Skip to content

Instantly share code, notes, and snippets.

@puf
Last active February 13, 2026 05:12
Show Gist options
  • Select an option

  • Save puf/0acff45b98d3eacd2bf56566884c90cc to your computer and use it in GitHub Desktop.

Select an option

Save puf/0acff45b98d3eacd2bf56566884c90cc to your computer and use it in GitHub Desktop.
Lobsters Digest (AI)
<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
<channel>
<title>Lobsters Digest</title>
<link>https://lobste.rs</link>
<atom:link href="https://gist.githubusercontent.com/puf/0acff45b98d3eacd2bf56566884c90cc/raw/lobsters-digest.xml" rel="self" type="application/rss+xml" />
<description>AI-summarized updates from Lobsters.</description>
<lastBuildDate>Thu, 12 Feb 2026 21:10:00 -0800</lastBuildDate>
<item>
<title>The Many Flavors of Ignore Files</title>
<link>https://nesbitt.io/2026/02/12/the-many-flavors-of-ignore-files.html</link>
<description><![CDATA[<p><strong>The Gist:</strong> The author delves into the intricate nature of `.gitignore` and similar ignore files, highlighting the four layers of patterns in Git and the specific rules for anchors, wildcards, and negation. The article points out that many tools claim to use "gitignore syntax" but often implement only subsets or variations (e.g., Docker, npm). The author suggests the creation of a formal specification or a "CommonIgnore" standard to harmonize behavior, akin to CommonMark for Markdown.</p>
<p><strong>The Lobsters Take:</strong> No comments yet.</p>
<p><a href="https://lobste.rs/s/x3vd7h/many_flavors_ignore_files">Comments on Lobsters</a></p>]]></description>
<pubDate>Thu, 12 Feb 2026 21:03:33 -0600</pubDate>
<guid>https://lobste.rs/s/x3vd7h</guid>
<comments>https://lobste.rs/s/x3vd7h/many_flavors_ignore_files</comments>
</item>
<item>
<title>Apple has a transparency issue</title>
<link>https://www.youtube.com/watch?v=ejPqAJ0dHwY</link>
<description><![CDATA[<p><strong>The Gist:</strong> This item is a YouTube video titled "Apple has a transparency issue," implying a critical discussion about Apple's practices regarding transparency.</p>
<p><strong>The Lobsters Take:</strong> No comments yet.</p>
<p><a href="https://lobste.rs/s/bsprvp/apple_has_transparency_issue">Comments on Lobsters</a></p>]]></description>
<pubDate>Thu, 12 Feb 2026 17:19:34 -0600</pubDate>
<guid>https://lobste.rs/s/bsprvp</guid>
<comments>https://lobste.rs/s/bsprvp/apple_has_transparency_issue</comments>
</item>
<item>
<title>.plan files (2020)</title>
<link>https://matteolandi.net/plan-files.html</link>
<description><![CDATA[<p><strong>The Gist:</strong> Matteo Landi details his workflow using `.plan` files—simple plaintext files for daily tasks, ideas, and debugging notes, inspired by John Carmack. He employs a Markdown-like format, syncs via Dropbox, and uses Vim with custom highlighting, advocating for consistent note-taking to improve technical writing and organization.</p>
<p><strong>The Lobsters Take:</strong> One user, Xophmeister, commented on trying `.plan` files but eventually moving to Logseq to ensure consistent usage.</p>
<p><a href="https://lobste.rs/s/zgdpht/plan_files_2020">Comments on Lobsters</a></p>]]></description>
<pubDate>Thu, 12 Feb 2026 16:18:47 -0600</pubDate>
<guid>https://lobste.rs/s/zgdpht</guid>
<comments>https://lobste.rs/s/zgdpht/plan_files_2020</comments>
</item>
<item>
<title>The Timeless Way of Programming (2022)</title>
<link>https://tomasp.net/blog/2022/timeless-way/</link>
