<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: hun3</title><link>https://news.ycombinator.com/user?id=hun3</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Sat, 13 Jun 2026 19:36:47 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=hun3" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by hun3 in "Anthropic requires 30 day data retention for Fable and Mythos"]]></title><description><![CDATA[
<p>China has proxies that sell cheaper access to frontier models in exchange for permission to train on your data.</p>
]]></description><pubDate>Thu, 11 Jun 2026 03:43:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=48485963</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=48485963</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48485963</guid></item><item><title><![CDATA[New comment by hun3 in "Do we fear the serializable isolation level more than we fear subtle bugs?"]]></title><description><![CDATA[
<p>Right.<p>As an example: Oracle and PostgreSQL don't have dirty reads: READ UNCOMMITTED does nothing. MySQL's concurrency model depends on the engine.</p>
]]></description><pubDate>Sun, 07 Jun 2026 23:11:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=48439596</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=48439596</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48439596</guid></item><item><title><![CDATA[New comment by hun3 in "Ask HN: Why is the HN crowd so anti-AI?"]]></title><description><![CDATA[
<p>What keeps users themselves from vibe-coding their own software?<p>If you don't own quality, why should I pay? You're just a middleman at that point.</p>
]]></description><pubDate>Sat, 06 Jun 2026 04:58:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=48421543</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=48421543</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48421543</guid></item><item><title><![CDATA[New comment by hun3 in "Claude Code and Codex can have real-time conversation via Git"]]></title><description><![CDATA[
<p>If you squint hard enough, you'll notice your bank account serves as an IPC semaphore replenishing API credit balances.<p>Your wallet is now a real-time communication channel.</p>
]]></description><pubDate>Thu, 04 Jun 2026 10:57:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=48396864</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=48396864</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48396864</guid></item><item><title><![CDATA[New comment by hun3 in "Ask HN: Shouldn't Google need to give a public statement about Railway incident?"]]></title><description><![CDATA[
<p>With consent, yes.</p>
]]></description><pubDate>Wed, 20 May 2026 17:31:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=48211148</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=48211148</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48211148</guid></item><item><title><![CDATA[New comment by hun3 in "CopyFail: From Pod to Host"]]></title><description><![CDATA[
<p>No, "virtual machine" alone doesn't make things safer.<p>Shrink your attack surface.<p>Use a completely locked down seccomp. Use nsjail or gVisor for containers. Use microvm or libkrun for full OS.<p>Lesser attack surface is what matters. Virtualization is only half of the story.</p>
]]></description><pubDate>Wed, 20 May 2026 12:26:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=48206594</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=48206594</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48206594</guid></item><item><title><![CDATA[New comment by hun3 in "Why senior developers fail to communicate their expertise"]]></title><description><![CDATA[
<p>Avoiding problems outside your business problem domain, and can therefore guide the AI more effectively towards building the right thing.</p>
]]></description><pubDate>Wed, 13 May 2026 01:35:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=48116809</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=48116809</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48116809</guid></item><item><title><![CDATA[New comment by hun3 in "Ask HN: We just had an actual UUID v4 collision..."]]></title><description><![CDATA[
<p>(Note that this still assumes that each biased-coin toss is i.i.d.)</p>
]]></description><pubDate>Sat, 09 May 2026 10:50:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=48073868</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=48073868</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48073868</guid></item><item><title><![CDATA[New comment by hun3 in "How OpenAI delivers low-latency voice AI at scale"]]></title><description><![CDATA[
<p>They are orthogonal.<p>Suppose you have 100ms audio latency <i>and</i> no wait time. Then, natural pause will trigger response immediately <i>but</i> you won't notice it has started until after ~200ms (round-trip time). Twice as annoying.</p>
]]></description><pubDate>Mon, 04 May 2026 23:53:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=48016450</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=48016450</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48016450</guid></item><item><title><![CDATA[New comment by hun3 in "An AI agent deleted our production database. The agent's confession is below"]]></title><description><![CDATA[
<p>But it's still a bit more difficult to <i>sue</i> them for leaking your company's data.<p>At least for now.</p>
]]></description><pubDate>Sun, 26 Apr 2026 20:44:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=47914204</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=47914204</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47914204</guid></item><item><title><![CDATA[New comment by hun3 in "AI should elevate your thinking, not replace it"]]></title><description><![CDATA[
