Maps and Territory

Earning knowledge through experience

Justin Andersun

Aug 8, 2023

I have this arbitrary goal to climb the highest point in each of the fifty United States, and enough other people share this arbitrary goal that it has a name: Highpointing. For the last ten years, highpointing’s taken me to all corners of the country—from a boulder in someone’s backyard to the tops of frozen volcanos—and, most recently, to a grizzly-infested mountain range.

Granite Peak towers 12,799 feet in the remote Beartooth Mountains and is the highest point in Montana. Two weeks ago, I ventured in with my dad for a three-day backpacking trip, hoping to snag my 43^rd state highpoint. The climb is notorious among the highpointing community, as the standard route is a grueling 22-mile trek, spanning the windswept Froze-to-Death plateau, treacherous snow bridge crossings, and a Class 4 climb over exposed granite to the peak. Carrying an extra twenty pounds of climbing gear didn’t appeal to us, so we opted for the trail less traveled.

To the south, a lesser-known route twisted along alpine lakes to a rock scramble up Granite’s southwest face. AllTrails didn’t have a map for this, so we drew one, marking the waypoints—places to camp, notes on the technical sections, and lakes for drinking water. While the elevation, mileage, and positioning of the lakes were correct, our experience in the wilderness was considerably more complex.

All maps are wrong

Whenever we do something new—start a job, cook a recipe, or summit a mountain—we rely on proxies to augment our knowledge gaps. All proxies are incomplete representations of reality, as they are simplified illustrations of a complex system. A company’s org chart is a simple map of reporting structures. It doesn’t convey all the interconnections, power dynamics, and centers of gravity that shape the working environment. A system diagram doesn’t depict every dimension of architecture and is often outdated.

Earlier in my product career, I relied heavily on mental models to make decisions. The RICE prioritization framework gave me the training wheels to prioritize features for my backlog. Still, it was never a perfect solution as it couldn’t capture the nuances of the space.

But some are useful

Although proxies of reality are limited and “wrong” because they aren’t perfect representations, some are helpful. Using the RICE prioritization framework wasn’t perfect, but the template was much better than prioritizing features on “gut instinct.” Likewise, the first attempt at a dish might taste meh, but a recipe could teach us the basics and produce a better meal than PB&J.

Frameworks and mental models are training wheels that provide enough knowledge to get us rolling.

The map is not the territory

On Granite, the terrain was varied—snow, scree, and boulders covered the trail, raging creek crossings flooded our boots, mosquitoes feasted on open skin, and mountain goats stalked us for our salt. Cell service was nonexistent, afternoon thunderstorms rained golf-ball-sized hail, and the final summit push required Class 5 climbing over wet granite. By some miracle, I managed to reach the peak.

As any climber knows, the adventure is only half over at the summit. After this selfie, I downclimbed and hiked a dozen miles back over snow and rock to the car. And the car sat at the end of the gnarliest dirt road we’d ever driven—which, of course, had looked perfectly smooth on the map.

The value of a guide

After a few years in the corporate world, I brimmed with sophomoric confidence. I job-hopped, didn’t heed the generous wisdom of more tenured peers, and learned too many lessons the hard way. My eagerness led me to gobble up documents and develop an inflated sense of competence.

People with experience know the territory—where the skeletons are, the pitfalls, and how reality compares to concepts. They can get things done quickly and guide new folks who are smart enough to listen.

As I drove along the dark and twisty Beartooth Highway late that night, I reflected on the difficulty of Granite compared to other climbs. Denali is objectively leagues harder than any other, but highpointers often argue about the second hardest: Rainier, Gannett, or Granite. I haven’t climbed Gannett, but I have thoughts on the others.

Although Rainier had more vertical gain, higher elevation, and scaling glaciers was technically challenging; I followed a guide. The seasoned mountaineer had led expeditions to the tops of Aconcagua and Everest, and not once did I fear the knife-edge walks between crevasses under his instruction.

Granite was a less technical endeavor. It featured rugged terrain, a dangerous rock climb, and necessitated bear spray to ward off grizzlies, but the elevation was less intense. However, we were on our own—following the opaque traces of a back-country trail and solving problems as we discovered them. More thinking and bravery were required because we only knew the map, not the territory.

Updating the map

Last year, I published the Highpoint Guide to help fellow highpointers reach the summits. I thought I could write guides for highpoints I hadn’t visited because I could synthesize logs from other climbers. But, as I sorely learned on Granite, my maps were insufficient without experiencing the terrain. Updates are coming soon.

Never trust a climber with clean boots.

Junk Drawer

Try to accept these terrible terms and conditions.
Play an oddly addicting blob game.
Read about other mental models on my latest web project, Model Garden.