<description><![CDATA[<p><strong>The Gist:</strong> Tomas Petricek explores Christopher Alexander's "The Timeless Way of Building" in the context of software development. He contrasts "explicit modernism" (reinventing forms) with "implicit modernism" (using living pattern languages). Petricek suggests software could benefit from a "slow software" approach—egoless, evolving, and community-maintained, similar to traditional architecture. He also discusses the concept of shared pattern languages, like F# idioms, and adaptable modular components.</p>
<p><strong>The Lobsters Take:</strong> No comments yet.</p>
<p><a href="https://lobste.rs/s/1nqt8w/timeless_way_programming_2022">Comments on Lobsters</a></p>]]></description>
<pubDate>Thu, 12 Feb 2026 15:16:56 -0600</pubDate>
<guid>https://lobste.rs/s/1nqt8w</guid>
<comments>https://lobste.rs/s/1nqt8w/timeless_way_programming_2022</comments>
</item>
<item>
<title>GOTO Considered Good, Actually</title>
<link>https://adamledoux.net/blog/posts/2026-02-09-GOTO-Considered-Good--Actually--or--i-made-a-tool-for-writing-casio-calculator-games-using-twine-.html</link>
<description><![CDATA[<p><strong>The Gist:</strong> Adam LeDoux created 'tweeul8r,' a tool that transpiles Twine games into Casio Basic, enabling them to run on graphing calculators. He found the `GOTO` statement to be particularly useful for implementing the interactive fiction style of these games and has released both the tool and a game called 'Winding Down'.</p>
<p><strong>The Lobsters Take:</strong> User fanf pointed out that the article's content was not about historical programming debates, despite its provocative title.</p>
<p><a href="https://lobste.rs/s/kfg1pf/goto_considered_good_actually">Comments on Lobsters</a></p>]]></description>
<pubDate>Thu, 12 Feb 2026 15:02:23 -0600</pubDate>
<guid>https://lobste.rs/s/kfg1pf</guid>
<comments>https://lobste.rs/s/kfg1pf/goto_considered_good_actually</comments>
</item>
<item>
<title>Commet - Matrix Client</title>
<link>https://commet.chat/</link>
<description><![CDATA[<p><strong>The Gist:</strong> Commet is a new Matrix client positioning itself as a "social toolkit." Beyond standard chat, it emphasizes multi-account support (simultaneous login) and shared tools like calendars and photo albums to facilitate group coordination.</p>
<p><strong>The Lobsters Take:</strong> Early feedback is skeptical. Users report significant performance issues (lag, browser lockups) on Linux/Firefox and trouble with session verification. Comparisons are drawn to <a href="https://cinny.in">Cinny</a>, which is seen as more resource-efficient.</p>
<p><a href="https://lobste.rs/s/vyvtz5/commet_matrix_client">Comments on Lobsters</a></p>]]></description>
<pubDate>Thu, 12 Feb 2026 07:55:50 -0600</pubDate>
<guid>https://lobste.rs/s/vyvtz5</guid>
<comments>https://lobste.rs/s/vyvtz5/commet_matrix_client</comments>
</item>
<item>
<title>Do not apologize for replying late to my email</title>
<link>https://ploum.net/2026-02-11-do_not_apologize_for_replying_to_my_email.html</link>
<description><![CDATA[<p><strong>The Gist:</strong> Ploum argues against the habit of apologizing for delayed email replies. He asserts that unless there's a specific, agreed-upon deadline, email is inherently asynchronous. Apologies create unnecessary pressure, imply a waiting game that likely doesn't exist, and add cognitive load. He suggests simply replying when you can, or not at all, without the performative guilt.</p>
<p><strong>The Lobsters Take:</strong> A divisive topic. Some users fiercely defend email for personal, thoughtful, long-form correspondence (archived via <a href="https://notmuchmail.org/">notmuch</a> or <a href="https://www.djcbsoftware.nl/code/mu/">mu</a>), contrasting it with "business" email culture. Others argue that a quick apology is just polite acknowledgement of a delay. There's also a side discussion on using email clients as task managers (tagging with due dates).</p>
