The Best Coders Should Exit the Feed  ## Metadata - Author: [[The Changelog: Software Development, Open Source]] - Full Title: The Best Coders Should Exit the Feed - Category: #podcasts - URL: https://share.snipd.com/episode/66d10289-c17f-4c94-9ef0-1a312161d1eb ## Document Notes ## Highlights - **Mass Data Loss From No Viable Off-Site Backups** - South Korea's National Information Resources Service lost work files from 750,000 civil servants after a fire destroyed primary data and backups. - Jerod notes this as a textbook failure of backup strategy and invokes the 3-2 rule for reliable backups. Transcript: Jerod Santo Did you hear what happened to South Korea's government recently? A fire broke out at their National Information Resources Service Headquarters, erasing work files saved by about 750,000 civil servants over the past seven years. The worst part? No backups. What, what, what? Well, that's not technically correct. They did have backups, but the backups were also destroyed in the same fire. This is why the 3-2 rule exists. Your backup strategy should have three copies, two different media types, and one off-site. ([Time 0:00:15](https://share.snipd.com/snip/de3a4fe7-cdf7-4d53-ae2f-1356883dc971)) - **Top Coders Should Leave Social Platforms** - Abner Coimbre argues top technical talent should leave for-profit social platforms because those platforms are collapsing under regulation and cultural backlash. - The shift away is driven by parents, school bans, visible bot farms, and links between being online and political violence. Transcript: Jerod Santo The best coders should exit the feed. Abner Quamber makes a compelling case why our biggest technical talent should abandon for-profit social platforms like X, Instagram, and Reddit. Quote, not as a moral crusade, but because the platforms are crumbling under heavy regulation and cultural backlash. If you're investing your peak years shaping public narratives there, you risk reputation and creativity. End quote. Abner has noticed four big forces pushing the shift away. One, the Anxious Generation book convincing parents to restrict phones. Two, bipartisan school smartphone bans. Three, growing evidence and visibility of bot farms. And four, a cultural link between being terminally online and political violence. My own social network of family, friends, and church, and my personal sentiment toward technology affirms what Abner is saying. Phone fasts and a return to analog are common discussions I'm having with people. Abner says, quote, because of everything I've already outlined, this way of life is barreling towards collapse. You can also find smaller signs, like the surge in dumb phone sales, that mainstream outlets are finally noticing. Everywhere around us, society is wondering, why do we keep a bottomless barrel of sugar in our pockets? End quote. His call for anyone in tech helping build these platforms is to stop that and start building technology for touching grass. And he says that movement has already begun. ([Time 0:00:55](https://share.snipd.com/snip/6f120626-224a-4f26-b4e6-c87c3de31939)) - **Use Cloud Code With An Obsidian Vault** - Use Cloud Code with an Obsidian vault configured for PARA to treat AI as a thinking partner rather than just a writing tool. - Maintain Git versioning and mobile access so your knowledge vault is portable and auditable. Transcript: Jerod Santo A clod code plus obsidian starter kit. Clod code and obsidian? Cue Julie Andrews because these are definitely two of my favorite things right now. The linked repo is a pre-configured obsidian vault structure that's designed to work seamlessly with clod code as an AI-powered second brain. This lets you, one, use AI as a thinking partner, not just a writing assistant. Two, organize knowledge using the PARA method, that's projects, areas, resources, and archives. Three, maintain version control with Git. And four, access your vault from anywhere, including mobile. The Starter Kit's creator, Noah Breyer, wrote all about why he loves this setup in The Magic of Cloud Code. ([Time 0:02:30](https://share.snipd.com/snip/6002d46c-7a2e-42e6-824f-7ef447094d18)) - **Vercel Versus Cloudflare Friction** - The Vercel vs Cloudflare fight centers on tradeoffs between beautiful abstractions and hidden costs or centralized security claims. - Each camp accuses the other of hiding complexity or masking costs while claiming better developer experience or security. Transcript: Jerod Santo Vercel versus Cloudflare. Vercel and Cloudflare are seriously duking it out for developer mindshare these days, and their founders aren't pulling any punches. In the linked post, Bharath Natarajan documents the fight. Quote, Cloudflare's camp frames Vercel as a front-end illusionist. Beautiful defaults masking AWS egress bills. They point to screenshots of cost breakdowns, claiming that what looks like free quickly becomes expensive once your app scales. Cloudflare's arguments against Vercel can be distilled into one line. The people designing your abstractions don't understand the systems they're built on. Vercel's camp fires back with equal conviction. Gishirimo accuses Cloudflare of security theater, centralizing the web under the guise of protection, hiding pricing under tooltips, selling bandwidth under new names. His tone is defensive but principled. Vercel's abstraction isn't ignorance, it's empathy. The idea is that the best infrastructure is invisible and the worst infrastructure is the one you have to keep configuring. End quote. Click through for many embedded tweets, a summary of the recurring themes, and Barath's personal take on who will win. ([Time 0:03:18](https://share.snipd.com/snip/bd4fbfe3-e6c4-4400-a89d-4623d58c6155)) - **Try Augment's Indie Plan** - Indie developers can get repo-aware AI coding features for an affordable monthly price to accelerate shipping. - Use Augment's indie plan to access refactors, context-rich reviews, and agent features for $20/month. Transcript: Jerod Santo Augment's indie plan is here. Augment just got a lot friendlier for indie devs. Until now, their pricing was geared towards teams. But with the new indie plan, you can get Augment's full AI coding superpowers for just 20 bucks a month. Why is this a big deal? Because indie hackers, solo founders, and side project dreamers don't always have an enterprise budget for AI tools. But they do need that same leverage. The indie plan brings Augment's repo-aware refactors, context-rich reviews, and agent features into reach for everyone. Now you can go from idea to shipped for $20 a month. ([Time 0:04:31](https://share.snipd.com/snip/e257148c-d339-4b38-ad94-db60dc942d08)) - **Toolbrew: No-Friction Utility Site** - Toolbrew provides many small utilities without signups or ads, making it handy for quick tasks like conversions or counters. - The site accepts suggestions and implements top-requested tools, emphasizing practical utility over monetization. Transcript: Jerod Santo Free little tools without signups or ads. Toolbrew is a well-designed website brimming with common utilities. If you ever need a quick case converter, a word counter, a URL encoder, a video downloader, a check splitter, a PDF converter, et cetera, check this out. There's no need to sign up, there's no ads, and you can suggest new tools to be added to the site. ([Time 0:05:12](https://share.snipd.com/snip/d52ac422-2725-42c0-aad0-1722ef8209c2)) - **Why Overengineering Happens** - Yusuf Aytas explains overengineering as impressive-on-paper systems solving nonexistent problems, likening them to Rube Goldberg machines. - Drivers include premature optimization, resume-driven development, management incentives, FOMO, and misaligned priorities. Transcript: Jerod Santo Why overengineering happens. In this article, Yusuf Atis does an excellent job defining overengineering, explaining why it happens, how much it costs, and what we can do to avoid it. The setup, quote, if you've worked in software long enough, you've probably seen it. A CRUD app serving a handful of users deployed on a Kubernetes cluster with half the CNCF landscape stitched together for good measure. On paper, it looks impressive. In reality, it's a Rube Goldberg machine solving problems the team doesn't actually have. End quote. Sound familiar? When it comes to the why, Yusuf blames good intentions gone sideways, misaligned incentives, or plain old human nature. The biggest drivers. 1. Premature optimization. 2. Resume-driven development. 3. Management incentives. 4. FOMO and trend chasing. And five, misaligned priorities. These forces are troublesome alone, but when combined, they can add up to some seriously over-engineered, seriously expensive systems. ([Time 0:05:39](https://share.snipd.com/snip/0a792f42-dcd3-4809-84a1-b593fa4ea634))