The Code Less Travelled


Consider all the technologies that you haven’t used. Of those, which would you really rather not use? Why?

I made a list recently, and it had a lot of items. There were languages (Scala, Perl, Go), web frameworks (Spring, Yii), source control systems (CVS, Darcs), and more besides.

I’m actually concerned by this. It’s very hard to form a well-informed view if you’ve never even played with something. Once I had my list, I grouped the items according to the reason I didn’t want to use them.

My most common reason was what others say about it. For example, some programming language implementers that I respect are critical of Go’s design. However, my sphere of influence is heavily skewed: many of the bloggers/twitter users I read have similar technology backgrounds. It’s unlikely that they’ll be the ones able to say what I’m missing out on! I spent some time looking at Go, and it does some very impressive tooling (I’m not aware of any equivalent of go fix anywhere else).

A related reason is reputation. Isn’t Spring just a framework for huge ugly enterprise codebases? So I’m told. However, many people use it, many projects have been created with it, and I am totally unable to say what Spring’s advantages are. It’s a blind spot.

I find it easy to dislike technologies that are similar to those I’m very familiar with. Why would I want a new web framework when Django exists already? I’ve written a lot of code using Django and I’m very productive with it. This is a sunk time cost, and it’s easy to resent the additional time required to reach the same skill level with something else. As a result, it’s hard to be open-minded and fair to new tools.

I’ve been forced to use other web frameworks when working on established projects. I’ve often found things that my Django projects lacked. For example, the Pyramid debug toolbar opens an interpreter if you have an unhandled exception. JavaScript frameworks often have great asset pipelines with auto-reloading (which Django doesn’t provide by default).

On the other hand, sometimes I dislike technologies because they’re too different to what I’m familiar with! For example, Darcs is a VCS that treats a repository as a partially ordered set of patches. I’m forced to learn the new conceptual models before I can understand the documentation and get things done. This effort requirement often feeds into an ill-informed dislike.

Note that I’m not saying that all languages are created equal or using badly-designed tools is ‘character-building’. If a technology has actual people doing real projects with it, we must understand the advantages, preferably from modest first-hand experience, before drawing conclusions. We must strive to be intellectually honest.

I’m also not saying that technology is a treadmill where we must keep learning new tools or risk an obsolete skillset. You can pick technologies that are sufficiently entrenched (e.g. C, JavaScript) so your experience will still be useful in five years time.

Instead, we should keep exploring, and be mindful of our blind spots. Somewhere on the web, there will always be an article saying $LANGUAGE is bad, giving us an excuse to miss that learning opportunity.

Stay open minded. No language has all the good ideas. Don’t miss out by only being a $LANGUAGE person. For 2015, I’m resolving to do some side projects using the tools in my dislike list.

Recent Posts

These Weeks in Remacs

This Week in Remacs

Announcing Remacs: Porting Emacs to Rust

Introspecting Glue Code

Searching A Million Lines Of Lisp