In my “9 qualities of worthwhile software“, I refer to user-friendliness as one of the qualities which makes software worthwhile. I ran into a morning battle with the bathroom faucet in my hotel room this morning which got me thinking more about the role interfaces play in the things we use all the time. All things we use have interfaces, and software is no different. These interfaces come in all shapes, sizes, colors, etc. Some seem to be easier to use than others, usually because they are more intuitive. Interfaces help hide the functional details that make a device do something useful for the user.

A faucet is one of the more universal and simple interfaces most of us interact with several times a day. In terms of functionality, they provide the user with the following

• turn water on and off
• control the amount of water flow (some might only allow on or off and no level control)
• select between cold and hot
• control the combination of cold and hot

Some “high tech” faucets might also provide the following

• auto on/off via a sensor which senses the presence/absence of the user’s hands
• specified length of time to be on in order to reduce the amount of unnecessary water flow, these are especially popular in high traffic areas – rest stops, airports, etc

So while all faucets provide pretty much the same functionality, the way in which they deal with this functionlity varies. For example

• push, pull, twist, press, turn
• separate the cold and the hot
• continuous lever which combines cold/hot and volume flow at the same time

The one in my hotel was a single handle through which I could select hot/cold and volume. The challenge for me was knowing which way to move the handle to get the hot water I wanted. So what’s the big deal? It was labeled with an H (presumably for hot) and C (presumably for cold). My challenge was I turned it in the direction I assumed was hot, waited a few seconds, wasn’t hot, so I moved it the other direction, waited a few seconds, it wasn’t hot. I repeated my test of the interface and waited a longer period of time. Still not hot and didn’t seem to be even getting warm! Being the patient person that I am (NOT!!) I went to the other sink and turned on the hot water immediately. Before leaving the questionable faucet, I decided to extend my test by leaving it run in the position I assumed was most likely hot (don’t report me to the water usage department!). When I came back, several minutes later, behold, I had hot water! So now I know which direction is hot and which is cold. 

I stood there a few seconds analyzing the situation to determine if this was the fault of the interface design. Perhaps the interface could have been improved by adding a couple of directional arrows to the H and C? But in this case I believe my challenge came not from a poor interface design, but from a performance problem! The hot water took way too long for me to be able to quickly tell if I had gone the right direction. 

My point for bit building is this – don’t focus on a single quality of worthwhile software, they must all be taken into account together. The software interfaces must be intuitive, but the software must also perform well – not too fast, not too slow. Can software perform too fast? Absolutely, for certainly classes of functionality. Error messages are good example Users get very irritated if a program’s error messages appear and then disappear too quickly. 

So when looking for good software, be sure to consider all of the qualities, don’t just focus on a single one. My job @ Worthwhile is to teach our bit builders how to balance all of the qualities into a single software product. Its sort of like putting together a great orchestra – all the stuff must be in balance with the right timing. So interface doesn’t make ALL the difference alone. It has to be properly combined with performance, security, correctness, etc. For a complete list see 9 qualities of worthwhile software.