Friday, August 5, 2011

It's (Almost Always) Not About You.

I've got a few minutes left in my lunch break, and have spent the last 15 minutes or so perusing a few user experience/interface blogs and websites. Some recurring themes are "keep it simple," and "provide feedback," among others. All good advice.

One site also mentioned that the language you use should be "conversational, not sensational." If you've happened across my previous blog entries where I talk about news headlines and how much I abhor them most of the time, you'll know that this strikes a chord with me. Users don't want software that puts them in a hammer lock, sweeps their legs, and pins them to the ground until they finally gasp, "Alright, you got me, this is awesome!" (Well, maybe some users do. I haven't met them, though.) Users are independent and, for the most part, capable of determining whether a piece of software is worthwhile or not, so you can avoid telling them.

So why is this overbearing self-hype so prevalent these days? Human nature. Pride. Whatever you want to call it. "I built this program/website/app, it's the best in the world, and you should use it to replace whatever you've been using" is a common mentality among developers. That may be an exaggeration, but when someone has spent a large chunk of time working on a project and is finally releasing it to the public, there's a certain amount of desperation for acceptance and recognition.

I'm guilty of this as well, of course, and the more projects I work on the more I see these principles in action. My summer has been spent coding up a concept my mentor/boss thought up for improving on a piece of optimization software. Goal #1 is accuracy. Since you either get the right answer or you don't, that's an easy one to nail down and meet expectations on. (Easy in the sense that determining success is straightforward. Getting correct results is certainly not an easy task.) Goal #2 is speed. The faster you can get the right answer, the better.

I spent the first couple months of the summer getting my head wrapped around the problem and rewriting the software to include the new concepts, and the last few weeks have been dedicated to small performance tweaks and testing. The first tests I ran were promising. Huge gains in performance, same correct answers, life was good. I could see myself standing up for my presentation to the department and saying, "This version is better. Period. It reduces time by 50% and gets the same results, and as you add more computers to the mix, that gain continues to increase. This is a game changer."

That's not what happened though. The next test I tried, my version ran slower than the previous one. It turns out, there are some tradeoffs taking place, and that the new solution works better on a specific subset of all problems it is written to tackle. Now I need to reign in my excitement and be objective about fully testing the software to determine what its strengths and weaknesses are.

That's a long, meandering way of getting to my real point. If you're presenting something you've worked on, be it software, a business proposal, a work of art, whatever, be sure to take a step back and be realistic about what you have in front of you. If you're unable to do that, have someone else do it for you, giving them permission to be perfectly honest. I'm terrible at this, so I know it's a big challenge. Your project will be received much more readily if you aren't oblivious to its shortcomings. (Hint: using copious amounts of superlatives to describe your project is a great way to imply it's not as good as you say it is. That turns people off.)

Why? Because when you're releasing your project on the world, it's not about you anymore. It's about your work and whether it can stand on its own. Let your audience or customers decide on their own.

Users appreciate arriving at a website and seeing a homepage with a welcome message simply stating what the site is meant to do. Readers take headlines more seriously if they're simple and appear unbiased. Audiences love when work is presented realistically and without misguided statements about how revolutionary it is. Well, at least I do.

One small caveat: If you actually have the best software or the most useful website or the most incredibly revolutionary research to present to the world, and you can't adequately describe it without spewing forth superlatives, have at 'em.

Just don't be wrong.