Tools matter

Banging my head since Thursday against a UX job I’ve been working on in OmniGraffle (my go-to design tool for the last six or eight years — before that I was a pretty solid Illustrator person).

Today I started the project over in Sketch and accomplished more in 2 hours than I did in the previous 72. Despite the fact that Sketch is arguably inferior to the task. I miss Omnigraffle’s magnets and marginally superior text handling, for example.

But Sketch provides me with a single, infinite blank canvas when I start, and that one UI design decision makes all the difference to my workflow. As I work I like to push objects around, clone them, move old ideas to the margins, delete bad ideas, define artboards and work areas on the fly, zoom in, zoom way in, zoom out, zoom way the fuck out. I expand artboards for big ideas, shrink them for little ones, lock them when they get solid. My work habits are spatial, and design tools that demand I define my workspaces as the very first step immediately disorient me. How do I know what I need before I’ve even started? That means I have to design something in my head before I commit a single pixel.

I think I prefer Vim to an IDE for the same reason. IDEs always want me to define a “project” before I start. Ditto my distaste for all-singing-all-dancing frontend frameworks. And CMSes that insist I want a forum, blog, comments, and workflow before I even start building the site. Give me a blank canvas.

In grad school I had a habit of getting a photocopy of every citation and then spreading them physically around the room, then grouping them into piles, tearing out the best pages, marking them up etc.

I don’t doubt lots of people work the other way around. People who won’t get into the car until they have a solid idea where they’re going. They make good use of the boost they get from templates and frameworks and functional proofs-of-concept. Good for them.

Tools matter.