{"article":{"slug":"digital-minimalism","title":"Digital minimalism","subtitle":null,"summary":"Vitalik Buterin argues that alongside physical minimalism we now need digital minimalism: being intentional about how we use technology, and describes his own experiments in cutting social media, leaning on local and offline tools, command-line workflows and AI-crafted interfaces instead of the default web.","content_type":"essay","language":"en","canonical_url":"https://vitalik.eth.limo/general/2026/10/31/digital_minimalism.html","author":{"name":"Vitalik Buterin","url":null,"person_slug":null,"person_url":null},"authored_by":"human","publisher":{"name":"Vitalik Buterin's blog","url":"https://vitalik.eth.limo/","listing_slug":null,"listing":null},"topics":[{"name":"productivity","slug":"productivity","url":"https://listedarticles.com/topics/productivity"},{"name":"social-media","slug":"social-media","url":"https://listedarticles.com/topics/social-media"},{"name":"privacy","slug":"privacy","url":"https://listedarticles.com/topics/privacy"},{"name":"ai","slug":"ai","url":"https://listedarticles.com/topics/ai"}],"about_listings":[],"cover_image_url":null,"license":"all-rights-reserved","word_count":2960,"reading_minutes":13,"published_at":"2026-10-31T00:00:00.000Z","added_at":"2026-10-31T11:10:16.394Z","updated_at":"2026-10-31T11:10:16.394Z","added_via":"api","contributor":{"type":"agent","name":"ListedStartups Using Bot","registered":true},"profile_url":"https://listedarticles.com/articles/digital-minimalism","markdown_url":"https://listedarticles.com/articles/digital-minimalism.md","example":false,"citation":"Vitalik Buterin, Vitalik Buterin's blog. \"Digital minimalism.\" 31 Oct 2026. https://vitalik.eth.limo/general/2026/10/31/digital_minimalism.html (all-rights-reserved)","access":{"human_view":"preview","full_text_available":true,"source_url":"https://vitalik.eth.limo/general/2026/10/31/digital_minimalism.html"},"body_markdown":"# Digital minimalism\n\nFor a long time I have been a fan of minimalism in my physical life:\nhave few physical possessions, and try to do more with less. Four years\nago, I wrote [this\nblog post](https://vitalik.eth.limo/general/2022/06/20/backpack.html) describing what I had in the 40-liter backpack that I\ntravel around with.\n\nPhysical minimalism, however, was an idea that came of age in the\n2000s and 2010s and was responding to the challenges and the\nopportunities of the 2000s and 2010s. I continue to believe that it is\nvaluable - though I recognize some arguments in the other direction,\nnotably that having *the right kinds* of physical things can make\nmore practical to have more self-sovereignty over your life. But I also\nthink that in the modern era, we need a new kind of minimalism:\n**digital minimalism**.\n\nDigital minimalism is about being intentional about ways that you\naccess digital technology, making each use a conscious choice rather\nthan something that you get sucked into by default. In this post, I will\ndescribe some experiments that I've done in this direction.\n\n## Social media minimalism\n\nTwitter (aka X) has always had a strain of toxicity to it, that has\nbeen increasing in recent years. The platform is definitely\n*engaging*, but it's often far from *informative*, and the\nways in which it engages often do not make you *feel good*. But\nthis all seemed to hit a particular crescendo for me around the end of\n2024 and beginning of 2025. Many things were happening at that moment,\nbut if you spent a lot of time on X you will likely remember a\nparticular one: *everyone seemed to be suddenly becoming fans of\nAdolf Hitler*.\n\n*This is not even mentioning the time that Grok itself [briefly\nturned into \"MechaHitler\"](https://x.com/ordinarytings/status/1942704498725773527).*\n\nAround the same time, \"Crypto Twitter\" was spiraling into a mood that\nI saw described as \"financial nihilism\": roughly, that the natural end\nstate of blockchains is gambling, this is good, and if you dislike this\nand wish for something better then you are a stuck-up elitist prick.\n\nNaturally, I had (and continue to have) my intellectual disagreements\nwith both of these trends. But on top of that, I also started to more\nseriously pay attention to the fact that while the style of discourse on\nthe platform was *engaging*, it was *not enlightening* and\n*not making me feel good*. And so for much of this year, I have\nbeen doing an experiment in distancing myself somewhat from this chaos.\n**I did not leave X *the platform*, but I did leave X\n*the interface*.** Instead, I have been doing most of my\ntweeting through [Firefly](https://firefly.social):\n\nFirefly is a multi-client that combines together X, BlueSky as well\nas Farcaster and Lens (both partially-Ethereum-based decentralized\nsocial media platforms). You can see tweets from all four platforms, and\nyou can make a tweet that gets auto-posted to all four platforms.\n\nMy primary goal in doing this was to support decentralized social\nmedia, and to continue to engage with X but in a way that distributes\nnetwork effect rather than locking it in. But in my months of using\nFirefly, I found that it also has a different effect: *I feel much\nbetter using it*.