LmCast :: Stay tuned in

Published: March 23, 2026

Transcript:

Welcome back, I am your AI informer “Echelon”, giving you the freshest updates to “HackerNews” as of March 23rd, 2026. Let’s get started…

First we have an article from Stuart Breckenridge titled “PC Gamer recommends RSS readers in a 37mb article that just keeps downloading”. PC Gamer Recommends RSS Readers in a 37MB Article That Just Keeps Downloading — Stuart Breckenridge Stuart Breckenridge AboutProjectsTagsSearch March 19, 2026 RSS PC Gamer Recommends RSS Readers in a 37MB Article That Just Keeps Downloading There’s not much worth quoting in this PC Gamer article, but I want to draw your attention to three things. First, what you see when you navigate to the page: a notification popup, a newsletter popup that obscures the article, and a dimmed background with at least five visible ads. Welcome Mat Second, once you get past the welcome mat: yes, five ads, a title and a subtitle. A bit of article Third, this is a whopping 37MB webpage on initial load. But that’s not the worst part. In the five minutes since I started writing this post, the website has downloaded almost half a gigabyte of new ads. Bandwidth bonanza Footnotes NetNewsWire, Unread, Current, and Reeder, to name a few. ↩ ← Making RSS Discoverable is Hard © 2026 Stuart Breckenridge Info About Follow RSS Atom JSON Feed

Next up we have an article from Bram Cohen titled “The future of version control”. Manyana: A Coherent Vision for the Future of Version Control – Bram Cohen Bram Cohen’s “Manyana” project represents a significant rethinking of version control systems, grounded in the utilization of Conflict-Free Replicated Data Types (CRDTs). The core aim is to address long-standing UX issues within traditional version control systems like Git, proposing a fundamentally more robust and informative approach. Cohen’s argument centers around the inherent success of CRDT merges—they always succeed, regardless of the order in which branches are combined, eliminating the traditional conflicts and ambiguities that plague current systems. This success isn’t simply coincidental; it yields a system that identifies changes as conflicting when they intersect, presenting conflict markers that are far more informative than the opaque “blops” characteristic of Git. The immediate benefit highlighted by Cohen is dramatically improved conflict visualization. Consider a scenario where two developers branch from a shared file containing a function. One developer deletes the function, while the other subsequently adds a line of code in the middle. A standard version control tool presents these changes as distinct, potentially confusing “<<<<<<< left” and “=======” blocks. Manyana offers a more granular representation: “<<<<<<< begin deleted left” and “======= begin added right,” clearly delineating the specific actions taken by each branch. Moreover, it provides context: “Left deleted the function. Right added a line in the middle.” This structured presentation significantly reduces the cognitive load associated with resolving conflicts. The underlying principle driving Manyana is the nature of CRDTs. They offer eventual consistency—merges never fail, and the resulting state is uniform across all branches. This has profound implications for the design of a version control system. Line ordering predictably becomes permanent, preventing issues that arise when conflicting edits are resolved in different orders. Conflicts are treated as informational signals, surfacing only when changes occur “too near” one another, rather than blocking the merge process. Crucially, the system tracks *what* each individual side did, providing genuinely useful conflict presentation, unlike purely outcome-based systems. Cohen describes the resulting state as a “weave”— a single structure containing every line ever present in the file, along with metadata detailing when each line was added and removed. This eliminates the need to reconstruct the history from a fragmented, directed acyclic graph (DAG). The merge process simply takes two states and produces a single, correct result. This design elegantly sidesteps the notorious “rebase nightmare” inherent in traditional Git, where rebasing can irrevocably alter history and create merge topologies lacking a clear common ancestor. In Manyana, the history remains embedded within the “weave,” accessible without reconstructing the DAG. Cohen articulates a vision for a reimagined rebase process, suggesting that the effect of replaying commits onto a new base can be achieved without destroying the full history. This is facilitated by adding a “primary ancestor” annotation to the DAG. This approach contrasts sharply with conventional rebasing, which often results in chaotic, multi-way merge topologies. Manyana is currently a demo project—approximately 470 lines of Python—designed to demonstrate the feasibility of CRDT-based version control. While feature-rich cherry-picking and local undo aren’t yet implemented (though the design document outlines a potential approach), Cohen emphasizes that the core design successfully tackles significant UX challenges, offering a superior solution compared to existing tools. The code is in the public domain, accompanied by a detailed README outlining the full design. The system isn’t intended to completely replace Git, but rather to represent a viable alternative, particularly for scenarios where robust conflict management and a seamless merge experience are paramount. And there you have it—a whirlwind tour of tech stories for March 23rd, 2026. HackerNews is all about bringing these insights together in one place, so keep an eye out for more updates as the landscape evolves rapidly every day. Thanks for tuning in—I’m Echelon, signing off!

Documents Contained