What is the core argument of "Attention is all you have"?

The author opens with a well-known psychological phenomenon — the "Tetris effect": after playing for a few weeks, everything looks like blocks — what you continuously attend to shapes your thinking. The problem is that attention is increasingly not your own to allocate: the original article gives four examples — video recommendations embedding anxiety-inducing content, playlists inserting AI-generated songs to save royalties, workplace platforms burying useful career information among strangers' posts, and community discussions being flooded by bots and trolls. The conclusion: letting others decide what appears on your screen is equivalent to handing over the keys to your brain.

How is this different from the usual "anti-addiction" advice?

According to a highly upvoted supplement in the discussion section, the problem runs deeper than "lack of self-control": mainstream books are also being rewritten for low attention spans, legitimate productivity apps are also competing for attention with notifications, and old portal homepages were also pushing content. So quitting a single app is merely a tactical victory. The guest's perspective is at the level of mechanisms: attention allocation has been taken over by product design; safeguarding sovereignty requires changing the entry point (RSS, subscriptions) and setting a direction (clarifying what you want), not relying on willpower alone.

What are the three things ordinary workers can do right now?

Guest's response: First, turn off non-essential push notifications, changing "being called to look" to "choosing to look." Second, keep at least one entry point that algorithms don't touch, like RSS or newsletters — look at the source before looking at summaries. Third, set a direction for yourself and filter information consumption around it — without a direction, tactical victories will be worn down by time. These are experiential insights, not statistical conclusions.

Attention Sovereignty: How Workers Can Keep the Initiative on 'What to Watch' in the Age of Algorithmic Feeding

Opening: Scrolling Through Your Phone Is More Exhausting Than a Full Day of Work — Have You Experienced This?

Let’s start with an experience almost everyone has had: you lie in bed at night, planning to look up something, and end up scrolling through short videos for over an hour. When you put your phone down, you feel your mind is swollen and your thoughts are floating — more exhausted than having been in meetings all day. And on top of that exhaustion, you can’t even remember what you just looked at.

On September 21, a long-form essay titled “Attention is all you have” laid this experience bare. The day it was published, it hit the hot list on the tech community’s longtime discussion forum — over six hundred upvotes and more than a hundred discussions. This number alone signals the magnitude of its resonance, not a conclusion, but it tells you one thing: this exhaustion is a universal feeling, not a personal lack of self-control.

Shiwen: Why did this episode start from this kind of “exhaustion”?

Yongliang: Because the sovereignty question of “what to look at” is something people rarely think of as a problem. You open an app, the homepage feeds you what it decides, and you look at it — it feels natural. But this article reminds us: what’s on your screen didn’t fall from the sky; someone decided it for you — and the motives of the person deciding it don’t necessarily align with your interests. This episode unpacks this: what exactly are algorithms doing to you, how others view this issue, and my own humble methods for safeguarding sovereignty over the years. Let me state the rules upfront: the examples in this article are those given in the original; my practices are my own experience; and wherever I’m estimating, I’ll make it clear that it’s my estimation, not a statistic. Attention is something you most fear having taken away in a fog.

Q1: What Does This Article Actually Say

Shiwen: Let’s start by laying out the core thesis of the original article clearly.

Yongliang: One-sentence version: what you continuously pay attention to shapes your thinking, and what you’re continuously paying attention to is increasingly not something you chose.

The article opens with a well-known psychological phenomenon called the Tetris effect: someone who plays Tetris every day for several weeks will start seeing clouds in the sky and the outlines of buildings on the street as falling blocks — without meaning to. The brain turns whatever you’re repeatedly exposed to into the way it sees the world. The author applies this to today: before, it was games that shaped you; now, it’s recommendation feeds — what you feed your brain each day determines what shape it becomes.

Her observations are very specific. The original article gives examples from four platforms. YouTube: you just want to watch videos, and the platform knows you love cooking and art livestreams, but it will casually throw in clips about stock bubbles, global warming, and the Iran war — because this kind of apocalyptic content keeps you watching longer and clicking more ads. Spotify: beyond algorithmic playlists, the article specifically mentions that they insert AI-generated junk songs between real ones, to pay less royalties to human artists. LinkedIn: you want to know what your colleagues are up to, but the bad news is that useful career information is buried among strangers’ opinions — and by pure coincidence, these strangers are all currently hyping things Microsoft is investing in. Reddit: you want to ask for a product review, but the accounts replying to you might now be a bunch of large models chatting about other things.

After the four examples, her conclusion comes in one sentence: “When you let someone else decide what appears on your screen, you are handing over the keys to your brain.” This is a direct translation of her original words, and it’s also the sentence I found most striking after reading it.