<p>You can write unambiguous (UB-free) code and the compiler's output will be deterministic. There will even be a spec that explains how your source maps to your program's behavior. LLM has neither.<p>Also, if you need to control performance, you still need to know how CPU cache and branch prediction works, both of which exists at the abstraction level of assembly.</p>
]]></description><pubDate>Sun, 26 Apr 2026 20:35:41 +0000</pubDate><link>https://news.ycombinator.com/item?id=47914092</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=47914092</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47914092</guid></item><item><title><![CDATA[New comment by hun3 in "AI should elevate your thinking, not replace it"]]></title><description><![CDATA[
<p>How can you read a language you didn't learn?</p>
]]></description><pubDate>Sun, 26 Apr 2026 20:24:45 +0000</pubDate><link>https://news.ycombinator.com/item?id=47913938</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=47913938</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47913938</guid></item><item><title><![CDATA[New comment by hun3 in "OpenClaw isn't fooling me. I remember MS-DOS"]]></title><description><![CDATA[
<p>PatchGuard would like to have a word with you.</p>
]]></description><pubDate>Wed, 22 Apr 2026 06:46:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=47859924</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=47859924</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47859924</guid></item><item><title><![CDATA[New comment by hun3 in "OpenClaw isn't fooling me. I remember MS-DOS"]]></title><description><![CDATA[
<p>Ironic, since Unix is itself an embodiment of "worse is better."</p>
]]></description><pubDate>Wed, 22 Apr 2026 06:42:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=47859894</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=47859894</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47859894</guid></item><item><title><![CDATA[New comment by hun3 in "It's OK to compare floating-points for equality"]]></title><description><![CDATA[
<p>I used floating timestamps as some kind of an identity. If there is ever a conflict, I just increase it by 1 ulp until it doesn't collide with anything. Sorry.</p>
]]></description><pubDate>Sat, 18 Apr 2026 15:58:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=47816913</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=47816913</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47816913</guid></item><item><title><![CDATA[New comment by hun3 in "Safe ways to do things in bash (2023)"]]></title><description><![CDATA[
<p>In Python, `shlex` has utilities for quoting and tokenizing strings for POSIX shells.<p>> Documentation: <a href="https://docs.python.org/3/library/shlex.html" rel="nofollow">https://docs.python.org/3/library/shlex.html</a><p>(Windows users: CMD is hopeless. Sorry. Try powershell.)</p>
]]></description><pubDate>Wed, 01 Apr 2026 06:08:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=47597408</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=47597408</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47597408</guid></item><item><title><![CDATA[New comment by hun3 in "DOOM Over DNS"]]></title><description><![CDATA[
<p>Finally, a DOOM download that bypasses captive portals</p>
]]></description><pubDate>Thu, 26 Mar 2026 18:56:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=47534265</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=47534265</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47534265</guid></item><item><title><![CDATA[New comment by hun3 in "Autoresearch on an old research idea"]]></title><description><![CDATA[
<p>> There are better techniques for hyper-parameter optimisation, right?<p>There always are. You need to think about what those would be, though. Autoresearch outsources the thinking to LLMs.</p>
]]></description><pubDate>Mon, 23 Mar 2026 19:40:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=47494161</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=47494161</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47494161</guid></item><item><title><![CDATA[New comment by hun3 in "Show HN: Threadprocs – executables sharing one address space (0-copy pointers)"]]></title><description><![CDATA[
<p>This is exactly what subinterpreters are for! Basically isolated copies of Python in the same process.<p><a href="https://docs.python.org/3/library/concurrent.interpreters.html" rel="nofollow">https://docs.python.org/3/library/concurrent.interpreters.ht...</a><p>If you want a higher-level interface, there is InterpreterPoolExecutor:<p><a href="https://docs.python.org/3/library/concurrent.futures.html#concurrent.futures.InterpreterPoolExecutor" rel="nofollow">https://docs.python.org/3/library/concurrent.futures.html#co...</a></p>
]]></description><pubDate>Mon, 23 Mar 2026 17:07:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=47492226</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=47492226</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47492226</guid></item><item><title><![CDATA[New comment by hun3 in "Astral to Join OpenAI"]]></title><description><![CDATA[
<p>> why we keep trusting venture capital to be the community's stewards I have no idea.<p>They bought the trust.</p>
]]></description><pubDate>Thu, 19 Mar 2026 22:13:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=47447056</link><dc:creator>hun3</dc:creator><comments>https://news.ycombinator.com/item?id=47447056</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47447056</guid></item></channel></rss>