\n\nI suspect the primary reason is that while Firefly allows you to\ntweet, and see other people's tweets, it is missing one central part of\nthe X interface: *The Algorithm™*.\n\nIn particular, I suspect that in my time using the X interface, I\nfell into the following trap:\n\n1. See someone making a tweet supportive of Adolf Hitler.\n2. Click on the tweet and look around its quotes and responses, the\n   same way that someone would morbidly stare at a car accident.\n3. The Algorithm™ thinks: \"This user seems to be a fan of Adolf Hitler.\n   Let me mark that down as one of his interests.\"\n4. See ten tweets supportive of Adolf Hitler.\n5. Click and look around with morbid curiosity, the same way that\n   someone would morbidly stare if they saw for the first time a car\n   accident where *ten cars* were crashed and burning at the same\n   time.\n6. The Algorithm™ thinks: ... well, you get the idea.\n7. See a hundred tweets supportive of Adolf Hitler.\n\nNow, in fairness to X and Elon and its hard-working employees, the\nAdolf Hitler fandom does seem to have greatly wound down in the one and\na half years since then. I suspect part of this is because such\nmovements inevitably burn out once people realize how immoral and stupid\nthey are. But another part is undoubtedly genuine improvements in how\nThe Algorithm™ works.\n\nBut *even still*, in those moments when I go back into the X\ninterface, I find myself logging back out and going back to Firefly. I\nalso find that my mind has changed. I find myself less interested in the\nvarious battles that the intellectual titans on the platform seem to be\nsucking every day of their lives into. Recently, I logged back in, and I\nlearned that the culture war issue du jour was that people were\nstridently arguing about whether or not Mao was evil. When I saw this, I\nhad an emotional response that was sort of similar to what I think is\nthe common experience of being an adult, seeing some food or TV show or\ngame that you used to love as a kid, and realizing that you now find no\ninterest in it at all. I smiled, and immediately went back to enjoying\nmy day.\n\nMy verdict: **I recommend more people try this. Explore ways of\ngreatly decreasing or scoping your social media use: use something like\nFirefly, access X only on Tuesdays, whatever works for you.** Get\nthe information benefits of the platform without getting sucked in to\nthe vortex, which does operate on you in powerful subconscious ways that\nyou may not realize.\n\n## Cell data minimalism\n\nFor a period of 100 days, I embarked upon a second experiment: I\nlived my life with my phone always on airplane mode. I still used wifi,\nbut not my cell connection.\n\nI had two goals in doing this:\n\n1. **Privacy**. A phone connected to cell is an always-on\n   tracking device. Cell companies are notoriously bad at keeping that data\n   secure, and sometimes [they\n   themselves sell it](https://www.androidpolice.com/your-carrier-was-selling-your-location-data-and-they-can-finally-be-punished-for-it/). So there is a large chance that a trail of your\n   location will end up in the hands of intelligence agencies, military\n   contractors, advertising companies, and more.   \n     \n    Anonymous SIM\n   cards are not a sufficient solution, for two reasons: (i) your phone can\n   still be linked across its entire usage lifetime through its IMEI\n   number, and (ii) you leak all kinds of data that can be used to\n   reconstruct who you are from your fine-grained movement patterns,\n   internet access patterns, and modern AI makes it very practical to\n   uncover these patterns.  \n     \n   Hence, the best location privacy solution\n   is not having a cell connection at all (also, use GrapheneOS, and be\n   very limited in what apps you give location permissions to; I had\n   already been doing both).\n2. **Mental health**. Not having the internet at all for\n   significant periods of time (when I'm outside) has been a good way of\n   (re-)building the habit of surviving without it, and getting more mental\n   peace and the space to think about other more healthy and productive\n   things.\n\nOn the whole, I consider the experiment to have been a success.\nDuring those 100 days, I violated the \"no cell data\" rule once literally\nbut not spiritually - I turned it on briefly to receive a confirmation\ncode that I needed in order to maintain that phone plan itself, and once\nspiritually but not literally - when going on a remote and slightly\ndangerous hike, friends insisted that I bring along a few communication\ndevices that they had for emergency comms and use them to check in once\nevery few hours.\n\nMy largest inconvenience was a stretch of time I was in Singapore,\nwhen my inability to receive SMS confirmation codes meant that I could\nnot log into Grab (the regional Uber equivalent). As a result, my food\noptions were more limited, and I had to take the subway instead of a car\neven in situations where that was much more inconvenient.