<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: lights0123</title><link>https://news.ycombinator.com/user?id=lights0123</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 13 Apr 2026 18:15:05 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=lights0123" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by lights0123 in "My university uses prompt injection to catch cheaters"]]></title><description><![CDATA[
<p>I use a prompt like this that asks for model name and version! It's been effective so far, especially since I have edit history.</p>
]]></description><pubDate>Mon, 06 Apr 2026 21:58:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=47667806</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=47667806</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47667806</guid></item><item><title><![CDATA[New comment by lights0123 in "Root Persistence via macOS Recovery Mode Safari"]]></title><description><![CDATA[
<p>Yep. While the Terminal is not an option from the 4 apps listed in the initial screen, it's available from Utilities → Terminal at the top. They even provide a convenient way to access the hard drive from another computer: <a href="https://support.apple.com/guide/mac-help/macos-recovery-a-mac-apple-silicon-mchl82829c17/26/mac/26#mchl6e3b9225" rel="nofollow">https://support.apple.com/guide/mac-help/macos-recovery-a-ma...</a></p>
]]></description><pubDate>Mon, 06 Apr 2026 21:10:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=47667164</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=47667164</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47667164</guid></item><item><title><![CDATA[New comment by lights0123 in "SSH certificates: the better SSH experience"]]></title><description><![CDATA[
<p>They do, for Enterprise customers only: <a href="https://docs.github.com/en/enterprise-cloud@latest/organizations/managing-git-access-to-your-organizations-repositories/about-ssh-certificate-authorities" rel="nofollow">https://docs.github.com/en/enterprise-cloud@latest/organizat...</a><p>They've rolled their host key one time, so there's little reason for them to use it on the host side.</p>
]]></description><pubDate>Fri, 03 Apr 2026 14:11:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=47626856</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=47626856</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47626856</guid></item><item><title><![CDATA[New comment by lights0123 in "Zstandard Across the Stack"]]></title><description><![CDATA[
<p>Nice! There's also zstd's flush ability that I've used for streaming robotics data. You can write data and flush it over the network for realtime updates, but the compression stream stays open so it can still reference past messages. This means messages get smaller over time so you don't need to share a dictionary ahead of time. I'm not aware of other compression algorithms that have flushing capability like this.<p>> binary data to connected clients in tiny messages, each saying “field 5 on object X is now 123”<p>I wonder how Meta's newer, format-understanding OpenZL would do. I imagine its schemas could be auto-generated from protobuf.</p>
]]></description><pubDate>Thu, 02 Apr 2026 19:21:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=47618968</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=47618968</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47618968</guid></item><item><title><![CDATA[New comment by lights0123 in "Zstandard Across the Stack"]]></title><description><![CDATA[
<p>They moved on to Courgette, then to Zucchini: <a href="https://chromium.googlesource.com/chromium/src/+/HEAD/components/zucchini/" rel="nofollow">https://chromium.googlesource.com/chromium/src/+/HEAD/compon...</a><p>These are optimized for compiled code though.</p>
]]></description><pubDate>Thu, 02 Apr 2026 19:16:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=47618905</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=47618905</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47618905</guid></item><item><title><![CDATA[New comment by lights0123 in "Show HN: Zerobox – Sandbox any command with file, network, credential controls"]]></title><description><![CDATA[
<p>> zerobox --secret OPENAI_API_KEY=$OPENAI_API_KEY<p>Linux by default allows all users to read CLI arguments of running processes. While it looks like your bwrap invocation prevents the sandbox from looking at this process (--unshare-pid), any other process running on your system can read the secret.</p>
]]></description><pubDate>Wed, 01 Apr 2026 21:01:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=47606505</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=47606505</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47606505</guid></item><item><title><![CDATA[New comment by lights0123 in "This specific GitHub issue is crashing"]]></title><description><![CDATA[
<p><i>all</i> of GitHub is crashing right now, even though githubstatus.com only mentions pull requests being an issue.</p>
]]></description><pubDate>Tue, 31 Mar 2026 15:44:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=47589093</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=47589093</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47589093</guid></item><item><title><![CDATA[New comment by lights0123 in "CSS is DOOMed"]]></title><description><![CDATA[
<p>> Interestingly, it was more choppy in Chromium.<p>Firefox's WebRender is truly a great creation. While Chrome is faster at most things especially involving JS, Firefox puts so much of its rendering on the GPU so moving elements around is incredibly fast.</p>
]]></description><pubDate>Sat, 28 Mar 2026 23:40:30 +0000</pubDate><link>https://news.ycombinator.com/item?id=47559059</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=47559059</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47559059</guid></item><item><title><![CDATA[New comment by lights0123 in "Cocoa-Way – Native macOS Wayland compositor for running Linux apps seamlessly"]]></title><description><![CDATA[
<p>While I agree with the rest of your comment, they do mention they use OrbStack as their hypervisor in their demo video.</p>
