouTube.

@reeflings

10 Likes

Thanks a lot @alberto and @joannes for this very important réflexion :smiling_face:

Will watch the YT episodes!

So, at yesterday’s full members’ meeting we reviewed the common spaces design process, and ended up with a suggestion that @joannes and I prepare a presentation and a proposal (to go “low road”) on this matter in the plenary of July 23rd. Joannes, is that acceptable?

I can certainly help but on 23 July I’m on holiday. Principles are there to guide us in making optimal, pragmatic decision.

This looks like it is being superseded by an available slot in the July 1st plenary, so all good.

While thinking about our future garden, I believe that we can certainly have a “garden that learns”.

  • A garden is more amenable to learning than a building; it is soft and we can try things out ourselves
  • A garden also needs a lot of learning. It’s difficult to predict how we will be using the garden and how nature takes a liking to our concept.

Here’s my conversation with chatGPT about the topic.

Agreed on all accounts. Gardens also learn much faster than buildings, on account on their malleable nature.

Small request: would it be OK to keep the use of gen AI to a minimum? Gen AI is wasteful, known to consume at least one order of magnitude more water and energy than regular web searches, all the while driving a super-concentrated industry with only 2-3 players worldwide. Sorry to be so neurotic, but I am so done with the tech industry.

2 Likes