When I started out programming the prevalent idea, which I shared at the time with many others, was that an artistic view was not going to be any part of the work. However, after a number of years in the business I began to come across moments of wonder when either I saw a great piece of coding or, very occasionally, managed to create something myself that hit the ‘sweet spot’. It was not until I happened upon Christopher Alexander’s work on patterns that I began to understand some of what was happening during these moments.

My introduction to the patterns movement occurred when reading the book Design Patterns written by the “Gang of Four”: Gamma, Helm, Johnson & Vlissides, this becoming a standard reference text. In trying to better understand the patterns vision I read some of Richard Gabriel who has some interesting ideas about the relationship between art and software. He has even come up with the idea of a Masters in Fine Arts in Software.

In Alexander’s earlier architectural patterns book he defines a library of external geometric entities to be used as design guidelines for buildings, for example: an alcove for chats that is separated off from a corridor. It is in his later masterwork: The Nature of Order that he describes his underlying ideas about ‘living structure’ and his thoughts about the perception of ‘goodness’ in design.

Alexander does not shy away from the moral dimension of his work. In a keynote speech he gave to the OOPSLA’96 conference in San Jose he stated that:

“One of the things we looked for was a profound impact on human life. We were able to judge patterns, and tried to judge them, according to the extent that when present in the environment we were confident that they really do make people more whole in themselves.” OOPSLA’96 keynote.

And later in the same talk:

“The pattern language that we began creating in the 1970s had other essential features. First, it has a moral component. Second, it has the aim of creating coherence, morphological coherence in the things which are made with it. And third, it is generative: it allows people to create coherence, morally sound objects, and encourages and enables this process because of its emphasis on the coherence of the created whole.” OOPSLA’96 keynote.

But how can we judge what is coherent? To understand Alexander’s approach we have to read the first book of ‘The Nature of Order’ series where he describes the ‘The Mirror of the Self’ test.

The Mirror of the Self

To develop this judgement of coherent living structure, Alexander identifies what he calls the ‘Mirror of the Self’ test. He highlights that there is a difference between what he calls ‘apparent liking’ and ‘true liking’. For example, when deciding which of two objects are liked the best, rather than accepting a quick ‘apparently liked’ judgement he asks for a ‘truly liked’ judgement:

“…which of the two objects seems like a better picture of all of you, the whole of you: a picture which shows you as you are, with all your hopes, fears, weaknesses, glory and absurdity, and which – as far as possible – includes everything that you could ever hope to be. In other words, which comes closer to being a true picture of you in all your weakness and humanity;…” Nature of Order: Book 1. p317.

Using this idea he has found that it is possible to have a high level of agreement (80-90%) between people when using their judgment to identify living structure for objects. So it seems that how we phrase the question is all important.

A Reappraisal of the Software Patterns Movement

So far the software patterns movement has tried to abstract out particular solution patterns to be used as guidelines when designing software structures. Despite the best intentions it has degenerated into being a set of document templates, rather than embodying the wider view of Alexander’s work. Once again we have become hooked on a results-oriented view of the world as if we can only feel comfortable with this approach in such a technical domain.

Erich Gamma, one of the co-authors of the Design Patterns book, said that referring to patterns is most useful when we already have a specific design ‘pain’ rather than trying to force patterns onto a particular project from the outset. This points to the fact that we cannot get away from being conscious of how we develop our judgement. How do we even identify that we have a design ’pain’ if not through discerning human judgement and a sense of rightness?

Along with other commentators like Jim Coplien, I consider that Alexander’s vision of patterns (the drive towards living structure and the big question of making human life more whole) has not been truly realized within the software discipline. We need to revisit the Alexandrian roots of the patterns movement and understand how these roots relate to software development.

In Alexander’s OOPSLA’96 talk he identified 3 key points in his vision for the patterns work: a moral component; coherent designs; generative process. Although there has been some discussion in the software community about Alexander’s later work, it is fair to say that it has been difficult to take these ideas further in the domain. However I have found that by connecting the ideas with those prompted by reading Bortoft and early Steiner we can get a bit more clarification which I will report on in my next post.

Thanks for reading so far and I wish you all the very best for 2014…