When transacting business in a store or a hospital or an auto repair shop I always watch what happens on the computer screen. I’ve never written line-of-business software but deeply respect those who do. It must be a huge challenge to abstract the data, terminology, and rules for some domain into software that can sell to a lot of businesses operating in that domain. Of course there’s a tradeoff. Line-of-business applications typically aren’t user-innovation toolkits. People who use them learn specific procedures, not general skills. Businesses can’t be creative in their use of the software, nor profit from that creativity.

One notable exception is Fix, an auto repair shop in my town owned by my friend Jonah Erikson. Fix doesn’t use any line-of-business software, it runs on LibreOffice, GMail, and Google Calendar. That’s only possible because the team at Fix has an intuitive grasp of the technical fluencies I outlined in Seven ways to think like the web. For example, when you open a case with Fix they create a new spreadsheet. The spreadsheet will have a name like 2013-12-11-Luann’s Passat.ods. No software enforces that convention, it’s just something the front-office folks at Fix invented and do consistently. I’ve long practiced this method myself, and it’s something I wish were widely taught.

Why does something so simple matter so much? Let’s count the reasons.

First, it’s portable. The computer at Fix runs Linux but if there were a need to switch platforms the choice would not be governed by the availability of a line-of-business application on that other platform. That kind of switch hasn’t happened but another did. The spreadsheet files used to reside on a local drive. Now, I noticed on my last visit, they’re on DropBox. Fix didn’t need to wait for a vendor to cloud-enable their estimation and billing, it just happened naturally. No matter where the files live, and no matter what system navigates and searches them, two things will always be true. Date-labelled file names can be sorted in ascending or descending order. And customer names embedded in those file names can be searched for and found.

Second, it’s flexible. There’s freeform annotation within a given job’s spreadsheet. That enables the capture of context that wouldn’t easily fit into a rigid template. But here too there are conventions. An annotation in bold, for example, signifies a task that is proposed but not yet accepted or completed.

Third, it’s free. Fix runs on a tight budget so that matters, but I think freedom to innovate matters more than freedom from a price tag. Using general-purpose rather than line-of-business software, Fix can tailor the software expression of its unique business culture, and both can evolve organically. That freedom is “priceless,” says Fix’s office manager Mary Kate Sheridan.

If you were to watch what happens on Fix’s computer screen you might object that the system requires users to know and do too much. People shouldn’t have to think about filenames and text-formatting conventions, right? Shouldn’t they just focus on doing their jobs? Shouldn’t the software know and enforce all the rules and procedures?

I’m not so sure. In another of my favorite examples, Hugh McGuire, creator of the free audiobooks service LibriVox, imagined a line-of-business application for LibriVox’s readers and quality checkers. He couldn’t afford to commission its development, though, so instead he adapted a web conferencing system, phpBB, to his needs. It remains the foundation of LibriVox to this day. Had Hugh been able to commission the application he wanted, I believe it would have failed. I don’t think lack of special-purpose software hampered the formation of LibriVox’s cuture and methods. On the contrary I think use of general-purpose software enabled that culture and those methods to emerge and evolve.

I realize this approach isn’t for everyone. We need to strike a balance between special-purpose software that’s too rigid and general-purpose software that’s too open-ended. I’m not smart enough to figure out what that middle ground should look like, but I think Bret Victor is and I’ve been inspired by his recent explorations that point the way to great user innovation toolkits. Give people the right tools and they’ll be happier and more effective — not only as employees, but also as citizens of the world.