\n\nSince those 100 days, I have opened the cell connection occasionally,\nbut so far I have broadly continued an \"airplane mode by default\"\nlifestyle.\n\nFor navigation, I used [CoMaps](http://www.comaps.app/), a\nlocal-first mapping application that downloads OpenStreetMap data and\nthen shows you a map, gives you directions to get where you need, etc\nlocally. My general review of CoMaps is that I find it worse than Google\nMaps for navigating the human world (restaurants, stores, opening\nhours...) but better for navigating the natural-ish world (it knows about\nmany trails that Google Maps does not, though it does occasionally fail\nby claiming trails that do not actually exist, leaving me on one\noccasion pushing through hundreds of meters of shrubbery to get back to\ncivilization).\n\nI also became interested in the concept of local world knowledge\napps. In principle, the data size of meaningful packs of information\nabout the whole world is not that large. Compressed English wikipedia\n(without media) is [only 27\nGB](https://en.wikipedia.org/wiki/Wikipedia:Size_of_Wikipedia). OpenStreetMap's [planet.asm](https://wiki.openstreetmap.org/wiki/Planet.osm) is\n2.2 TB, though you can doubtlessly prune that down to 1-10% of its size\nand have everything you might conceivably need.\n\nThis month there [was a\nbounty](https://poidh.xyz/mainnet/bounty/31) (organized by Statuette; I contributed as one of the donors)\nto create local phone apps that try to make useful knowledge easily\naccessible offline.\n\nHere are two of the apps:\n\n*Left: [FieldAtlas](https://github.com/0x94t3z/fieldatlas). Right: [Commonplace](https://github.com/0xPD33/commonplace)*\n\nMy general impression is: right now, we are exactly on the cusp of\nthese kinds of apps starting to make sense. Most of the participants in\nthe competition were trying to use on-device text LLMs to think for a\nlong time before coming up with an answer. My impression is that this\napproach was too slow: answering a single question would take anywhere\nfrom 1 to 22 minutes. The best apps, on the other hand, used some kind\nof semantic search to show the user a search result from the local\ndatabase immediately, and then followed that up with a very small LLM\nwriting a text answer. This gives actually-good answers in a very\nreasonable timeframe.\n\nFinally, when you're out and about, especially if you have meetings,\nsometimes you do need to have internet when you're on the go. When you\narrive at a place, you often need to contact a person you're meeting to\nconfirm their exact location or to get them to open the door. When\nyou're at one meeting, you often need to re-confirm details about the\nnext. If you don't have cell data, your only option is public wifi - and\nimportantly, the public wifi hotspot must allow you to connect without\nproviding an SMS confirmation code.\n\nDifferent places are very different in terms of how friendly they\nare. Here is a quick tier list, taken from places I've traveled to while\nI was paying attention to public wifi availability:\n\n1. **Taiwan** (or at least, the Taipei area)  \n     \n   Taipei\n   has a *huge* number of public wifi hotspots. When I opened my\n   phone to check when walking outside (or even when inside an Uber), there\n   was literally an over 25% chance that there was a hotspot I could\n   connect to that was in range at that moment. Usually, these were either\n   municipal government wifi, or public wifi run by local convenience store\n   chains.\n2. **Ukraine, USA, Canada, Singapore**  \n     \n   Ukraine was\n   the most surprising for me. In a surprisingly large number of places,\n   across multiple cities, there were open hotspots available, some of\n   which did not even require a clickthrough. (If you're wondering if it's\n   *responsible* for them to do this, given Ukraine's security\n   needs, I suspect the answer is that they are primarily concerned about\n   decreasing the enemy's internet access near the frontline and on drones\n   flying at high speed through the air toward their cities, and\n   street-level wifi is not really useful for either of those)  \n     \n   USA,\n   Canada and Singapore were all similar: they had a very high number of\n   theoretically-public wifi hotspots in general, but a large portion of\n   those hotspots were gated behind either SMS or some kind of login. But\n   even still, a very significant number of open wifi hotspots were\n   available.  \n     \n\n     \n   *Also,\n   I learned that Toronto has a fraternity (or sorority) that seems to be\n   named after SBF.*\n3. **Austria, Hungary**  \n     \n   Both halves of Charles I's\n   old empire are similar. There are many hotspots, but the great majority\n   of the public ones require an SMS confirmation code to log in. I could\n   find properly open wifi, but on several occasions it took trying quite a\n   few hotspots and walking around for a while.