Reading the McMansion thread put me in mind of this excellent book. Has anyone here read it? For folks who haven't, it's a book about "what works" in architecture and in town planning. It's basically a distilled collection of rules of thumb regarding good design and building practices. An example is the rule that states that a good room will have light coming from at least two directions. In my old added-on house, we have several submarine rooms that have all their windows in one wall. Bad. The nicest rooms follow the rule.
You're referring to Christopher Alexander's book, correct? Interestingly (and OT), his work inspired a major movement within software architecture circles, as well.
A year and a half ago, my wife and I moved into our newly-built house near Port Townsend, Washington. Our architect was Ross Chapin, who lives and works on Whidbey Island. His specialty is small houses, which suited us perfectly because we're childless and have no need for a large house, and were also operating under tight budget constraints.
Once we had agreed that he would design our house for us, the first 'assignment' he gave us was to read A Pattern Language, and to note the patterns that resonated most strongly with us. The idea was that, having read and absorbed it, we would be communicating with a common language of design concepts. My wife and I each read it independently, and marked it up with Post-It notes as page markers as we read it. In our discussions with the architect during the design process, we frequently referred to specific design patterns and worked out how we would incorporate them in the final design. From our viewpoint it was an extremely valuable exercise, and I believe that it contributed to the excellence of the final product.
The house we ended up with (entirely unlike our original notions, as it turned out) is only 1157 square feet, but because of the care that went into its design it seems like a much larger house. Visitors are regularly surprised at how light-filled and spacious it seems.
As to the book itself, I think that the part of it that works the best is the part that talks about individual dwellings. Discussions of patterns that relate to public spaces - universities, town commons, and so forth, were illuminating as well, since I've personally seen public spaces that were awkward or impersonal and off-putting, and I can see from the book's treatment of those patterns how they could have been made more pleasant and useful.
When the book moves on to the design of towns and cities, I think it's going a little bit off track. The authorship of the book came out of Berkeley academia, if I remember correctly, and I think it's tinged with a little bit of that old hippie influence. It's sweet, but impractical, and I doubt if that part of the story gets much attention from the people who design at that level (if, in fact, any design goes on at all at that level!)
With all due respect, I doubt if A Pattern Language was actually the inspiration for the emergence of software design patterns as a software architecture issue. I think it's a sort of parallel evolution of thinking, but coming from an independent origin.
On the other hand, if you can come up with some seminal reference from the literature, I'd be interested in knowing about it.
We really ought to take this discussion elsewhere, though. This is felony thread drift in the first degree. If you'd like to pursue this, my real email address can be worked out pretty easily.
Best regards, Tom Dac>> Reading the McMansion thread put me in mind of this excellent book. Has
"Tom Dacon" wrote in news: snipped-for-privacy@corp.supernews.com:
Here's a start:
formatting link
you DAGS "design patterns software history" or variants of same you'll come up with a lot of references. The original software to which patterns were applied ultimately was a collosal failure, though through no fault of Cunningham, Beck, or the application of Patterns.
Perhaps it's my identity as an ancient hippie that makes me think this way, but in my opinion the town planning stuff is pretty good. As you probably know, several intentional communities (made up generally of hip yuppies, not ancient hippies) have based their design on A Pattern Language, and those communities appear to be vastly more livable, at least to my eyes, than what passes for planned communities in general.
I'll give a couple examples to support my opinion. In most places, zoning laws make it impossible to have a small commercial operation attached to one's dwelling. The laws are applied inflexibly: an artist's studio or antique shop is regarded by planners with the same horror as a hog farm.
Another: If you've traveled widely in foreign countries, you've probably been through a town where the houses fronted on the street, so as to have more garden space behind a privacy wall. Try to build a house like that in a typical sprawlburb.
But you're right. In general, whatever wisdom is contained in A Pattern Language is almost universally ignored. I can think of only one major development I've personally seen which owes a lot to the ideas in that book. About 40 miles down the coast is a famous planned community called Seaside. There's not a lawn to be seen, all the million dollar cottages front directly on the street, there's a good mix of residential and small business usage, the streets are closed to all but residents' vehicles and so forth. It has been wildly successful in a commercial sense and in terms of livability.
As those who haven't read the book may not know, the driving force behind the book is what some call vernacular architecture-- houses designed by those who live in them, rather than by an architect. The main idea is that houses people like to live in survive the centuries, and bad design gets neglected and disappears. The same might be said of city design. Good city design gets preserved, bad city design gets renovated. So I don't think the Berkeley origins of some of the authors had much impact on the patterns they isolated in old but highly livable cities.
