Are you an A-list programmer? Do you care?

Programmers are a diverse lot. Not merely in gender, ethnic background, or age, but in programming talent. Tests of programming skills show a large difference between the best and worst programmers: some tests report a factor of twenty-five.

Such a large difference may offend the sensibilities of some managers. (Especially managers in organizations that consider programming staff to be interchangeable cogs.) Yet I think that we can agree that some programmers are better than others. I classify programmers into three tiers.

The A-list programmers are those few who are superb at the craft. Not only productive, they are knowledgeable and creative. These are the folks who create Twitter and FaceBook and Amazon.com. They do not fit in to large, stodgy bureaucracies; the only large companies they stay with are the ones that they found. These folks work with the cutting edge languages. Today, that list is: Haskell, Scala, Lua, and perhaps Python. These programmers are mobile. If the A-list programmer does not like his current work, he moves on to something more appealing.

The B-list programmers are hard-working and passionate, but they work at a less entrepreneurial level than the A-list folks. They don't found companies, but they do have lots of ideas and creativity. They don't use the cutting-edge languages, but do use modern languages such as Ruby, Python, Objective-C, or Perl. They are not as mobile as the A-list programmers, but they do move from opportunity to opportunity.

The C-list folks are the determined programmers. They are not entrepreneurs, and they don't have ideas that are burning to get out. They are quietly sensible, and know that the paycheck is there to pay the rent. These folks avoid the (chaotic) start-ups and are comfortable with the (predictable) large bureaucracy. They use the languages that the corporation deems "acceptable": Java, C#, or maybe C++ or Visual Basic. (The last two are from legacy systems which must be maintained but not converted to new languages.) These folks move rarely, being risk-averse and preferring the known devil to the unknown demon.

The difference in performance matches the risk tolerance of the company. Small start-ups require talented performers, and cannot survive mediocrity. Large corporations find it difficult to tolerate creative folks who break the rules, and prefer conformance over performance.

I am sure that there are exceptions to these broad categories. The correlation of performance and language selection is yet to be shown, and some projects will require the use of an older language. But I think this alignment is pretty accurate. The top performers are where they are because they pick the best tools -- and languages -- for the job. The worst performers know only one language, try to solve every problem with it, and have picked a language that is "safe" (one that matches lots of job openings). They are not on the cutting edge of programming.

Where you fit into this scheme is, for the most part, up to you. If you are willing to take risks and are creative, you can be an A-list programmer. (You need to be a good programmer, too.) If you prefer a safer solution, a large company with older technology is a better fit for you. (Although I don't know that larger companies are safer, in these days.)