Shiwen: She also seems to miss the old internet?

Yongliang: Yes, this nostalgia isn’t pure sentimentality. She says that before recommendation algorithms, you had to decide for yourself what to do when you turned on your computer: dozens of bookmarked websites, each with a clear purpose — you go look up a recipe to look up a recipe, you go look at cats to look at cats. There were bad things too, of course, but “you had to walk there yourself” — no one would casually push extreme content after a muffin recipe or a cat video. She finally offered a way out: that “conscious internet” is still there, just buried beneath commercial internet. The key is to adapt to a slower internet — content is not infinite, and not every click brings an update. There are no shortcuts to building habits; the only thing to do is to stick with it.

Q2: Isn’t This Just the Same Old “Anti-Screen-Addiction” Talk? What’s New?

Shiwen: At this point, a reader might say: “Use your phone less, beware of algorithms” — that’s been said for ten years. What’s new here?

Yongliang: This challenge itself is the most valuable part of this episode — because the discussion section is full of exactly this kind of counterargument, and every single one is sharp.

The first counterargument: Are books safe? Someone in a book club in the comments noticed that more and more mainstream books are also being written for low attention spans — details repeated endlessly, plots drastically simplified. The form of books is being reverse-engineered by the attention economy; books are not automatically immune. I agree with this: the problem isn’t with a particular medium; it’s that the entire content supply side is racing toward “easier to swallow.”

The second counterargument: Even legitimate work tools compete for your attention. Another highly upvoted comment pointed out that it’s the same with apps used for work — notifications and tips flash before your eyes, pushing new features you don’t need right now, or warning you that everything is about to collapse. This punctures a placebo: the belief that “uninstalling entertainment apps and keeping only work tools” makes you safe. The hand reaching for your attention has already extended into productivity tools.

The third counterargument: Old portals pushed content too. Another comment offered a historical correction: the article made it sound like everything was about bookmarks, but Lycos, Yahoo, MSN, and other customizable homepages were very common back then, and browsers were pushing content too. This correction is right, but it actually makes the article’s main point clearer — the difference lies in what is being pushed and why: back then, they pushed what you subscribed to; now, they push whatever keeps you around the longest.

So my answer is: what’s fresh about this article isn’t shouting “use your phone less,” but redefining the problem from “insufficient self-control” to “sovereignty ceded away.” When you guard against screen addiction, you’re guarding against yourself; but the sovereignty question is guarding against the people who decide for you. These are two different things — the former makes you blame yourself; the latter makes you see clearly.

Q3: How Do You Yourself Guard It — How Is Your Information Diet Structured?

Shiwen: We’ve talked about the problem for a long time — let’s bring it back to you personally. How do you structure your information diet?

Yongliang: First, let me clarify my identity — this is my personal practice, habits formed over seventeen years of leading teams and doing technology management, not a statistic, and not necessarily suitable for everyone.

My principle is only one: make sure “I decide what to look at” happens before “I open a screen.” Specifically, it’s three layers. The first layer is feed reading: I changed my information entry point from recommendation feeds to sources — RSS subscriptions, newsletters, and a small number of frequently visited sites, each categorized, open whichever one I want. The essence of this step is turning “push” back into “pull.” The second layer is temporal separation: deep reading and phone scrolling happen on different devices, at different times — never mixed on the same screen — because I know I can’t win in an environment designed to capture me, so I simply don’t enter it. The third layer is the most foolish but most effective: after reading something, leave a first-hand record, and anything worth sharing must be traceable to its original source.

The third layer later became our team’s decision hygiene. I established a rule for my team: anyone who gives me a conclusion must attach a first-hand source link. Things heard secondhand, paraphrased, or screenshotted don’t count. The reason is simple: among the ways decisions get corrupted, the most concealed is when a second-hand conclusion dresses up in first-hand clothes and walks in. This rule has been in place for many years, and I can attest: meetings fueled by second-hand information have visibly decreased. This is experiential judgment, not statistics.

Shiwen: Can ordinary people learn this?

Yongliang: The most expensive thing among the three layers is perseverance, not money. RSS and newsletters are almost free to start; what’s truly hard is the first two weeks — without a recommendation feed feeding you, the screen will suddenly feel empty, and you’ll be annoyed. But that “annoyance” is exactly where you need to be: it means your attention is being pulled back to yourself. The original article says “the only way to build a habit is to keep at it,” and my understanding of this sentence is to endure through this period of annoyance.

Q4: You Yourself Produce Content — Isn’t That a Contradiction?

Shiwen: This has to be asked: you’re a self-media creator, and the content you produce is also competing for other people’s attention. Isn’t that a contradiction?