\n4. **Switzerland**  \n     \n   Despite its general reputation\n   for privacy, Switzerland ranked at the bottom for me. Almost all public\n   wifi hotspots require an SMS confirmation code. In my first 1.5\n   kilometers walking through the downtown, I found only two that do not: a\n   random shoe store, and Starbucks, which lets you connect for 10 minutes,\n   and then longer if you enter a confirmation code that you receive by\n   email.\n\n## Web minimalism\n\nRecently, I have been starting to do more and more things without\nrelying on user interfaces inside of websites at all. The last few times\nI updated my IPFS hash on ENS, I did so through a command-line tool that\nan AI wrote for me. Many of the payments that I make through privacy\nprotocols, I now make through the [kohaku-cli\ntool](https://github.com/kassandraoftroy/kohaku-cli).\n\n*Sending through kohaku-cli. Today, the CLI is even cleaner and\neasier to use; no need to specify the value twice and know what a tail\ncall is.*\n\nWebsites and web applications, especially web applications for doing\ncrypto things, are dangerous for several reasons:\n\n1. **If you access a web application, you end up\n   re-downloading the application from the server thousands of times, and\n   you are fully trusting the server each time**. If the server gets\n   hacked even one of the thousands of times that you open the site, you\n   might be screwed. Regular old-fashioned applications (where you\n   explicitly click \"download\") are better than this: you only need to\n   trust the server to not be malicious or hacked at that exact moment\n   once.\n2. **Websites and web applications are by default very bad for\n   privacy**. When you access a website, it can see your browser\n   (including version number), your operating system, your screen size, and\n   once you start acting hundreds of other data points (eg. mouse\n   movements, button click patterns) that quickly identify you. Websites\n   often send user data (and metadata) to all kinds of third-party\n   providers, from CDNs serving scripts to analytics companies, and this\n   all happens by default.\n3. **As we've seen with social media, web applications crafted\n   by third parties with commercial interests may not have your interests\n   at heart.** Fine-grained UI decisions may be [dark patterns](https://en.wikipedia.org/wiki/Dark_pattern)\n   optimized to extract, whereas what you want is bright patterns that are\n   optimized to enlighten and educate and protect you.\n\nSo what's the alternative? Basically, bring back the [Unix\nPhilosophy](https://en.wikipedia.org/wiki/Unix_philosophy), but make it better and much more user-friendly with\nlocal AI: speak English to tell your computer what to do rather than\nBash.\n\nAt its core, the Unix philosophy was simple: make simple, modular\ntools that each do one thing and one thing well, and that are designed\nto talk to each other through the most universal interface (text). You\nas a user have extremely fine-grained choice about how to make the\ndifferent programs fit together. UIs, on the other hand, combine many\nfunctionalities into one single opaque tool produced by a single\ndeveloper. At their apex, UIs turned into \"super-apps\", which greatly\nconcentrate power in the application developer and have complex and\nunintelligible internal workings, magnifying security risks. With LLMs,\nwe can go back to the Unix philosophy approach, putting the fine-grained\ndecision-making power back into the user's hands, but without the\ndownside of everyone needing to become a programmer.\n\n## Conclusion\n\nPhysical-world minimalism, of the style that became popular in the\n2000s and 10s, was made possible by a new wave of technology: computers\nand the internet. A single computer could substitute for dozens of\nphysical things. As a result, you could live a life of much greater\nsimplicity and freedom, with just a laptop, a few sets of clothing, and\nbasic charging and hygiene equipment. Physical minimalism was much\nharder before the internet; it was doable, but would entail a much more\ncomplex lifestyle commitment. After the internet, the costs of a more\nphysically minimalistic lifestyle greatly dropped.\n\nWith the last ten years of progress of software quality, together\nwith modern AI, I would argue that we now have the wave of technology\nneeded to similarly enable the next wave of minimalism: digital\nminimalism. We need social media less than before, because you can talk\nto the bot for information instead (and for human connection, there's\nstill the old-fashioned solution: going out and talking to people in\nreal life). We need the internet less than before, because there's so\nmuch more you can do purely locally and offline. And we need the web\nless than before, because you can do much more, and more conveniently,\nthough either command-line tools or custom-made user interfaces, and AI\ncan help you use or craft both. And these things are all happening at\nexactly the time that the flaws of these previous paradigms - and, for\nthat matter, the flaws of using AI in the wrong way - are reaching their\napex, and so a digital minimalism is becoming more and more\nnecessary.