Is Good Software the Same as Valuable Software?

Posted by Andy Singleton on Oct 27, 2007 5:12:00 PM

I am not sure I agree with this 37 Signals post: Is it really the number of features that matter? On the face of it, it seems impossible to argue with. You need to edit your software so that it includes the features that people will actually use and understand, and removes features that are just confusing. That approach will give you better software, from the point of view of the user. But, will it give you more valuable software?  Not necessarily.

Saying that you can make better software by including the "right" mix if features is a tautology.  The real meat here is in the strategy for finding the "right" mix.  The recommended strategy is to "edit" the feature list to solve 80% of the problem for 20% of the effort.  I do that too, because it's a good way to get software out fast.  But I am willing to admit that this is a strategy to aid the developers, not a product strategy which is necessarily optimal for users.  Is there a difference between the 80% definition of good software, and truly valuable software?

In a lot of cases, the value is in the extreme. Would a Porsche be good if it only went fast enough to satisfy 80% of the drivers? Would Exxon be the world's most profitable company if it only drilled to 80% of the maximum possible depth?

The Segway scooter has an elegant user interface (you push the stick in the direction that you want to go) and is by all accounts easy to use even for beginners. On the other hand, it takes months to learn to drive a car fast. But on the highway, I will run over your Segway over with my Porsche, and by the way, my vendor outsells yours by hundreds to one.

The argument for simplicity falls apart pretty quickly when applied to the modern work environment. If you are the CEO of Exxon, you have 123,000+ employees. You have More. A Less is Less philosophy is an irrelevant fantasy. You also have ridiculously complicated systems and processes that cost billions of dollars and allow you to drill a hole in the gulf of Mexico five miles deep. For dealing with this, you are extremely well paid.

I’m going to guess that you and I would consider much of the software used at Exxon to be pretty crappy. But it’s valuable.

Topics: development process, project management

Follow Assembla

Get Started

blog-CTA-button

Subscribe to Email Updates