Joel is wrong

Like many of you (I think, and possibly hope), I've read Joel of Fog Creek Software's blog on occassion.

The man tends to be correct. Every now and then though, the man is just completely and totally wrong.

I remember a blog post where he advocates hiring the best programmers exclusively. What on earth is wrong with the man!?

He's wrong on two levels:

• Hiring only "the best" creates a terrible culture of elitism and exclusivity. Do you really want to make the cost of hiring escallate that much as time goes on?
• "The best" are, by definition, an extremely finite resource. You want to cut down your potential workforce and then make it all the more difficult to integrate them?

Good grief. These two facts are actually the lesser of your problems.

Let's assume you get a team of 4 brilliant developers who write amazing code, backed by perfect design. Your problems have begun.

Over the next 5 years, all 4 of your developers move on. Face it, they're "the best" - if you've only got 4 developers what are the chances you can keep them interested? What happens when you replace them? Are the replacements going to be equiped to handle whatever they produced in terms of maintanance? Will they be able to support your sales staff, support staff, technical writers, strategists and other stakeholders?

These are questions that need to be asked - Joel doesn't seem to cover them all that thoroughly at all.