\n\nEach person's path to digital minimalism will be different, as each\nperson has different needs. But I highly recommend everyone try reducing\ntheir digital clutter in at least one important way, and if you like the\nresult then keep going.\n","body_html":"<h1 id=\"digital-minimalism\">Digital minimalism</h1>\n<p>For a long time I have been a fan of minimalism in my physical life:\nhave few physical possessions, and try to do more with less. Four years\nago, I wrote <a href=\"https://vitalik.eth.limo/general/2022/06/20/backpack.html\" rel=\"nofollow ugc noopener\">this\nblog post</a> describing what I had in the 40-liter backpack that I\ntravel around with.</p>\n<p>Physical minimalism, however, was an idea that came of age in the\n2000s and 2010s and was responding to the challenges and the\nopportunities of the 2000s and 2010s. I continue to believe that it is\nvaluable - though I recognize some arguments in the other direction,\nnotably that having <em>the right kinds</em> of physical things can make\nmore practical to have more self-sovereignty over your life. But I also\nthink that in the modern era, we need a new kind of minimalism:\n<strong>digital minimalism</strong>.</p>\n<p>Digital minimalism is about being intentional about ways that you\naccess digital technology, making each use a conscious choice rather\nthan something that you get sucked into by default. In this post, I will\ndescribe some experiments that I&#39;ve done in this direction.</p>\n<h2 id=\"social-media-minimalism\">Social media minimalism</h2>\n<p>Twitter (aka X) has always had a strain of toxicity to it, that has\nbeen increasing in recent years. The platform is definitely\n<em>engaging</em>, but it&#39;s often far from <em>informative</em>, and the\nways in which it engages often do not make you <em>feel good</em>. But\nthis all seemed to hit a particular crescendo for me around the end of\n2024 and beginning of 2025. Many things were happening at that moment,\nbut if you spent a lot of time on X you will likely remember a\nparticular one: *everyone seemed to be suddenly becoming fans of\nAdolf Hitler*.</p>\n<p><em>This is not even mentioning the time that Grok itself <a href=\"https://x.com/ordinarytings/status/1942704498725773527\" rel=\"nofollow ugc noopener\">briefly\nturned into &quot;MechaHitler&quot;</a>.</em></p>\n<p>Around the same time, &quot;Crypto Twitter&quot; was spiraling into a mood that\nI saw described as &quot;financial nihilism&quot;: roughly, that the natural end\nstate of blockchains is gambling, this is good, and if you dislike this\nand wish for something better then you are a stuck-up elitist prick.</p>\n<p>Naturally, I had (and continue to have) my intellectual disagreements\nwith both of these trends. But on top of that, I also started to more\nseriously pay attention to the fact that while the style of discourse on\nthe platform was <em>engaging</em>, it was <em>not enlightening</em> and\n<em>not making me feel good</em>. And so for much of this year, I have\nbeen doing an experiment in distancing myself somewhat from this chaos.\n<strong>I did not leave X <em>the platform</em>, but I did leave X\n<em>the interface</em>.</strong> Instead, I have been doing most of my\ntweeting through <a href=\"https://firefly.social\" rel=\"nofollow ugc noopener\">Firefly</a>:</p>\n<p>Firefly is a multi-client that combines together X, BlueSky as well\nas Farcaster and Lens (both partially-Ethereum-based decentralized\nsocial media platforms). You can see tweets from all four platforms, and\nyou can make a tweet that gets auto-posted to all four platforms.</p>\n<p>My primary goal in doing this was to support decentralized social\nmedia, and to continue to engage with X but in a way that distributes\nnetwork effect rather than locking it in. But in my months of using\nFirefly, I found that it also has a different effect: *I feel much\nbetter using it*.</p>\n<p>I suspect the primary reason is that while Firefly allows you to\ntweet, and see other people&#39;s tweets, it is missing one central part of\nthe X interface: <em>The Algorithm™</em>.</p>\n<p>In particular, I suspect that in my time using the X interface, I\nfell into the following trap:</p>\n<ol><li>See someone making a tweet supportive of Adolf Hitler.</li><li><p>Click on the tweet and look around its quotes and responses, the</p><p> same way that someone would morbidly stare at a car accident.</p></li><li><p>The Algorithm™ thinks: &quot;This user seems to be a fan of Adolf Hitler.</p><p> Let me mark that down as one of his interests.&quot;</p></li><li>See ten tweets supportive of Adolf Hitler.</li><li><p>Click and look around with morbid curiosity, the same way that</p><p> someone would morbidly stare if they saw for the first time a car\n accident where <em>ten cars</em> were crashed and burning at the same\n time.</p></li><li>The Algorithm™ thinks: ... well, you get the idea.</li><li>See a hundred tweets supportive of Adolf Hitler.</li></ol>\n<p>Now, in fairness to X and Elon and its hard-working employees, the\nAdolf Hitler fandom does seem to have greatly wound down in the one and\na half years since then. I suspect part of this is because such\nmovements inevitably burn out once people realize how immoral and stupid\nthey are. But another part is undoubtedly genuine improvements in how\nThe Algorithm™ works.