Comments for Michael Groner's Blog

Comment on MapReduce talk from Lambda Lounge Presentation by Introduction to MapReduce: Q&A with Michael Groner

Introduction to MapReduce: Q&A with Michael Groner — Mon, 21 Dec 2009 15:23:19 +0000

[...] gave a talk on the MapReduce framework. The 5-minute Lightening Talk was a condensed version of the 45-minute presentation he delivered originally at the St. Louis Lambda Lounge (a local group focused on dynamic and [...]

Comment on How TDD clarified the Open/Closed Principle for me by michael

michael — Thu, 29 Jan 2009 18:25:53 +0000

Brian -

Thanks for the question. It really got me thinking.

I hope you don’t mind, but my response grew into a blog post of its own. Please take a look and let me know what you think: http://bit.ly/6mZ0b

- Michael

Comment on Which principle was it? Open-closed or dependency inversion? by zyle

zyle — Wed, 30 Nov -0001 00:00:00 +0000

..and I just had an epiphany, and realized that the STL “for_each” may be another example where OCP seems to apply without the use of base class interfaces; different functors can inject behavior (as you term) it, to perform different tasks, without changing the for_each() library code itself.

Of course, you could argue that even though dynamic polymorphism through a base class isn’t being used here to adhere to an interface, we are in a sense using static polymorphism to enforce the template contract (i.e. for the functor being passed in), but still….

Comment on Which principle was it? Open-closed or dependency inversion? by zyle

zyle — Wed, 30 Nov -0001 00:00:00 +0000

I’ve been reading Robert Martin’s “Agile software development” book, and found myself struggling over the difference between OCP and DIP. No matter which way I looked at it, they both always seemed to converge on the same solution i.e. use abstractions rather than direct dependence. The statement you posted, i.e. “OCP states the goal of not having to modify a class to extend behavior. DIP provides an implementation guideline that we should depend on abstract classes as much as possible.” clears up a lot. Thanks.

Comment on Which principle was it? Open-closed or dependency inversion? by Matt Campbell

Matt Campbell — Wed, 30 Nov -0001 00:00:00 +0000

Brian Button recently posited on Twitter:

OCP in 6 words – Structure systems around abstractions, not concretions

To which Robert Martin replied:

That’s also a reasonable definition of DIP.

So no, you’re not alone in thinking that there’s some heavy overlap here.

Comment on Which principle was it? Open-closed or dependency inversion? by Zac

Zac — Wed, 30 Nov -0001 00:00:00 +0000

To me, OCP has always seemed like a direct effect of adherence to the other SOLID principles. Kinda the inverse of testability, OCP is a litmus test of the flexibility and extensibility of your design rather than a guide to achieving that flexibility and extensibility.

I’m probably wrong, but that’s how I see it.

Comment on How TDD clarified the Open/Closed Principle for me by Brian Button

Brian Button — Wed, 30 Nov -0001 00:00:00 +0000

Michael,

I’m glad you get these principles. I learned them from Bob in the very early ’90s and have held them very dear since. But in this case, where the act of injecting behavior and data into the classes being tested is the effect you’re seeing, does this seem more closely aligned with the DIP?

The SRP drives you down the road of creating small, coherent, but loosely coupled classes, while the DIP teaches you how to keep that coupling loose.

Admittedly, I usually think of the OCP from a statically typed language point of view, but even when viewed with an eye towards ruby, injecting behavior seems to favor the DIP.

What do you think?

– bab

Comment on How TDD clarified the Open/Closed Principle for me by Michael Groner

Michael Groner — Wed, 30 Nov -0001 00:00:00 +0000

In my example the “DummyList” was a dummy object, not mock class. That way I was sure I had removed any dependencies on concrete classes. Python shielded me from having to build abstract interfaces with separate implementations but it would be necessary in C++ and Java.

I wanted to go into more code detail in the blog, but was concerned it was getting too long already. If there is some element of my post that you like to see in more detail in a later entry, please let me know.

- Michael

Comment on How TDD clarified the Open/Closed Principle for me by Robert 'Groby' Blum

Robert 'Groby' Blum — Wed, 30 Nov -0001 00:00:00 +0000

I might be confused, but isn’t DummyList just a mock for your list?

Or do you mean you implemented DummyList so it only responds to the necessary methods? (In that case, I’m SOL – static typing…

Comment on How TDD clarified the Open/Closed Principle for me by Victor noagbodji

Victor noagbodji — Wed, 30 Nov -0001 00:00:00 +0000

Great article. It would be nice if you write some tutorials on TDD for the rest of us.