Clay Shirky’s essays are usually something that ends up in my inbox and I skip over, usually because they’re too long to rip through and move on to other things.

Today I tried to make good on my promise to myself to not put things off and read it right away. Dealing with it right away takes less time than passing it over all the time, but it makes it difficult to focus on the highest priority items.

(Regrettably I wasn’t able to follow through on my promise of actually finishing this and posting it, hence why it’s appearing more than a week after Clay’s essay.)

A number of points he made in his latest essay, “Situated Software” rang clear with me.

First, I wholeheartedly agree that when people build products for themselves or their friends, the result is often way better than a traditional reauirements-gathering, design and planning cycle. It’s like hyper-accelerated Extreme Programming. You are the beta users.

Sidenote: I worked with Clay a long time ago when his design firm worked on the never launches redesign of CRAYON. More on that in a future blog.

I’ve experienced this with some projects I’ve done at work.

The problem is that people who have this product instinct are rare. Steve Jobs is probably one well-known one. People who have good product instinct AND can actually create it are even harder to find. Hiring is hard enough. Where do you find those people?

While I may claim to know how to make a good tool for say, Geocaching, because I spend a lot of time doing it, ask me to design an accounting tool and I’ll probably get it all wrong.

There are some areas where I just don’t know if there are people that could wholly conceive and design it. The space shuttle is one that comes to mind, but I’m sure there are other examples on a smaller scale.

At work we fight this every time there’s a build or buy decision. Sure, we can write some software to do most of what we want, but might it be cheaper to use some thing already out there, free or not. But it might be more trouble than it’s worth and not meet our customized needs. Instead, the generalized nature of any repurposeable tool might make it obtuse, such as forcing you to wade through features you don’t use.

The ‘feature’ of small scale often enables something to succeed.

One last point I wanted to nod at was the coming of age of MySQL as the “apache of databases”. We use it at work. It’s certainly not perfect and some of the bugs are pretty bad, but the cost (free) has enabled us to build systems faster than before that anyone can understand. It has allowed relational databases to spread virually through a company that has been tradionally adverse to databases. This is even more suprising when many in the company including executives are refugees of Oracle.

I guess now I’ve have less excuses to put off reading Clay since he was so on target this time.