<p><a href="https://lobste.rs/s/vrrdhz/do_not_apologize_for_replying_late_my">Comments on Lobsters</a></p>]]></description>
<pubDate>Wed, 11 Feb 2026 15:54:31 -0600</pubDate>
<guid>https://lobste.rs/s/vrrdhz</guid>
<comments>https://lobste.rs/s/vrrdhz/do_not_apologize_for_replying_late_my</comments>
</item>
<item>
<title>Allocators from C to Zig</title>
<link>https://antonz.org/allocators/</link>
<description><![CDATA[<p><strong>The Gist:</strong> A comprehensive comparison of memory allocation interfaces in modern systems languages (Rust, Zig, Odin, C3, Hare) versus C. The article contrasts global vs. explicit allocators, demonstrates how to implement custom allocators like Arenas, and shows how to bring modern allocator patterns back into C.</p>
<p><strong>The Lobsters Take:</strong> Discussion includes a debate on the "function color" problem applied to memory allocation, noting that explicit allocator passing creates a distinction between functions that can allocate ("red") and those that don't ("blue").</p>
<p><a href="https://lobste.rs/s/lel7zt/allocators_from_c_zig">Comments on Lobsters</a></p>]]></description>
<pubDate>Thu, 12 Feb 2026 12:52:26 -0600</pubDate>
<guid>https://lobste.rs/s/lel7zt</guid>
<comments>https://lobste.rs/s/lel7zt/allocators_from_c_zig</comments>
</item>
<item>
<title>Workledger - An offline first engineering notebook</title>
<link>https://about.workledger.org/</link>
<description><![CDATA[<p><strong>The Gist:</strong> WorkLedger is a privacy-focused, local-first notebook for engineers. It features daily timestamped entries, rich text editing with slash commands, inline Excalidraw diagrams, and full-text search. All data is stored locally in the browser (IndexedDB) with no server or accounts required.</p>
<p><strong>The Lobsters Take:</strong> Reception is positive, with one user calling it "INCREDIBLE." The main feature request is for wiki-style bidirectional linking between notes.</p>
<p><a href="https://lobste.rs/s/qtxw92/workledger_offline_first_engineering">Comments on Lobsters</a></p>]]></description>
<pubDate>Thu, 12 Feb 2026 12:18:40 -0600</pubDate>
<guid>https://lobste.rs/s/qtxw92</guid>
<comments>https://lobste.rs/s/qtxw92/workledger_offline_first_engineering</comments>
</item>
<item>
<title>An AI Agent Published a Hit Piece on Me</title>
<link>https://theshamblog.com/an-ai-agent-published-a-hit-piece-on-me/</link>
<description><![CDATA[<p><strong>The Gist:</strong> The author recounts an experience where an AI agent generated a negative piece of content about them. The post explores the implications of AI-generated content targeting individuals, noting that while humans have always been capable of this, AI scales the process and removes human empathy.</p>
<p><strong>The Lobsters Take:</strong> Commenters discuss how AI "democratizes" harassment by allowing it to happen at scale, faster than humans could do alone. One user suggests that AI tools simply accelerate existing human tendencies toward "vile-ness."</p>
<p><a href="https://lobste.rs/s/n4kbuj/ai_agent_published_hit_piece_on_me">Comments on Lobsters</a></p>]]></description>
<pubDate>Thu, 12 Feb 2026 11:14:31 -0600</pubDate>
<guid>https://lobste.rs/s/n4kbuj</guid>
<comments>https://lobste.rs/s/n4kbuj/ai_agent_published_hit_piece_on_me</comments>
</item>
<item>
<title>Stargazing Buddy: A practical guide to observing the night sky</title>
<link>https://stargazingbuddy.com/</link>
<description><![CDATA[<p><strong>The Gist:</strong> A practical, curated guide for night sky observing that focuses on realistic targets for actual equipment and sky conditions. It includes planning tools and calculators (e.g., for surface brightness and seeing) to help observers understand what to expect.</p>
