<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: 4diii</title><link>https://news.ycombinator.com/user?id=4diii</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Fri, 04 Sep 2026 09:20:45 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=4diii" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by 4diii in "GPT-6 Astra"]]></title><description><![CDATA[
<p>Can confirm this as well, mostly VHDL and HLS. Sol and Fable can reason about performance and designs consistently. Whereas Opus and others seem to just throw generic optimisation techniques at the wall unprovoked (while hallucinating a justification + expected improvement) until the synth reports improve.</p>
]]></description><pubDate>Fri, 04 Sep 2026 04:49:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=49560638</link><dc:creator>4diii</dc:creator><comments>https://news.ycombinator.com/item?id=49560638</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49560638</guid></item><item><title><![CDATA[How to Build a Minimal ZFS NAS Without Synology, QNAP, TrueNAS (2024)]]></title><description><![CDATA[
<p>Article URL: <a href="https://neil.computer/notes/how-to-setup-minimal-zfs-nas-without-truenas/">https://neil.computer/notes/how-to-setup-minimal-zfs-nas-without-truenas/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48827325">https://news.ycombinator.com/item?id=48827325</a></p>
<p>Points: 346</p>
<p># Comments: 250</p>
]]></description><pubDate>Wed, 08 Jul 2026 03:59:35 +0000</pubDate><link>https://neil.computer/notes/how-to-setup-minimal-zfs-nas-without-truenas/</link><dc:creator>4diii</dc:creator><comments>https://news.ycombinator.com/item?id=48827325</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48827325</guid></item><item><title><![CDATA[New comment by 4diii in "Last.fm is now independent"]]></title><description><![CDATA[
<p>Sleeve on MacOS, last.fm app on iOS (have to manually check for new scrobbles every so often and it only counts songs from your library) and a web extension scrobbler for when I’m on Linux using the web version.</p>
]]></description><pubDate>Thu, 28 May 2026 03:10:41 +0000</pubDate><link>https://news.ycombinator.com/item?id=48303974</link><dc:creator>4diii</dc:creator><comments>https://news.ycombinator.com/item?id=48303974</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48303974</guid></item><item><title><![CDATA[Are LLM merge rates not getting better?]]></title><description><![CDATA[
<p>Related: <i>Many SWE-bench-Passing PRs would not be merged</i> - <a href="https://news.ycombinator.com/item?id=47341645">https://news.ycombinator.com/item?id=47341645</a> - March 2026 (149 comments)</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47349334">https://news.ycombinator.com/item?id=47349334</a></p>
<p>Points: 174</p>
<p># Comments: 156</p>
]]></description><pubDate>Thu, 12 Mar 2026 11:49:05 +0000</pubDate><link>https://entropicthoughts.com/no-swe-bench-improvement</link><dc:creator>4diii</dc:creator><comments>https://news.ycombinator.com/item?id=47349334</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47349334</guid></item><item><title><![CDATA[New comment by 4diii in "Obsidian Sync now has a headless client"]]></title><description><![CDATA[
<p>someone reverse engineered obsidian sync a couple years ago, but obsidian ended up “patching” it. Saw some recent discussion on here about it: 
<a href="https://news.ycombinator.com/item?id=44768641">https://news.ycombinator.com/item?id=44768641</a></p>
]]></description><pubDate>Sun, 01 Mar 2026 07:41:41 +0000</pubDate><link>https://news.ycombinator.com/item?id=47204600</link><dc:creator>4diii</dc:creator><comments>https://news.ycombinator.com/item?id=47204600</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47204600</guid></item><item><title><![CDATA[Exploring pq, a plain-text database connector for plan9]]></title><description><![CDATA[
<p>Article URL: <a href="https://mccd.space/posts/pq/">https://mccd.space/posts/pq/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46575717">https://news.ycombinator.com/item?id=46575717</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Sun, 11 Jan 2026 13:45:29 +0000</pubDate><link>https://mccd.space/posts/pq/</link><dc:creator>4diii</dc:creator><comments>https://news.ycombinator.com/item?id=46575717</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46575717</guid></item><item><title><![CDATA[New comment by 4diii in "Pixel 10 Phones"]]></title><description><![CDATA[
<p>Definitely agree about those nice touches. Just a few thoughts from my experience as of late.<p>- Window management wise the new tiling controls/keyboard shortcuts added last year have replaced third party tools for me. It's not as customisable as rectangle or moom but it covers halves and quarters which is mostly all I need. Additionally the "arrange" feature is nice letting you automatically tile N most recently used windows in a given layout.<p>- For window rules you can right click on an app in the dock and under options assign it to the current workspace and it should reopen there.<p>- 3rd party software like BetterTouchTool can "throw" windows to other spaces. But it did feel a bit hacky from memory (small but perceptible lag)<p>- The new spotlight features coming this year look quite promising for keyboard driven workflows.</p>
]]></description><pubDate>Thu, 21 Aug 2025 05:25:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=44969309</link><dc:creator>4diii</dc:creator><comments>https://news.ycombinator.com/item?id=44969309</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44969309</guid></item></channel></rss>