Article 011: Social Media
URL: https://thelongrun.work/articles/011-working-from-a-phone
Medium
Tags (5 max)
- Technology
- Future Of Work
- Software Engineering
- Productivity
- Artificial Intelligence
Subtitle
“Mobile work reveals what must stay central. If it works on a phone, it’s probably supervisory.”
LinkedIn Posts
Post 1: Publish-day
There’s a useful diagnostic hidden in the question: could I do this from my phone?
If the answer is yes, the work is probably supervisory. You’re reviewing, approving, checking, deciding. The artifacts came from somewhere else. You’re the decision point, not the execution engine.
If the answer is no, that’s equally informative. Debugging, architecture, exploratory coding — these don’t work on a phone because they require an environment that expands to match the problem. The phone form factor isn’t just inconvenient; it can’t hold what the work requires.
As more software work becomes agent-assisted, the supervisory layer grows. More of what senior developers do becomes review and direction rather than direct implementation. That work travels well. It can be distributed across time and device.
But the execution capacity matters too. Not continuously, but enough. The judgment required to supervise well comes from having done the work. That reservoir doesn’t replenish itself.
Working from a phone is a clue, not a goal.
New article: https://thelongrun.work/articles/011-working-from-a-phone
Post 2: Insight (target: 2026-04-06)
There’s a useful warning hidden in the “work from anywhere” framing.
The supervisory work that travels well — the approvals, reviews, decisions — draws on something that doesn’t travel as easily: the accumulated experience of having done the execution work.
You can evaluate an architecture because you’ve designed architectures that failed in specific ways. You can supervise debugging because you’ve spent hours tracing wrong assumptions through tangled codebases. You can tell when a spec has a dangerous ambiguity because you’ve seen how those play out.
That’s the reservoir. And it drains if it’s not replenished.
As more software work becomes agent-assisted, the supervisory layer expands. That’s not bad. But it creates a slow risk: the ratio shifts, execution experience accumulates more slowly, and the quality of the supervision quietly degrades. Not immediately. Not visibly. Just gradually.
The phone test is a way of noticing where you sit in the ratio. Work that functions on a phone is probably supervisory. That’s useful to know. What it can’t tell you is whether the underlying reservoir is still full.
https://thelongrun.work/articles/011-working-from-a-phone
Post 3: Reflection (target: 2026-04-27)
The phone test has a second use that’s harder to sit with.
The first use is descriptive: could I do this from my phone? If yes, the work is probably supervisory.
The second use is diagnostic in a more uncomfortable direction: why am I trying to make this work on my phone?
Sometimes the answer is just convenience. But sometimes it reveals something else — a pull toward the work that travels lightly, the decisions that don’t require full context, the approvals that feel like progress without demanding the deep engagement. The phone becomes a way of staying in motion without going into the harder territory.
Debugging a concurrency issue doesn’t fit in a phone session. Neither does working through a difficult architectural decision. These things require the full environment not because of screen size, but because they require a kind of sustained presence that is easy to defer.
If you notice yourself reaching for the phone version of your work more often than before, that’s worth pausing on. Not as a judgement. Just as a signal. The same way the work that won’t move to mobile tells you something important, so does the direction of your own gravitational pull.