David Malouf wrote:
> At the heart of Alexander's work is that Patterns ARE. They are  
> primarily a
> description of what already exists, but then through conversion of  
> being
> described as a pattern is canonized. They are not rules for what  
> you have to
> do, but rather suggestions for how previously similar "problems"  
> have been
> solved "with success".

I agree that they're not at all rules, but I disagree with your  
characterization of his purpose: it's not that Alexander's patterns  
ARE existent in architecture, it's that they're what he thinks  
architecture SHOULD BE based on. He was citing examples of patterns  
that were consistent with his particular social and political world  
view, things that he thought were good. They're very idiosyncratic:  
they're incredibly poetic and spiritual, very socialist (even in many  
cases communist), and almost never based on any sort of empirical  
argument. Many of them are even, in my opinion, dead wrong. And they  
are far from comprehensive -- he doesn't describe lots of very common  
architectural patterns, such as how to house a prisoner, where a CEO  
should sit with respect to his employees, how to structure a sales  
floor to move lots of merchandise, or how to instill fear in  
religious worshippers.

I don't even think of them as "suggestions". Alexander's pattern  
language is a very particular cultural manifesto, which taken as a  
whole comprises a way of building that he and his followers think  
results in great architecture and shiny happy people.


> Patterns further can be innovated, they can be nested and scaled.
>
> To answer Christian. I "believe" in patterns, but am uncertain to  
> how to
> best make them work in a living real workspace given my experience.  
> I fall
> between the line academic who loves "pattern recognition" and a  
> "gotta get
> it done" person.

My take on patterns is this: They should usually be thought of as a  
"language", not as a "reference library" or even worse as a "solution  
locator". You don't take a problem you have (say, "I need to present  
the user with a small number of multiple mutually exclusive  
selections") and then look up the pattern that best matches the  
problem (in this case, a "radio button"). This just isn't how most  
people work. Instead, you immerse yourself in the language of your  
field and then when you face a problem the solution can be found in  
your head, in your learned vocabulary, accessible in the same way  
that a word might be accessible when you are forming a sentence. The  
advantage of assembling a collection of patterns into a book like  
Alexander's is that in a single sitting a reader can imbibe the full  
breadth of his language, and can later look back to the book when  
they need inspiration.

When a pattern is put into a library for reference, it becomes,  
functionally, a style guide. I think of Yahoo's library as a style  
guide when I go there for an answer to a particular problem, but I  
see it as a pattern language when I go there simply to read through  
it and absorb the holistic view of interaction design it embodies.  
It's also something of a dictionary or thesaurus so you can see  
proper usage within a given style idiom.

But the overarching language of interactive design should be seen as  
something fluid and hard to pin down, much like a real language. The  
whole world is a pattern language. But it's great when people like  
Alexander, and organizations like Yahoo, attempt to carve out a set  
of it as a good and consistent way of doing things. Just keep in mind  
that it's just one way of doing things.

-Cf

Christopher Fahey
____________________________
Behavior
biz: http://www.behaviordesign.com
me: http://www.graphpaper.com

________________________________________________________________
Welcome to the Interaction Design Association (IxDA)!
To post to this list ....... [EMAIL PROTECTED]
Unsubscribe ................ http://gamma.ixda.org/unsubscribe
List Guidelines ............ http://gamma.ixda.org/guidelines
List Help .................. http://gamma.ixda.org/help

Reply via email to