]]></description><pubDate>Sat, 28 Mar 2026 17:41:47 +0000</pubDate><link>https://news.ycombinator.com/item?id=47556749</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=47556749</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47556749</guid></item><item><title><![CDATA[New comment by lights0123 in "Health NZ staff told to stop using ChatGPT to write clinical notes"]]></title><description><![CDATA[
<p>Azure OpenAI is not the same as paying OpenAI directly. While you may not be able to pay OpenAI for them to run models in Australia, you can pay Azure: <a href="https://azure.microsoft.com/en-au/pricing/details/azure-openai/" rel="nofollow">https://azure.microsoft.com/en-au/pricing/details/azure-open...</a><p>The models are licensed to Microsoft, and you pay them for the inference.</p>
]]></description><pubDate>Thu, 26 Mar 2026 01:45:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=47525769</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=47525769</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47525769</guid></item><item><title><![CDATA[New comment by lights0123 in "Health NZ staff told to stop using ChatGPT to write clinical notes"]]></title><description><![CDATA[
<p>Many AI companies, including Azure with their OpenAI hosting, are more than willing to sign privacy agreements that allow processing sensitive medical data with their models.</p>
]]></description><pubDate>Wed, 25 Mar 2026 22:46:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=47524310</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=47524310</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47524310</guid></item><item><title><![CDATA[New comment by lights0123 in "Why does C have the best file API"]]></title><description><![CDATA[
<p>Yes, it’s the SIGBUS signal.</p>
]]></description><pubDate>Mon, 02 Mar 2026 00:48:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=47212514</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=47212514</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47212514</guid></item><item><title><![CDATA[New comment by lights0123 in "C++26: Std:Is_within_lifetime"]]></title><description><![CDATA[
<p>Python removes features all the time in 3.x releases. For example, I was not a fan of the distutils removal in 3.12 which broke many legacy but otherwise functional packages. Deprecated functions and classes are also removed from packages regularly.<p>They do publish removal plans years in advance, e.g. see Python 3.17's plans: <a href="https://docs.python.org/3/deprecations/pending-removal-in-3.17.html" rel="nofollow">https://docs.python.org/3/deprecations/pending-removal-in-3....</a></p>
]]></description><pubDate>Thu, 19 Feb 2026 15:09:42 +0000</pubDate><link>https://news.ycombinator.com/item?id=47074607</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=47074607</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47074607</guid></item><item><title><![CDATA[New comment by lights0123 in "Show HN: Extracting React apps from Figma Make's undocumented binary format"]]></title><description><![CDATA[
<p>I agree. It would likely have identified the separate deflate and zstd chunks automatically.</p>
]]></description><pubDate>Wed, 28 Jan 2026 16:15:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=46797299</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=46797299</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46797299</guid></item><item><title><![CDATA[New comment by lights0123 in "Dell admits it made a mistake when it abandoned XPS"]]></title><description><![CDATA[
<p>It is slightly wider than the space bar. I've never had an issue with mine, as it is located exactly where I expect it to be.</p>
]]></description><pubDate>Sun, 11 Jan 2026 19:16:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=46578850</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=46578850</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46578850</guid></item><item><title><![CDATA[New comment by lights0123 in "I program on the subway"]]></title><description><![CDATA[
<p><a href="https://zealdocs.org/" rel="nofollow">https://zealdocs.org/</a> is surprisingly decent.</p>
]]></description><pubDate>Mon, 22 Dec 2025 02:21:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=46350668</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=46350668</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46350668</guid></item><item><title><![CDATA[New comment by lights0123 in "Show HN: Shittp – Volatile Dotfiles over SSH"]]></title><description><![CDATA[
<p>Yes, Firefox 147 will respect XDG dirs.</p>
]]></description><pubDate>Sun, 21 Dec 2025 17:49:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=46346649</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=46346649</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46346649</guid></item><item><title><![CDATA[New comment by lights0123 in "Rust's Block Pattern"]]></title><description><![CDATA[
<p>GCC adds similar syntax as an extension to C: <a href="https://gcc.gnu.org/onlinedocs/gcc/Statement-Exprs.html" rel="nofollow">https://gcc.gnu.org/onlinedocs/gcc/Statement-Exprs.html</a><p>It's used all throughout the Linux kernel and useful for macros.</p>
]]></description><pubDate>Fri, 19 Dec 2025 20:36:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=46330595</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=46330595</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46330595</guid></item><item><title><![CDATA[New comment by lights0123 in "Shai-Hulud compromised a dev machine and raided GitHub org access: a post-mortem"]]></title><description><![CDATA[
<p>noexec now prevents mmaping files on that filesystem as executable.</p>
]]></description><pubDate>Mon, 15 Dec 2025 03:34:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=46270141</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=46270141</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46270141</guid></item><item><title><![CDATA[New comment by lights0123 in "Show HN: Epstein's emails reconstructed in a message-style UI (OCR and LLMs)"]]></title><description><![CDATA[
<p>See also: <a href="https://jmail.world/" rel="nofollow">https://jmail.world/</a></p>
]]></description><pubDate>Sat, 13 Dec 2025 00:57:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=46250908</link><dc:creator>lights0123</dc:creator><comments>https://news.ycombinator.com/item?id=46250908</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46250908</guid></item></channel></rss>