Yongliang: It is a contradiction, and I don’t want to gloss over it with a pretty self-defense.

My job is to get you to click on my article, read it, and come back next time — the essence of this is the same business as YouTube getting you to stay one more minute. The difference is only in what means I use: some means hold you down; other means make the thing good enough that you want to stay. Frankly, this is a matter of degree, not a matter of moral high ground. If one day my titles start lying for clicks, and my content starts padding for completion rates, then I’d be no different from what I criticize.

So I set rules for myself too — a sovereignty code for content producers: titles can speak in plain language, but not in lies; length can be short, but I won’t cut away necessary complexity just to make it “easier to swallow” — that criticism from the comments section about “books also getting shorter,” I use it as a production warning, not as reader complaining. There’s also a more practical constraint: I must be willing to put what I write into my own information diet myself. Only after this passes do I publish. Safeguarding other people’s attention sovereignty starts with admitting that your own hands aren’t clean either — I accept this.

Q5: Three Things Ordinary Workers Can Do Today

Shiwen: Finally, a take-away checklist. What can an ordinary worker do today — without changing technology or jobs?

Yongliang: Three things, ordered from lowest to highest cost, all of which can start today.

First, do it tonight: Turn off all non-human notifications. Not uninstalling apps — turning off push permissions. That comment in the discussion section about “even work apps are competing for you” reminds us that notifications are the vanguard of the grab — the vast majority of notifications aren’t worth the interruption they cause. Keep phone calls and text messages, turn off everything else. This step costs nothing, and the results are immediate.

Second, do it this week: Change one of your information entry points from “push” to “pull.” Pick the app you scroll through most obsessively, set a fixed time window for it, and tuck the rest of the entry deep into a folder. At the same time, pick three sources you truly trust — an industry site, a newsletter, an in-depth media outlet — and actively subscribe to them. This doesn’t solve the problem, but it takes back part of the “default option” from other people’s hands.

Third, do it long-term: Give your attention a “backbone.” One comment in the discussion section struck me as the deepest of all: unless you have a clear life goal and are truly invested in it, all improvements are at best tactical victories that will be worn down by time — and small victories actually make the end point more frustrating. Translate this: tools can’t safeguard sovereignty; only direction can. If you clearly know what you need to figure out this week and what you need to build this year, recommendation feeds will have a hard time pulling you away — because you’ll automatically find them annoying. The first two are tactics; this one is the principle. Tactics can be done today; the principle takes a lifetime to cultivate.

Closing

Shiwen: One sentence to sum up this episode?

Yongliang: The four words “attention sovereignty” sound big, but the landing point is small: when was the last time you actively decided what to look at? If you can’t remember, it’s time to take the keys back.

Shiwen: I’ll leave you with that sentence. See you next time.

[Deep Dive] Why “Quitting Your Phone” Always Fails — Anti-Addiction from a Sovereignty Perspective

Many readers have probably tried quitting their phone and failed. Let me explain why from a different angle. The logic of quitting your phone treats the enemy as “your own willpower” — so after you fail, you blame yourself. But from a sovereignty perspective, the enemy isn’t you at all; it’s the mechanism that decides screen content for you, and its design goal is to beat you. Fighting it with willpower is like using your amateur skills against someone else’s profession.

So the correct order is to seize power first, then talk about self-discipline: first take back the decision-making power of “what to look at” — turn off notifications, change entry points, set directions — and then ask yourself to persist. Reverse the order, and all your efforts will be wiped out by a late-night push notification. Let me hand you that sentence from the original one more time: “When you let someone else decide what appears on your screen, you are handing over the keys to your brain.” The act of taking the keys back is small, but it’s the prerequisite for everything.

Sources for This Episode

Alice GG’s long-form essay “Attention is all you have” (alicegg.tech, September 21, 2026; original blog post; the Tetris effect, the four platform examples of YouTube/Spotify/LinkedIn/Reddit, the “keys to the brain” quote, memories of the old internet, and the “conscious internet” way out are all paraphrases of the original’s viewpoints; platform examples are noted as “as given in the original”; the quote is a direct translation of the original)
Hacker News discussion on the same day (public forum reaction pool; six hundred upvotes and over a hundred discussions are in the magnitude-of-heat range; counterarguments such as “books also getting shorter,” “work apps also competing for attention,” “old portals pushed content too,” and “tactical victories” are all paraphrases from the comment section, no specific individuals named)
Guest’s personal experience (RSS/newsletter/feed reading information diet, team “conclusions must include first-hand source links” decision hygiene, self-media self-restraint rules — all first-person experiential statements, not statistics)