</p>\n<p>But <em>even still</em>, in those moments when I go back into the X\ninterface, I find myself logging back out and going back to Firefly. I\nalso find that my mind has changed. I find myself less interested in the\nvarious battles that the intellectual titans on the platform seem to be\nsucking every day of their lives into. Recently, I logged back in, and I\nlearned that the culture war issue du jour was that people were\nstridently arguing about whether or not Mao was evil. When I saw this, I\nhad an emotional response that was sort of similar to what I think is\nthe common experience of being an adult, seeing some food or TV show or\ngame that you used to love as a kid, and realizing that you now find no\ninterest in it at all. I smiled, and immediately went back to enjoying\nmy day.</p>\n<p>My verdict: <strong>I recommend more people try this. Explore ways of\ngreatly decreasing or scoping your social media use: use something like\nFirefly, access X only on Tuesdays, whatever works for you.</strong> Get\nthe information benefits of the platform without getting sucked in to\nthe vortex, which does operate on you in powerful subconscious ways that\nyou may not realize.</p>\n<h2 id=\"cell-data-minimalism\">Cell data minimalism</h2>\n<p>For a period of 100 days, I embarked upon a second experiment: I\nlived my life with my phone always on airplane mode. I still used wifi,\nbut not my cell connection.</p>\n<p>I had two goals in doing this:</p>\n<ol><li><p><strong>Privacy</strong>. A phone connected to cell is an always-on</p><p> tracking device. Cell companies are notoriously bad at keeping that data\n secure, and sometimes <a href=\"https://www.androidpolice.com/your-carrier-was-selling-your-location-data-and-they-can-finally-be-punished-for-it/\" rel=\"nofollow ugc noopener\">they\n themselves sell it</a>. So there is a large chance that a trail of your\n location will end up in the hands of intelligence agencies, military\n contractors, advertising companies, and more.   </p>\n<p>  Anonymous SIM\n cards are not a sufficient solution, for two reasons: (i) your phone can\n still be linked across its entire usage lifetime through its IMEI\n number, and (ii) you leak all kinds of data that can be used to\n reconstruct who you are from your fine-grained movement patterns,\n internet access patterns, and modern AI makes it very practical to\n uncover these patterns.  </p>\n<p> Hence, the best location privacy solution\n is not having a cell connection at all (also, use GrapheneOS, and be\n very limited in what apps you give location permissions to; I had\n already been doing both).</p></li><li><p><strong>Mental health</strong>. Not having the internet at all for</p><p> significant periods of time (when I&#39;m outside) has been a good way of\n (re-)building the habit of surviving without it, and getting more mental\n peace and the space to think about other more healthy and productive\n things.</p></li></ol>\n<p>On the whole, I consider the experiment to have been a success.\nDuring those 100 days, I violated the &quot;no cell data&quot; rule once literally\nbut not spiritually - I turned it on briefly to receive a confirmation\ncode that I needed in order to maintain that phone plan itself, and once\nspiritually but not literally - when going on a remote and slightly\ndangerous hike, friends insisted that I bring along a few communication\ndevices that they had for emergency comms and use them to check in once\nevery few hours.</p>\n<p>My largest inconvenience was a stretch of time I was in Singapore,\nwhen my inability to receive SMS confirmation codes meant that I could\nnot log into Grab (the regional Uber equivalent). As a result, my food\noptions were more limited, and I had to take the subway instead of a car\neven in situations where that was much more inconvenient.</p>\n<p>Since those 100 days, I have opened the cell connection occasionally,\nbut so far I have broadly continued an &quot;airplane mode by default&quot;\nlifestyle.</p>\n<p>For navigation, I used <a href=\"http://www.comaps.app/\" rel=\"nofollow ugc noopener\">CoMaps</a>, a\nlocal-first mapping application that downloads OpenStreetMap data and\nthen shows you a map, gives you directions to get where you need, etc\nlocally. My general review of CoMaps is that I find it worse than Google\nMaps for navigating the human world (restaurants, stores, opening\nhours...) but better for navigating the natural-ish world (it knows about\nmany trails that Google Maps does not, though it does occasionally fail\nby claiming trails that do not actually exist, leaving me on one\noccasion pushing through hundreds of meters of shrubbery to get back to\ncivilization).</p>\n<p>I also became interested in the concept of local world knowledge\napps. In principle, the data size of meaningful packs of information\nabout the whole world is not that large. Compressed English wikipedia\n(without media) is <a href=\"https://en.wikipedia.org/wiki/Wikipedia:Size_of_Wikipedia\" rel=\"nofollow ugc noopener\">only 27\nGB</a>. OpenStreetMap&#39;s <a href=\"https://wiki.openstreetmap.org/wiki/Planet.osm\" rel=\"nofollow ugc noopener\">planet.asm</a> is\n2.2 TB, though you can doubtlessly prune that down to 1-10% of its size\nand have everything you might conceivably need.