Peter Seale's weblog

Wednesday, February 27, 2008 10:00:17 PM UTC 

"Learn at least one new language every year."

    -Andy Hunt and Dave Thomas, The Pragmatic Programmers

For a while now many of us in the .NET space have complained of the never-ending torrent of new technologies and frameworks coming from Microsoft. While excited by the potential benefits (and yes, the shiny bits), we're so inundated with NEW that it's impossible to keep up. Something has to give.

Arrival stage left: the Pragmatic Programmers, circa 2000/2001, who inform us in their "Your Knowledge Portfolio" section that in order to be well-rounded developers, in addition to the learning we do related to our job, we need to set a goal to learn at least one new programming language every year.

What's unfortunate about this whole situation is that I actually think this advice is harmful.

One language a year: unattainable

No one learns a language a year, year after year. Except Ted Neward, and let's face it, he's nuts, and therefore doesn't count. So why should we set it as a goal? So we can fail, unless we're Ted Neward? GREAT.

Now I'm depressed

Why set unrealistic goals for ourselves?

Why a year?

Why not 18 months? Am I still broadening my horizons if I take 18 months to learn Lisp? or 10 years to learn C++?

Alternative to '1+ language/year'

Instead of hyperfocusing on '1+ language/year', why not take other advice offered by the Pragmatic Programmers on the very same page? I particularly like the way they present learning as an investment [the following is paraphrased]:

• Invest regularly
• Diversify portfolio
• Balance portfolio for risk
• Buy low/sell high (get in on technology early)
• Portfolio should be rebalanced periodically.

See! All this is useful, spot-on, helpful advice! And what's more, I don't even feel a sense of crushing despair!

Generalizing specialists

Let's work this from another angle: on this page Scott Ambler discusses what it means to be a generalizing specialist. The summary, by example, is that instead of sticking your head in the sand and learning SharePoint Designer 2007 to the exclusion of all other things (i.e. becoming a 'specialist'), you make a conscious effort to learn things that are NOT related to SPD 2007, however lucrative it may be at the time.

Nowhere on the description of generalizing specialists is there a mention of yearly language learning quotas.

So what language are you learning, Peter?

I'm going to make a conscious effort this year to learn C#. Wait a minute dude, I'm supposed to know C# already! Well, sure, yeah, but I would like to, you know, REALLY know C#. I would like learn C# such that I'm effective with it—such that when I read Bill Wagner's Effective C#, I'm not surprised by any of his 50 discussions of C# programming. Think about it.

I want to learn all the C# 3.0 features, aside from LINQ. I don't even honestly know if there are other 3.0 features, oops. Wait, extension methods. Awesome, I didn't totally forget.

I'd like to learn C# to the point that the very sight of Java code disgusts me. I want to see an Eclipse window and feel the physical urge to puke. That's how awesome at C# I want to be.

I'm not going to promise I will be C# dominant by EOY 2008. It may take longer. But I will work at it, on a regular basis. And I will get better.

And I think this is all the Pragmatic Programmers ever intended in the first place: pick a language; learn it and gain some perspective; pick another language; absolutely don't stop learning after your first language; don't take too long to learn the new one; constantly learn. There we go; that wasn't so bad.