The Eroding Capacity for Focus

"I am feeling it right now."

To produce this very piece of writing, I had to implement a strict 15-minute timer on my machine and aggressively block every possible distraction. Without this forced discipline, my attention would have drifted within minutes.

The most frustrating part is that this happens even when I am engaging in activities I actually look forward to. I feel a constant, nagging impulse to switch tasks.

The Struggle for Deep Work

For several years, I have balanced a demanding trifecta: studying, professional work, and open-source contributions. Managing these simultaneously requires a high degree of signal-to-noise filtering.

To be clear, my objective isn't some grueling 90-hour work week. Rather, I want the time I dedicate to programming, writing, or learning to be singular. I want $\text{Focus} = 100\%$.

The Reality Gap

Instead of deep immersion, my current workflow looks like this:

• Plan to work for several hours uninterrupted.
• Work for 10 minutes.
• Get distracted.
• Attempt to refocus.
• Repeat.

The Anatomy of a Distraction

My brain seems to have developed a sophisticated mechanism for avoiding "painful" states—specifically boredom or cognitive strain.

Common scenarios include:

1. Searching for a specific post $\rightarrow$ clicking tangential links $\rightarrow$ reading unrelated content.
2. Waiting for a process to finish $\rightarrow$ browsing the web $\rightarrow$ losing the thread of the original task.
3. Facing a difficult problem $\rightarrow$ standing up for water $\rightarrow$ checking the phone $\rightarrow$ total distraction.

---

A History of Attention

Back in 2015, I noticed how a smartphone in bed could swallow hours of one's life. Even then, "attention-economy" apps were becoming potent. My solution was simple: never keep a charger near the bed.

At that time, my computer usage was primarily active. I was either coding, reading, or playing strategic games like chess. I avoided passive consumption.

Eventually, I discovered platforms like Medium and HackerNoon. While they initially inspired my programming journey, they are now often filled with high-quality insights monetized clickbait and "slop." To combat this, I began curating a personal reading list for my commutes and utilized $\text{\LaTeX}$ for my notes to ensure I was synthesizing information rather than just copying it.

---

The Professional Paradox

I first noticed a disturbing trend at work: people who provide only $\approx 10\%$ of their attention are often viewed as successful, provided they can talk the talk.

This differs from university. While students often cheat or copy homework to get grades, it is immediately obvious when you speak to them that they are clueless. In the corporate world, the "low-effort, high-visibility" strategy actually works.

I once tracked my time and discovered I spent 8 hours a week just chatting on Slack. This habit essentially trained my brain to embrace distraction while programming.

The LLM Effect: Efficiency vs. Overstimulation

The rise of Large Language Models (LLMs) has introduced a new layer of fragmentation.

• The Research Loop: I'll outsource a task to an LLM, then start something else. Or, I'll enter a "research mode" where I chat with the AI for hours instead of actually building.
• The Mental Overhead: Even with notifications off, I find myself preoccupied with the prompt I sent, unable to focus on the present task.
• The Latency Frustration: Interacting with an AI can feel slow; I'm stuck waiting for responses and correcting hallucinations.

The most effective use of AI is outsourcing tasks that can be completed end-to-end without error. However, when I manage too many AI-driven threads at once, I become overstimulated.

def focus_mode(task):
    if task.is_hard():
        # Avoid the urge to use LLM for everything
        return "Work through the problem manually"
    else:
        return "Outsource to AI and move on"

Finding the Way Back

I used to live-stream my work as a form of "forced accountability"—the camera prevented me from escaping into my phone. Unfortunately, due to the current instability of internet connections in Iran, this is no longer viable.

I've noticed a decline in my ability to commit to long-term projects. If a result takes days to manifest, my motivation wanes. Conversely, "vibe coding"—writing quick utility scripts—is addictive because the reward is instant. This reinforces a dangerous habit: waiting for an answer rather than working through a problem.

My current countermeasures:

• Analog Reading: When motivation fails, I pick up a physical book.
• Tactile Gardening: I maintain a balcony garden. Pruning plants and moving soil is a grounding experience that is far less addictive than a movie or a screen.

Writing this article took significantly more effort than usual, but that struggle provided the initial spark of motivation I needed.