<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: exclipy</title><link>https://news.ycombinator.com/user?id=exclipy</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Sun, 19 Apr 2026 12:45:15 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=exclipy" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by exclipy in "OpenAI Symphony"]]></title><description><![CDATA[
<p>The specs are inscrutable agent slop. I want it to tell me what it <i>does</i> and instead it just lists database fields. It mentions a state machine and then proceeds to not describe the state machine.<p>If I wasn’t so attuned to agent slop, I’d be thinking I’m just too dumb to get it but no, whatever GPT model wrote this is not good enough (and the fact this was published by OpenAI suggests this is not an operator skill issue).</p>
]]></description><pubDate>Sat, 07 Mar 2026 01:01:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=47283220</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=47283220</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47283220</guid></item><item><title><![CDATA[New comment by exclipy in "Anthropic tries to hide Claude's AI actions. Devs hate it"]]></title><description><![CDATA[
<p>The root cause is that the TUI is just not the right surface an agentic coding tool. If it were a GUI with mouse support and other rich affordances, it'd be trivial to click to expand</p>
]]></description><pubDate>Tue, 17 Feb 2026 08:41:53 +0000</pubDate><link>https://news.ycombinator.com/item?id=47045109</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=47045109</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47045109</guid></item><item><title><![CDATA[New comment by exclipy in "Claude Opus 4.6 extra usage promo"]]></title><description><![CDATA[
<p>I think the first message consumes disproportionately much percentage because it's not cached and includes the system prompt, tools, etc.</p>
]]></description><pubDate>Fri, 06 Feb 2026 05:06:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=46909352</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=46909352</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46909352</guid></item><item><title><![CDATA[New comment by exclipy in "SVG Path Editor"]]></title><description><![CDATA[
<p>If anyone sees this, I tried it and unfortunately am not getting better results on the pelican-on-bicycle test. I think the vision models just aren't good enough yet (I tried Claude and Gemini)<p>I can share the code if there's interest.</p>
]]></description><pubDate>Fri, 30 Jan 2026 03:25:36 +0000</pubDate><link>https://news.ycombinator.com/item?id=46820193</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=46820193</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46820193</guid></item><item><title><![CDATA[New comment by exclipy in "SVG Path Editor"]]></title><description><![CDATA[
<p>Free idea: turn this into an MCP server. Give the agent the ability to virtually "hover" a path and see which part of the final render it corresponds to</p>
]]></description><pubDate>Wed, 28 Jan 2026 12:02:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=46794249</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=46794249</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46794249</guid></item><item><title><![CDATA[New comment by exclipy in "I ported JustHTML from Python to JavaScript with Codex CLI and GPT-5.2 in hours"]]></title><description><![CDATA[
<p>Can you port tsc to go in a few hours?</p>
]]></description><pubDate>Wed, 17 Dec 2025 12:53:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=46301416</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=46301416</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46301416</guid></item><item><title><![CDATA[New comment by exclipy in "Ask HN: How to deal with long vibe-coded PRs?"]]></title><description><![CDATA[
<p>I made a /split-commit prompt that automatically splits a megacommit into smaller commits. I've found this massively helpful for making more reviewable commits. You can either run this yourself or send this to your coworker to have them run it before asking you to re-review it.<p>Sometimes it doesn't split it among optimal boundaries, but it's usually good enough to help. There's probably room for improvement and extension (eg. re-splitting a branch containing many not-logical commits, moving changes between commits, merging commits, ...) – contributions welcome!<p>You can install it as a Claude Code plugin here: <a href="https://github.com/KevinWuWon/kww-claude-plugins" rel="nofollow">https://github.com/KevinWuWon/kww-claude-plugins</a> (or just copy out the prompt from the repo into your agent of choice)</p>
]]></description><pubDate>Tue, 04 Nov 2025 02:46:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=45806956</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=45806956</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45806956</guid></item><item><title><![CDATA[New comment by exclipy in "Magit Is Amazing"]]></title><description><![CDATA[
<p>Idea: a general tool that helps you write command lines. It reads the help or man page and generates a GUI or TUI to present all the options in a logical way. You just fill out the form and it does the thing</p>
]]></description><pubDate>Wed, 22 Oct 2025 01:45:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=45664051</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=45664051</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45664051</guid></item><item><title><![CDATA[New comment by exclipy in "DeepMind and OpenAI win gold at ICPC"]]></title><description><![CDATA[
<p>In what way did the computer compress time? It completed it in 5 hours and I'm pretty sure they didn't invent a time machine</p>
]]></description><pubDate>Thu, 18 Sep 2025 03:43:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=45285097</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=45285097</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45285097</guid></item><item><title><![CDATA[New comment by exclipy in "Jujutsu for everyone"]]></title><description><![CDATA[
<p>git-branchless implements revsets for git <a href="https://github.com/arxanas/git-branchless/wiki/Reference:-Revsets" rel="nofollow">https://github.com/arxanas/git-branchless/wiki/Reference:-Re...</a></p>
]]></description><pubDate>Mon, 01 Sep 2025 05:30:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=45089737</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=45089737</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45089737</guid></item><item><title><![CDATA[New comment by exclipy in "Cognitive load is what matters"]]></title><description><![CDATA[