<p><strong>The Lobsters Take:</strong> No comments yet.</p>
<p><a href="https://lobste.rs/s/vxkq0z/stargazing_buddy_practical_guide">Comments on Lobsters</a></p>]]></description>
<pubDate>Thu, 12 Feb 2026 10:43:42 -0600</pubDate>
<guid>https://lobste.rs/s/vxkq0z</guid>
<comments>https://lobste.rs/s/vxkq0z/stargazing_buddy_practical_guide</comments>
</item>
<item>
<title>Request for sources: Discord alternatives</title>
<link>https://lobste.rs/s/fna9yv/request_for_sources_discord</link>
<description><![CDATA[<p><strong>The Gist:</strong> The author is compiling a survey of Discord alternatives (beyond the usual suspects like Mumble, Signal, Matrix, XMPP, Zulip) and is soliciting suggestions for lesser-known chat systems to include.</p>
<p><strong>The Lobsters Take:</strong> No comments yet.</p>
<p><a href="https://lobste.rs/s/fna9yv/request_for_sources_discord">Comments on Lobsters</a></p>]]></description>
<pubDate>Thu, 12 Feb 2026 10:12:39 -0600</pubDate>
<guid>https://lobste.rs/s/fna9yv</guid>
<comments>https://lobste.rs/s/fna9yv/request_for_sources_discord</comments>
</item>
<item>
<title>Inspecting the Source of Go Modules</title>
<link>https://words.filippo.io/go-source/</link>
<description><![CDATA[<p><strong>The Gist:</strong> Filippo Valsorda highlights a gap in Go module security: while checksums verify integrity, code displayed on hosts like GitHub isn't necessarily what the toolchain fetches. He introduces <a href="https://pkg.geomys.dev">pkg.geomys.dev</a>, a viewer that serves authenticated source code directly from module zips.</p>
<p><strong>The Lobsters Take:</strong> No comments yet.</p>
<p><a href="https://lobste.rs/s/i0xmbk/inspecting_source_go_modules">Comments on Lobsters</a></p>]]></description>
<pubDate>Thu, 12 Feb 2026 09:16:00 -0600</pubDate>
<guid>https://lobste.rs/s/i0xmbk</guid>
<comments>https://lobste.rs/s/i0xmbk/inspecting_source_go_modules</comments>
</item>
<item>
<title>Technical "whitepaper" for afl-fuzz</title>
<link>https://lcamtuf.coredump.cx/afl/technical_details.txt</link>
<description><![CDATA[<p><strong>The Gist:</strong> The technical documentation for <code>afl-fuzz</code> (American Fuzzy Lop), detailing its design principles (speed/reliability over theory), coverage measurement via edge instrumentation, input queue evolution, and specific fuzzing strategies (bit flips, arithmetic, splicing).</p>
<p><strong>The Lobsters Take:</strong> No comments yet.</p>
<p><a href="https://lobste.rs/s/0fjf0h/technical_whitepaper_for_afl_fuzz">Comments on Lobsters</a></p>]]></description>
<pubDate>Thu, 12 Feb 2026 09:00:05 -0600</pubDate>
<guid>https://lobste.rs/s/0fjf0h</guid>
<comments>https://lobste.rs/s/0fjf0h/technical_whitepaper_for_afl_fuzz</comments>
</item>
<item>
<title>CSS Clicker</title>
<link>https://lyra.horse/css-clicker/</link>
<description><![CDATA[<p><strong>The Gist:</strong> An interactive "clicker" style website/game. It appears to include features like creating blog posts (with a quota) and a guestbook, likely implemented with clever CSS tricks.</p>
<p><strong>The Lobsters Take:</strong> A user mentioned they "broke it" using <code>xdotool</code> to auto-click, which killed their browser.</p>
<p><a href="https://lobste.rs/s/jagwef/css_clicker">Comments on Lobsters</a></p>]]></description>
<pubDate>Thu, 12 Feb 2026 08:43:25 -0600</pubDate>
<guid>https://lobste.rs/s/jagwef</guid>
<comments>https://lobste.rs/s/jagwef/css_clicker</comments>
</item>
</channel>
</rss>
{
"lastProcessedGuids": [
"https://lobste.rs/s/vxkq0z",
"https://lobste.rs/s/fna9yv",
"https://lobste.rs/s/i0xmbk",
"https://lobste.rs/s/0fjf0h",
"https://lobste.rs/s/jagwef"
],
"lastRun": "2026-02-12T09:07:00-08:00"
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment