Posted 8 years ago

Every feature I add to the website has a cost, and often the cost is too high. This is something I didn't understand at first.

When I first started work on the website, it was very much a small hobby. I had an interest in making websites with HTML, but I knew very little; and web development was completely different back then too.

And so, naively, I piled on features. I added so many things that made the website more complicated.

A lot of these things were technical, and I promised I wouldn't talk about that stuff so much. So without going into details, let's just say that I made the task of creating and maintaining the website harder, simply because I liked adding clever features and following "best practice". Here are some that I implemented or at least toyed with:

• An "elastic" layout that zoomed with the text size
• Text resizing widgets for accessibility
• Alternate colour schemes
• Access keys (for keyboard users)

All of these are features that at least some people might appreciate. And while many of them seem quaint now, at the time they were all considered "best practice" by many in the web design community.

The trouble is, features have a cost. Often they have a cost to users of the site, because adding features usually makes the site more complex to use: I have to highlight the feature somewhere so that people realise it exists, and doing so is a distraction for people who don't use that feature.

They also have a cost to me. Each feature takes time to build, and more importantly takes time to maintain. But there's also a more subtle cost sometimes: an opportunity cost, the cost of what I could have done without the additional restrictions imposed by a feature.

Does anyone remember the "sources" section that used to live at the bottom of article pages? The idea was to list my sources, in the style of an academic paper. That way people could be suitably impressed by the thoroughness and correctness of whatever I said. Or more usefully, perhaps it would let them decide on a case-by-case basis whether to listen to my advice.

It seemed like a good idea at the time, but it was hopelessly unrealistic. Badminton coaching is nothing like academia. As time went on and I became more experienced in coaching, I realised that I needed to use ideas that did not come with multiple references from authoritative sources.

Having the sources section imposed an unwelcome restriction: I couldn't say anything unless I could back it up with convincing sources, and I certainly couldn't say anything original! Why the hell would I place that restriction on myself? What was I thinking?

And of course, adding the sources was Yet More Work. It was ultimately motivated by a lack of confidence in the credibility of my own ideas --- a hallmark of academia, which stifles creativity among the dusty exhaustion of established thought.

So I dumped the sources feature. And no one has ever complained about that. It vanished without a trace, and no one missed it.

It's easy to confuse my concerns with the concerns of my website users or customers. This is an extremely common mistake, but it's much easier to spot when someone else is making it. With hindsight, the sources section didn't help my users. It wasn't for them; it was for me! Its sole purpose was to make me feel more credible and boost my confidence.

Often this stuff seems obvious from the outside. But that's the whole point: you're only prone to these kind of mistakes when you're on the inside, when it's your creation.

I have tried to learn from these mistakes. I try not to add features unless I'm pretty sure many users would benefit from them. One feature that I would really like is a forum, because that's a great way to gather feedback and community. A forum would help me make decisions based more on user feedback than just my own instinct.

Another feature I would like to add (or restore!) is Paypal support. Why? Because plenty of people have actually asked for it!

Removing features is good. It helps you focus on the most important things. Better yet, don't add the feature in the first place unless it brings a clear benefit to the user.