Design patterns criticism sources

Question

I was reading the design patterns page on Wikipedia, particularly the \"Criticism\" section.

Could you point me to some articles or books about shortcomings of desig

Answer 1

Alan Kay himself is very critical about patterns because he doesn't believe software should be so vaunted. Here's a 2012 Dr. Dobbs interview with Kay.

"The most disastrous thing about programming — to pick one of the 10 most disastrous things about programming — there's a very popular movement based on pattern languages. When Christopher Alexander first did that in architecture, he was looking at 2,000 years of ways that humans have made themselves comfortable. So there was actually something to it, because he was dealing with a genome that hasn't changed that much. I think he got a few hundred valuable patterns out of it. But the bug in trying to do that in computing is the assumption that we know anything at all about programming. So extracting patterns from today's programming practices ennobles them in a way they don't deserve. It actually gives them more cachet."

Answer 2

Design patterns are usually presented as a set of tricks in a specific programming language, usually Java or C++. It is usually not explained when and why a pattern needs to be used, and when it is best not to use it. It is usually not explained what would happen in a totally different programming language.

So one gets an impression that 1) design patterns come from the sky, invented by geniuses, 2) the GoF book needs to be memorized, 3) patterns need to be applied always and ever more in every situation.

In my view, design patterns are highly language specific (LISP has a totally different set of "design patterns" than Java) and problem specific (depending on the project, you use or don't use a pattern even though it is "theoretically" applicable to this place in your code). The GoF book is a useful companion to a Java language manual, but not a set of eternal principles about "programming in general".

Answer 3

Design patterns were hyped a lot about ten years ago; it seems to me that most of the criticism is about overarchitecting applications and applying all the patterns you can think of wherever you can. This heated debate is pretty boring when you take the rant factor out - yes, too much of anything is not good and to inexperienced programmer with a hammer everything looks like a nail.

From time to time, somebody will discover that something he's been doing for the whole time has a name and comment that it doesn't deserve to have a name (missing the point that design patterns are about naming sometimes obvious things so that you can talk about them).

Apart from that, you're basically left with the fact that some languages have a few patterns built-in, and some other don't, and academic debate about how with time some patterns become programming language features.

I haven't seen much valid criticism related to design patterns beside that. They definitely exist, sometimes they are useful, you don't have to know all of them when somebody wakes you up at 3 AM, and that's about it. :)

Answer 4

I am sure that design patterns help to structure educational courses on programming techniques but I don't think it helpful that they become a lingua franca in the software industry. They can be counter productive as a way of communicating design because they force the design to conform with established patterns, they allow the design to be described too casually and they force anyone trying to understand the design to read a book on design patterns.

Answer 5

Having lived through the times when the GoF book was published (ca 1995), I recall reading that a major impetus (or at least a reason for so much attention), was that OOD was still new to most developers. There was a dearth of understanding of how to make Objects at all.

So this book came out showing some common patterns of making classes and how they worked together.

So a criticism to this: it may not be as relevant now as it was then. Back then, and for a few years afterward there were not as many frameworks (especially in Java) as there would come to be in the years afterward. Now, you can find a lot of examples of good Object Oriented code. Wouldn't it be nice if some authors took snapshots of very good OO codebases aside, and wrote books about why they think they are so wonderful?

To me, the best use of these patterns is to know them when you see them, understand the drawbacks (see 'Visitor Pattern' and note which parts of the system are more easily expanded), and take those lessons as you write your own code.

Answer 6

Most of the criticisms of design patterns that I have come across relate to a distaste for the structuring and labeling of what they consider to be just good object-oriented practices. Most patterns boil down to programming to interfaces, and other SOLID principles. The feeling is that when we teach patterns we cause developers, especially junior developers to try to cram all problems into the set of patterns that they have learned, which can create more obtuse and cumbersome problems than if they had taken a more 'straightforward' approach.

I tend to agree with the sentiment that once you begin to learn patterns you tend to overuse them, however, typically you very quickly move out of that stage, and into a much more productive and professional software career afterward.

As a bonus, here is a bit of mild criticism from Jeff Atwood and some critical insights from Mark Dominus