<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: mijoharas</title><link>https://news.ycombinator.com/user?id=mijoharas</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Fri, 20 Nov 2026 09:38:58 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=mijoharas" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by mijoharas in "Synaptics patent claims stop work on DisplayLink drivers"]]></title><description><![CDATA[
<p>My company used to have some display link docks, which I was unable to use due to bugs in the Linux implementations.<p>For a while I'd keep an eye on updates to see if they'd been fixed but they never were before we moved office.<p>I'm going to be quite sure we don't get any more in any future offices (we don't have them at our current office)</p>
]]></description><pubDate>Tue, 10 Nov 2026 19:03:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=50444676</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=50444676</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50444676</guid></item><item><title><![CDATA[New comment by mijoharas in "Libghostty 1.0"]]></title><description><![CDATA[
<p>I had a look to see how Rex was coming along the other day and it seems there's still no Linux beta (though MacOs users have one).<p>I'm still interested to check it out though none of these things have ever interested me too much so far.</p>
]]></description><pubDate>Sat, 07 Nov 2026 10:39:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=50403331</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=50403331</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50403331</guid></item><item><title><![CDATA[New comment by mijoharas in "OpenSSH 10.6p2 has been released with a workaround for a critical vulnerability"]]></title><description><![CDATA[
<p>There is this write-up/fancy page from the person who found the vulnerability[0] (HN discussion[1], how I found this thread).<p>At the bottom of that it says dropbear is unaffected.<p>> on debian, ubuntu, fedora or anything else with glibc? going through pam? dropbear? openbsd? you're fine<p>[0] <a href="https://sshhole.com/" rel="nofollow">https://sshhole.com/</a><p>[1] <a href="https://news.ycombinator.com/item?id=50378228">https://news.ycombinator.com/item?id=50378228</a></p>
]]></description><pubDate>Thu, 05 Nov 2026 12:41:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=50378414</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=50378414</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50378414</guid></item><item><title><![CDATA[New comment by mijoharas in "Kimi K3.1 scores 55 on Artificial Analysis, placed on the Pareto line"]]></title><description><![CDATA[
<p>Maybe I need to try it again, but I have never liked the opus models.<p>I keep having problems with them not doing the thing I ask, (and also failing to call tools correctly in pi. I think they've very much trained to the harness, which is frustrating.).</p>
]]></description><pubDate>Fri, 23 Oct 2026 20:50:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=50215141</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=50215141</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50215141</guid></item><item><title><![CDATA[New comment by mijoharas in "US threatens Europe over Starlink restrictions"]]></title><description><![CDATA[
<p>brent crude futures are trading above 109$/barrel right now. America didn't want that to happen, Iran did.
They won that one.</p>
]]></description><pubDate>Sun, 18 Oct 2026 20:10:53 +0000</pubDate><link>https://news.ycombinator.com/item?id=50144973</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=50144973</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50144973</guid></item><item><title><![CDATA[New comment by mijoharas in "New Withings Users Must Pay $99/Year for Apple Health Syncing"]]></title><description><![CDATA[
<p>Used to be a thing before LLMs too.<p>Here is a starting point. It includes a link to instructions on how to reverse engineer a Bluetooth scale[0], and it looks like someone dumped some data already.<p>[0] <a href="https://github.com/oliexdev/openScale/issues/786" rel="nofollow">https://github.com/oliexdev/openScale/issues/786</a></p>
]]></description><pubDate>Thu, 15 Oct 2026 20:54:27 +0000</pubDate><link>https://news.ycombinator.com/item?id=50110068</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=50110068</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50110068</guid></item><item><title><![CDATA[New comment by mijoharas in "Redesigning source control at Anthropic"]]></title><description><![CDATA[
<p>So... Anthropic reinvents SVN?</p>
]]></description><pubDate>Tue, 13 Oct 2026 17:17:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=50075274</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=50075274</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50075274</guid></item><item><title><![CDATA[New comment by mijoharas in "Show HN: SolidCraft – Open-Source Parametric CAD in Rust (With MCP Support)"]]></title><description><![CDATA[
<p>Interested to try this out.<p>I played around with Freecad mcp, and then also with having an agent generate some openscad before.<p>For the Freecad mcp I didn't like it at all. The agent didn't work in a way that would let me easily modify things. I hadn't given it instructions, but it didn't really create sketches and then extrude them (after checking they're all fully parameterized.)<p>For openscad the problem was just that while the model could do it, the lack of "visual reasoning" (for want of a better word) really held it back. There was a nice workflow there though by giving it a little "generate_image" of part tool and it could make a nice iterative loop.<p>Overall for both approaches it was slower than if I'd done it myself (and I'm not particularly good at 3d modelling). Now, that's not that surprising given that I was trying something new, but without being able to edit things generated by the AI myself I wouldn't want to try it again.<p>Now, maybe this solution can work in the way I'd want to. I may give this a go next time I have something I want to make.</p>
]]></description><pubDate>Tue, 13 Oct 2026 09:04:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=50068758</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=50068758</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50068758</guid></item><item><title><![CDATA[New comment by mijoharas in "Strands Decider 2B: a small, open-source, decision model"]]></title><description><![CDATA[
<p>so, how are people running these locally? is there a llama.cpp/ollama solution for systemone/jev style apis?</p>
]]></description><pubDate>Wed, 07 Oct 2026 13:48:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=49992865</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=49992865</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49992865</guid></item><item><title><![CDATA[New comment by mijoharas in "Google Playground: Create and play custom games"]]></title><description><![CDATA[
<p>UK too.</p>