Anyway, I think everyone thinking about building a new home would benefit enormously from reading the book. And every builder ought to have to pass a test on the book. Pipedream, I guess.
On Fri, 08 Oct 2004 13:59:35 -0500, Ray Aldridge calmly ranted:
I had read both of Sarah's books prior to finding that at the library and found it to be similar but not as good/insightful. I also remember disagreeing with some of his architectural thoughts but can't recall what they were specifically. Your experience may be different.
-=-=-=-=-=-=-=-=-=-=-=-==-=--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Life is full of little surprises. * Comprehensive Website Development --Pandora *
OK, I Googled around and found mention of Sarah Susanka: If this is the person, I also found an article in Residential Architecture. An excerpt:
Others claim Susanka was not the first to argue for better-designed houses or to explain the importance of intimate, human-scaled spaces within the home. And she would agree: "I've always described myself as a child brought up on Christopher Alexander's A Pattern Language. It's a fundamental text--something we can grow from. It's a great beginning. But it would give wings to the profession if we weren't afraid to be derivative."
Good work can certainly be derivative. That's the whole point of A Pattern Language, I think.
I'll have to see if I can get copies of the Susanka book. They look very interesting.
Damn, you had to bring that up didn't you. I owned both "The Fountainhead" and "Atlas Shrugged". Had them books for years and hauled them around every time I moved. Finally, I sold them on Ebay and never got a chance to read them past chapter one. Then I read somewhere that most engineers have at one time in their lives owned those 2 books but ended up not reading them through - I felt vindicated. I am going on a tangent. I also owned "The Pattern Language" never read that through either, but I liked the pictures.
I met a nearly-graduated architecture student recently. I asked her if she'd learned about passive solar and other alternative architecture approaches. She said no. I asked if she'd been introduced to "A Pattern Language". No again.
Architects are taught to build ticky-tacky. IMNSHO - architecture has failed society. Too pretentious and not practical. Afraid to lead and to quick to copy. Claiming to be artists and originators but inclined to follow blindly the latest fad. An architect that is actually good is an oddball within the profession.
If she's somebody you care about you might want to get her a copy of "A Pattern Language" and of Goswami, Kreith, and Kreider's "Principles of Solar Engineering".
Sturgeon's law--90% of anything is crap. But Susanka (IIRC it was her) points out that much residential "architecture" isn't, the houses were designed by a builder not by an architect.
I read the book several years ago, after many more years of hearing or seeing references to it (must have heard about it shortly after it came out--probably 'cause I read that "hippie" rag Whole Earth Review. Those folks loved it!). A minor revelation for me...I highly recommend it.
About three years ago our neighborhood association bought several copies to lend to residents as part of a planning process. They also got copies of the two other books in the trilogy, the titles of which I cannot recall. One was (IIRC) about the application of the patterns to a specific building project, and the other focused more on the city and regional design implications, but I could be wrong. Not always the most readable book(s), but well worth the effort IMO.
It was. Gamma, et al., _Design Patterns_. Everyone uses this book now.
| I think it's a sort of parallel evolution of thinking, but coming | from an independent origin.
Well, I suppose, since you can't directly connect architecture to software engineering in any but a few limited ways. But _Design Patterns_ quotes liberally from Alexander in its introductory material and is organized the same way as _A Pattern Language_. Clearly Gamma intended computer scientists to use his book the way architects use Alexander.
We have always had "design patterns" both in architecture and in software design, but they were basically just elements of technique. A wood-framed stud wall is a technique just as a binary search tree is. Design patterns, as expressed both in Gamma and in Alexander, reach deeper than that into problem-solving and use.
My sister is an architect and she's never read all of Alexander (and was surprised that I had -- it's voluminous). There are other approaches to architecture just as there are other approaches to making software. But Alexander and Gamma stand respectively as milestone works in their fields.
I've also read Susanka, and I think she has some good insight. I'm sure you'll recognize some of her ideas in what I wrote in the McMansion thread. I find her book less practical, ironically, than Alexander. As for Alexander, one has to approach him as one would an encyclopedia. One does not need to read him from cover to cover in order to make use of the principles he teaches. In fact, the later chapters dealing with construction methods didn't seem very practical by today's standards.
--Jay (who worked his way through engineering and computer science degrees as an architecture librarian)
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.