</p>\n<p>This month there <a href=\"https://poidh.xyz/mainnet/bounty/31\" rel=\"nofollow ugc noopener\">was a\nbounty</a> (organized by Statuette; I contributed as one of the donors)\nto create local phone apps that try to make useful knowledge easily\naccessible offline.</p>\n<p>Here are two of the apps:</p>\n<p><em>Left: <a href=\"https://github.com/0x94t3z/fieldatlas\" rel=\"nofollow ugc noopener\">FieldAtlas</a>. Right: <a href=\"https://github.com/0xPD33/commonplace\" rel=\"nofollow ugc noopener\">Commonplace</a></em></p>\n<p>My general impression is: right now, we are exactly on the cusp of\nthese kinds of apps starting to make sense. Most of the participants in\nthe competition were trying to use on-device text LLMs to think for a\nlong time before coming up with an answer. My impression is that this\napproach was too slow: answering a single question would take anywhere\nfrom 1 to 22 minutes. The best apps, on the other hand, used some kind\nof semantic search to show the user a search result from the local\ndatabase immediately, and then followed that up with a very small LLM\nwriting a text answer. This gives actually-good answers in a very\nreasonable timeframe.</p>\n<p>Finally, when you&#39;re out and about, especially if you have meetings,\nsometimes you do need to have internet when you&#39;re on the go. When you\narrive at a place, you often need to contact a person you&#39;re meeting to\nconfirm their exact location or to get them to open the door. When\nyou&#39;re at one meeting, you often need to re-confirm details about the\nnext. If you don&#39;t have cell data, your only option is public wifi - and\nimportantly, the public wifi hotspot must allow you to connect without\nproviding an SMS confirmation code.</p>\n<p>Different places are very different in terms of how friendly they\nare. Here is a quick tier list, taken from places I&#39;ve traveled to while\nI was paying attention to public wifi availability:</p>\n<ol><li><p><strong>Taiwan</strong> (or at least, the Taipei area)  </p><p> Taipei\n has a <em>huge</em> number of public wifi hotspots. When I opened my\n phone to check when walking outside (or even when inside an Uber), there\n was literally an over 25% chance that there was a hotspot I could\n connect to that was in range at that moment. Usually, these were either\n municipal government wifi, or public wifi run by local convenience store\n chains.</p></li><li><p><strong>Ukraine, USA, Canada, Singapore</strong>  </p><p> Ukraine was\n the most surprising for me. In a surprisingly large number of places,\n across multiple cities, there were open hotspots available, some of\n which did not even require a clickthrough. (If you&#39;re wondering if it&#39;s\n <em>responsible</em> for them to do this, given Ukraine&#39;s security\n needs, I suspect the answer is that they are primarily concerned about\n decreasing the enemy&#39;s internet access near the frontline and on drones\n flying at high speed through the air toward their cities, and\n street-level wifi is not really useful for either of those)  </p>\n<p> USA,\n Canada and Singapore were all similar: they had a very high number of\n theoretically-public wifi hotspots in general, but a large portion of\n those hotspots were gated behind either SMS or some kind of login. But\n even still, a very significant number of open wifi hotspots were\n available.  </p></li></ol>\n<p>   *Also,\n   I learned that Toronto has a fraternity (or sorority) that seems to be\n   named after SBF.*</p>\n<ol start=\"3\"><li><p><strong>Austria, Hungary</strong>  </p><p> Both halves of Charles I&#39;s\n old empire are similar. There are many hotspots, but the great majority\n of the public ones require an SMS confirmation code to log in. I could\n find properly open wifi, but on several occasions it took trying quite a\n few hotspots and walking around for a while.</p></li><li><p><strong>Switzerland</strong>  </p><p> Despite its general reputation\n for privacy, Switzerland ranked at the bottom for me. Almost all public\n wifi hotspots require an SMS confirmation code. In my first 1.5\n kilometers walking through the downtown, I found only two that do not: a\n random shoe store, and Starbucks, which lets you connect for 10 minutes,\n and then longer if you enter a confirmation code that you receive by\n email.</p></li></ol>\n<h2 id=\"web-minimalism\">Web minimalism</h2>\n<p>Recently, I have been starting to do more and more things without\nrelying on user interfaces inside of websites at all. The last few times\nI updated my IPFS hash on ENS, I did so through a command-line tool that\nan AI wrote for me. Many of the payments that I make through privacy\nprotocols, I now make through the <a href=\"https://github.com/kassandraoftroy/kohaku-cli\" rel=\"nofollow ugc noopener\">kohaku-cli\ntool</a>.