We were working with a potential client a few weeks ago, trying to figure out if we could help them improve some seismic processing software. The software had excellent science under the covers, but the visual interface was old and tired. Could Palladium help rejuvenate their user experience? Old looking software can imply old or out of date capabilities. Could we make it, well, better?
I asked if we could look at a seismic panel as a wiggle plot. So we loaded up a dataset, created a panel, typed in an inline number, typed in a cross-line number, and saw a huge blob of black. Hm. Now let’s find the traces-per-inch setting and make it smaller. Now just tiny lines. So of course then we updated the gain so we could see wiggles.
(If you’re not familiar with seismic data, this is a bit like asking to see a Google Maps street view by bringing up dialogs for exact latitude and longitude, then an explicit scale for your view (10 pixels == 100 feet?), and choosing a good orientation, like 32.5 degrees)
The potential customer asked if we could make the buttons look better. You know, freshen up the look and feel.
Don’t get me wrong — getting whitespace and buttons right isn’t always easy, and it can make a big difference. I’ve used applications with buttons so big they looked like ads. Invisible, even when you’re looking for them. And applications where obvious operations were so small, or so hidden in menus, that I couldn’t make progress.
We might make the buttons a little nicer to click. But nice buttons weren’t what they were missing. What was the actual use case here? Quickly bring up a detail from a large dataset so we can “look around”. The way I’d expect this to work now would be to load up the dataset, immediately see an overview map (with lots of nice detail from, say, Google Maps) and be able to click anywhere to generate an instant panel along an inline. And that panel would use AGC (automatic gain control) and some basic heuristics to make a view that was instantly usable. We’d have worked with a few users to figure out what a reasonable trace spacing was, what a sensible default color bar was, and how big wiggles should be.
In other words, to make this app seem current and usable, we couldn’t just make some buttons simpler, or bigger, or smaller. We had to add a map and do some user studies.
The end application would have actually looked smaller and felt simpler, even though it required quite a bit more under the covers. As it turned out, the potential client wasn’t ready at that time to engage in that sort of re-work. It’s not always the right investment, after all. It would have been a good project to work on, and right in our sweet spot. Perhaps they’ll get some budget next year to work on the problem.
Next time you look at making a particular dialog nicer, or tweaking a wizard to work more smoothly, take a step back and try to figure out if it’s not nicer buttons you’re missing. Maybe it’s a whole map you’re missing.