]]></description><pubDate>Wed, 07 Oct 2026 13:45:41 +0000</pubDate><link>https://news.ycombinator.com/item?id=49992825</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=49992825</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49992825</guid></item><item><title><![CDATA[New comment by mijoharas in "What is Codemode"]]></title><description><![CDATA[
<p>not at all, it's nice! solves the composability problem while keeping a trust separation between harness/ai bash calls.</p>
]]></description><pubDate>Wed, 07 Oct 2026 10:11:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=49990641</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=49990641</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49990641</guid></item><item><title><![CDATA[New comment by mijoharas in "What is Codemode"]]></title><description><![CDATA[
<p>I agree with most of what you said, but wanted to add that another point is there is a different level of trust in code executing from the harness, and code executing from the bash shell by the agent.<p>I block all network access from bash calls, but want to provide certain tools that can call certain services in a specific way. I can have a tool (or mcp) that does just that, and runs outside of the sandbox.</p>
]]></description><pubDate>Wed, 07 Oct 2026 10:06:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=49990595</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=49990595</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49990595</guid></item><item><title><![CDATA[New comment by mijoharas in "Strands Decider 2B: a small, open-source, decision model"]]></title><description><![CDATA[
<p>oh god, you're completely right! nice and obvious on a reread. (I also just learnt about "garden path" sentences.)<p>I'd googled "bernoulli map" after reading that, and came across this[0], which I wasn't aware of and thought was somehow related so I entrenched my misunderstanding (I didn't dig in.)<p>Side note: I just wrote the sentence "you're completely right!", and almost changed it because it sounds like slop now. I wonder if we're gonna get an increase in these sentences in human written prose over time as people mimic these sentences, or a decrease as people shy away from them to not sound like AI. :)<p>[0] <a href="https://en.wikipedia.org/wiki/Dyadic_transformation" rel="nofollow">https://en.wikipedia.org/wiki/Dyadic_transformation</a></p>
]]></description><pubDate>Wed, 07 Oct 2026 09:47:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=49990439</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=49990439</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49990439</guid></item><item><title><![CDATA[New comment by mijoharas in "Strands Decider 2B: a small, open-source, decision model"]]></title><description><![CDATA[
<p>From Bernoulli maps apparently. I still don't understand why[0].<p>[0] <a href="https://news.ycombinator.com/item?id=49723267">https://news.ycombinator.com/item?id=49723267</a> (see parent for reference)</p>
]]></description><pubDate>Wed, 07 Oct 2026 06:21:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=49989021</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=49989021</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49989021</guid></item><item><title><![CDATA[New comment by mijoharas in "Git 3.0's upcoming SHA-256 default will be a costly mistake"]]></title><description><![CDATA[
<p>> There is no "initialize a new thing on push" and there never has been<p>Well, there hasn't ever been a real reason for it, and now there is?</p>
]]></description><pubDate>Fri, 02 Oct 2026 08:59:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=49931355</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=49931355</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49931355</guid></item><item><title><![CDATA[New comment by mijoharas in "Pi 1.0"]]></title><description><![CDATA[
<p>Ooi, can I ask for the reasoning for the new default of fullscreen?<p>The codemode stuff/mcp has been explained to death at this point but I haven't seen anything around fullscreen.<p>(I don't have strong feelings either way, just curious)</p>
]]></description><pubDate>Fri, 02 Oct 2026 08:15:30 +0000</pubDate><link>https://news.ycombinator.com/item?id=49931083</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=49931083</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49931083</guid></item><item><title><![CDATA[New comment by mijoharas in "Show HN: Ledge.sh – Runnable Markdown Notes"]]></title><description><![CDATA[
<p>I was thinking the same in terms of similarity to org-babel (which can make some very nice runnable notebooks).<p>To be clear, is the issue with org-babel that it's not accessible to other people on the team? or something else that this solves?</p>
]]></description><pubDate>Thu, 01 Oct 2026 10:08:42 +0000</pubDate><link>https://news.ycombinator.com/item?id=49919822</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=49919822</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49919822</guid></item><item><title><![CDATA[New comment by mijoharas in "Pi.dev: You Said No MCP"]]></title><description><![CDATA[
<p>I find it pretty handy. I run any bash tools in a restricted sandbox, and so that it cannot hit the network (usually).
I don't want any credentials to be available to the agent, and my trust boundary for that is precisely the harness vs. the agent. Harness can touch secrets, agent can't (filesystem is restricted from it being able to read any secrets as well).<p>that's my one problem with the cli vs mcp debate. I prefer cli, because they're tools I'm familiar with, and they're composable. My problem with it is some cli tools need to use secrets to access things.
By making sure they only go to the harness, then I've got my problem solved.</p>
]]></description><pubDate>Wed, 30 Sep 2026 14:25:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=49909465</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=49909465</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49909465</guid></item><item><title><![CDATA[New comment by mijoharas in "What AI-Native Looks Like"]]></title><description><![CDATA[
<p>Getting a 502 error. I had to come to the comments to check whether that was the joke.</p>
]]></description><pubDate>Wed, 23 Sep 2026 13:43:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=49816059</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=49816059</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49816059</guid></item><item><title><![CDATA[New comment by mijoharas in "Introducing System One Models and Jev"]]></title><description><![CDATA[
<p>It is Bernoulli (mentioned in the comment I replied to). I just failed to guess that myself :)</p>
]]></description><pubDate>Wed, 16 Sep 2026 08:42:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=49723672</link><dc:creator>mijoharas</dc:creator><comments>https://news.ycombinator.com/item?id=49723672</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49723672</guid></item></channel></rss>