Perpendicular Angel Design

I’m thinking about writing a book, and the ^[1]nonfiction book I’m thinking about writing is about building design systems and making sure they’re accessible. ^[2]The list of fiction books in progress is embarrassing.

It’s a big undertaking. ^[3]Building a design system, I mean, not the book. Although I think the book will be a big undertaking too — it’s just that books tend to be finite creatures that you can finish, and design … Continue reading

It’s an undertaking that I’ve been involved with for three different companies now — as a consumer of the system at Vanguard, and as a designer at both Boomi and Vertex.

But really, I wouldn’t have made the jump from consumer to designer successfully at all if Jeremy Keith hadn’t given a talk in 2018 called The Way of the Web. ^[4]It is not one of his best-known talks and for that matter if I hadn’t found these sketchnotes I’d really be wondering if I made it up right now.

In that talk, Keith summarized the concept of pace layers, presented by Stewart Brand. It surmised that culture builds from a bottom layer of nature (which changes very slowly) through multiple layers until it reaches fashion (which changes almost too rapidly to track), and that by understanding the relationships between the layers, we can better understand how and where to effect change.

Keith’s suggestion in his talk was that a similarly-structured pace layer could be built from TCP up through Javascript and other scripting libraries, from the slowest changing layers of the internet through the fastest.

At the time I was prepping for my own talk so the idea kind of grabbed on to the back of my brain… and there it started gnawing on things.

Building a design system is often challenging and frustrating because some things about it (such as, for example, the organization’s voice and tone, or the corporate font) change rarely, but have huge impacts. Others, such as the fashion of flat design or the implementation of specific interactions, change frequently but only impact small subsets of the full design system. As a designer, I’ve found that understanding the difference between scheduling a change to the corporate font and scheduling a change to the 4th carousel owned by the 2nd subdivision of importance in the organization is the key to ensuring that the design system is given the budget to grow — and the key to ensuring that the design and development partners don’t want to string me up.

But to explain all of that, first I had to be able to explain pace layers. And while I was pretty sure I did understand them, if one wants to start writing a book it’s a really good idea to read the source material before running one’s mouth.

Then I bought the wrong book.

See,  my brain is a muddled mess of memories from various conferences and web events. Learning about pace layers is definitely one that sticks out. Another that sticks out is discussion of the Long Now Foundation.

This foundation is dedicated to taking the long view on time. The very long view. Like, a clock that runs for ten thousand years long. People who place bets that take a minimum of two years to prove out true long. Storing a sample of all the world’s languages in nickel long.

And while The Clock of the Long Now does, indeed, discuss pace layers in chapter 7, the bulk of the book is about taking the long view on what we put into the world. In the case of web design and development this can mean everything from choosing HTML as your primary markup language (and not coding everything in JavaScript) to deciding which companies and projects you’ll still be proud to be associated with 20+ years into your career.

(I also remember a lot of conversations about how future forward design is progressive enhancement from older technology, and a lot of men getting their pictures taken in a space helmet, but once again the terminology to find proof of these things eludes me tonight.)

It turned that reading The Clock of the Long Now was not a mistake on my part (more of a lucky accident) because it’s helped me understand and maybe even figure out how to explain that one of the biggest challenges with a design system is that it’s a system. You’re designing a system that, if successful, cuts design and development time for the whole organization. But that means you’re also designing a system that will require time to develop, time to adopt, time to grow, time to stress-test. It is not a short-term two-quarter project with a payout at the end.

In other words, it works in long time, but in corporate-driven show-me-the-returns time, it may be hard for our Product people to buy into.

One of the amusing things about the book is that the edition I read was published right after the turn of the millennium (or the year 2000 for you pedants, don’t @ me about when centuries start), which is itself a millennia in web years. The Long Now Foundation still exists, which is a bit of a relief because basing my ideas on design systems on a long vision that subsequently died in the 20 years since the book was written would’ve been embarrassing.

Brand did seem to go out of his way to not make predictions about what the future would bring, outside of the fact that it 10,000 years from now life will likely be very different from today.  Still, there was more than one point where something said would pull me out of the book. Brand seemed quite optimistic, that the web would have the power to prevent people in power from spreading lies, for example, at which point I almost spat my drink across the room.

The book’s thesis, that we should take a long view if we want to work together to solve problems, is still valid 20 years later, and has helped me better shape the kinds of discussion points I’ll need to get the funding necessary for a quality design system. (I’m pretty sure Brand didn’t see that coming, but then, he probably wouldn’t be surprised either.)

So it’s a good, relatively short read, not directly about the web but definitely about the underpinnings of being a developer or a designer, and I recommend it to anyone interested in the longer view.