<p>I actually think this is antithetical to the philosophy. Cyclometic complexity is very much <i>not</i> the same as "is this code difficult to understand".<p>Arbitrary structure rules like "do_thing_a(); do_thing_b(); do_thing_c();" also is not unless you can explain how this helps make it easier to understand compared to say, one big function with "// DO THING A" comments.</p>
]]></description><pubDate>Mon, 01 Sep 2025 00:41:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=45088387</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=45088387</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45088387</guid></item><item><title><![CDATA[New comment by exclipy in "Cognitive load is what matters"]]></title><description><![CDATA[
<p>Oh! I was surprised you didn't link or mention the book</p>
]]></description><pubDate>Sat, 30 Aug 2025 20:59:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=45077958</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=45077958</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45077958</guid></item><item><title><![CDATA[New comment by exclipy in "Cognitive load is what matters"]]></title><description><![CDATA[
<p>This was my main takeaway from A Philosophy Of Software Design by John Ousterhout. It is the best book on this subject and I recommend it to every software developer.<p>Basically, you should aim to minimise complexity in software design, but importantly, complexity is defined as "how difficult is it to make changes to it". "How difficult" is largely determined by the amount of cognitive load necessary to understand it.</p>
]]></description><pubDate>Sat, 30 Aug 2025 20:51:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=45077894</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=45077894</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45077894</guid></item><item><title><![CDATA[New comment by exclipy in "Typed languages are better suited for vibecoding"]]></title><description><![CDATA[
<p>The closest we got to vibe coding pre-LLMs was using a language with a very good strong type system in a good IDE and hitting Ctrl-Space to autocomplete your way to a working program.<p>I wonder if LLMs can use the type information more like a human with an IDE.<p>eg. It generates "(blah blah...); foo." and at that point it is constrained to only generate tokens corresponding to public members of foo's type.<p>Just like how current gen LLMs can reliably generate JSON that satisfies a schema, the next gen will be guaranteed to natively generate syntactically and type- correct code.</p>
]]></description><pubDate>Mon, 04 Aug 2025 01:54:52 +0000</pubDate><link>https://news.ycombinator.com/item?id=44781430</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=44781430</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44781430</guid></item><item><title><![CDATA[New comment by exclipy in "Cerebras Code"]]></title><description><![CDATA[
<p>Windsurf also has Cerebras/Qwen3-Coder. 1000 user messages per month for $15<p><a href="https://x.com/windsurf/status/1951340259192742063" rel="nofollow">https://x.com/windsurf/status/1951340259192742063</a></p>
]]></description><pubDate>Sat, 02 Aug 2025 06:06:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=44765201</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=44765201</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44765201</guid></item><item><title><![CDATA[New comment by exclipy in "Jujutsu for busy devs"]]></title><description><![CDATA[
<p>I tried it for a bit and went back to git-branchless (<a href="https://github.com/arxanas/git-branchless">https://github.com/arxanas/git-branchless</a>).<p>git-branchless is just a better set of tools for working on a git repo so local tools like the JetBrains git integration will work just fine.<p>It operates on a similar philosophy to Jujutsu (make it easy to manipulate the commit tree) and the authors did exchange ideas with each other.</p>
]]></description><pubDate>Tue, 22 Jul 2025 10:33:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=44645256</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=44645256</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44645256</guid></item><item><title><![CDATA[New comment by exclipy in "Jujutsu for busy devs"]]></title><description><![CDATA[
<p>But how do you split commits? Command line tools for splitting commits suck because they force you to look at one hunk at a time without getting the context from the rest of the change.</p>
]]></description><pubDate>Tue, 22 Jul 2025 10:30:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=44645236</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=44645236</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44645236</guid></item><item><title><![CDATA[New comment by exclipy in "Attended Windsurf's Build Night 18 hours before founders joined Google DeepMind"]]></title><description><![CDATA[
<p>You can select an element by clicking on it and it knows where in the React code it came from</p>
]]></description><pubDate>Tue, 15 Jul 2025 09:44:05 +0000</pubDate><link>https://news.ycombinator.com/item?id=44569526</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=44569526</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44569526</guid></item><item><title><![CDATA[New comment by exclipy in "Kiro: A new agentic IDE"]]></title><description><![CDATA[
<p>That's a compelling three file format.<p>Have you considered a fourth file for Implemented such that Spec = Implemented + Design?<p>It would serve both as a check that nothing is missing from Design, and can also be an index for where to find things in the code, what architecture / patterns exist that should be reused where possible.<p>And what about coding standards / style guide? Where does that go?</p>
]]></description><pubDate>Mon, 14 Jul 2025 21:40:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=44565672</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=44565672</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44565672</guid></item><item><title><![CDATA[New comment by exclipy in "Solar electricity every hour of every day is here and it changes everything"]]></title><description><![CDATA[
<p>The article doesn't say how they got that levellized cost figure. How much do they think 17kWh of battery costs?<p>From my research, a home battery costs, per kWh, about 8x an electric car battery and I don't know why.</p>
]]></description><pubDate>Fri, 11 Jul 2025 22:39:08 +0000</pubDate><link>https://news.ycombinator.com/item?id=44537489</link><dc:creator>exclipy</dc:creator><comments>https://news.ycombinator.com/item?id=44537489</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44537489</guid></item></channel></rss>