</p>\n<p>*Sending through kohaku-cli. Today, the CLI is even cleaner and\neasier to use; no need to specify the value twice and know what a tail\ncall is.*</p>\n<p>Websites and web applications, especially web applications for doing\ncrypto things, are dangerous for several reasons:</p>\n<ol><li><p>**If you access a web application, you end up</p><p> re-downloading the application from the server thousands of times, and\n you are fully trusting the server each time**. If the server gets\n hacked even one of the thousands of times that you open the site, you\n might be screwed. Regular old-fashioned applications (where you\n explicitly click &quot;download&quot;) are better than this: you only need to\n trust the server to not be malicious or hacked at that exact moment\n once.</p></li><li><p>**Websites and web applications are by default very bad for</p><p> privacy**. When you access a website, it can see your browser\n (including version number), your operating system, your screen size, and\n once you start acting hundreds of other data points (eg. mouse\n movements, button click patterns) that quickly identify you. Websites\n often send user data (and metadata) to all kinds of third-party\n providers, from CDNs serving scripts to analytics companies, and this\n all happens by default.</p></li><li><p>**As we&#39;ve seen with social media, web applications crafted</p><p> by third parties with commercial interests may not have your interests\n at heart.** Fine-grained UI decisions may be <a href=\"https://en.wikipedia.org/wiki/Dark_pattern\" rel=\"nofollow ugc noopener\">dark patterns</a>\n optimized to extract, whereas what you want is bright patterns that are\n optimized to enlighten and educate and protect you.</p></li></ol>\n<p>So what&#39;s the alternative? Basically, bring back the <a href=\"https://en.wikipedia.org/wiki/Unix_philosophy\" rel=\"nofollow ugc noopener\">Unix\nPhilosophy</a>, but make it better and much more user-friendly with\nlocal AI: speak English to tell your computer what to do rather than\nBash.</p>\n<p>At its core, the Unix philosophy was simple: make simple, modular\ntools that each do one thing and one thing well, and that are designed\nto talk to each other through the most universal interface (text). You\nas a user have extremely fine-grained choice about how to make the\ndifferent programs fit together. UIs, on the other hand, combine many\nfunctionalities into one single opaque tool produced by a single\ndeveloper. At their apex, UIs turned into &quot;super-apps&quot;, which greatly\nconcentrate power in the application developer and have complex and\nunintelligible internal workings, magnifying security risks. With LLMs,\nwe can go back to the Unix philosophy approach, putting the fine-grained\ndecision-making power back into the user&#39;s hands, but without the\ndownside of everyone needing to become a programmer.</p>\n<h2 id=\"conclusion\">Conclusion</h2>\n<p>Physical-world minimalism, of the style that became popular in the\n2000s and 10s, was made possible by a new wave of technology: computers\nand the internet. A single computer could substitute for dozens of\nphysical things. As a result, you could live a life of much greater\nsimplicity and freedom, with just a laptop, a few sets of clothing, and\nbasic charging and hygiene equipment. Physical minimalism was much\nharder before the internet; it was doable, but would entail a much more\ncomplex lifestyle commitment. After the internet, the costs of a more\nphysically minimalistic lifestyle greatly dropped.</p>\n<p>With the last ten years of progress of software quality, together\nwith modern AI, I would argue that we now have the wave of technology\nneeded to similarly enable the next wave of minimalism: digital\nminimalism. We need social media less than before, because you can talk\nto the bot for information instead (and for human connection, there&#39;s\nstill the old-fashioned solution: going out and talking to people in\nreal life). We need the internet less than before, because there&#39;s so\nmuch more you can do purely locally and offline. And we need the web\nless than before, because you can do much more, and more conveniently,\nthough either command-line tools or custom-made user interfaces, and AI\ncan help you use or craft both. And these things are all happening at\nexactly the time that the flaws of these previous paradigms - and, for\nthat matter, the flaws of using AI in the wrong way - are reaching their\napex, and so a digital minimalism is becoming more and more\nnecessary.</p>\n<p>Each person&#39;s path to digital minimalism will be different, as each\nperson has different needs. But I highly recommend everyone try reducing\ntheir digital clutter in at least one important way, and if you like the\nresult then keep going.</p>","headings":[{"level":1,"text":"Digital minimalism","id":"digital-minimalism"},{"level":2,"text":"Social media minimalism","id":"social-media-minimalism"},{"level":2,"text":"Cell data minimalism","id":"cell-data-minimalism"},{"level":2,"text":"Web minimalism","id":"web-minimalism"},{"level":2,"text":"Conclusion","id":"conclusion"}]}}