<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: kajman</title><link>https://news.ycombinator.com/user?id=kajman</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Sat, 02 May 2026 10:30:22 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=kajman" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by kajman in "Flock cameras keep telling police a man who doesn't have a warrant has a warrant"]]></title><description><![CDATA[
<p>Maybe the practice is what <i>you</i> would always want. Not me.</p>
]]></description><pubDate>Fri, 01 May 2026 20:39:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=47979998</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47979998</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47979998</guid></item><item><title><![CDATA[New comment by kajman in "How we run iSCSI over the internet"]]></title><description><![CDATA[
<p>That is an interesting example, thank you for the insight. It is amazing how simple logic is when you're able to front-load pre-reqs into a pattern match, but the idea of defining a function multiple times still feels "wrong" to me for now.<p>Overall, it's really cool to see Elixir applied like this. Congrats on the launch!</p>
]]></description><pubDate>Fri, 01 May 2026 20:08:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=47979604</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47979604</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47979604</guid></item><item><title><![CDATA[New comment by kajman in "Flock cameras keep telling police a man who doesn't have a warrant has a warrant"]]></title><description><![CDATA[
<p>I personally don't want people violently dragged out of their cars at gunpoint when they don't actually have warrants.</p>
]]></description><pubDate>Fri, 01 May 2026 19:33:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=47979167</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47979167</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47979167</guid></item><item><title><![CDATA[New comment by kajman in "Flock cameras keep telling police a man who doesn't have a warrant has a warrant"]]></title><description><![CDATA[
<p>I think the assumption is most criminals won't bother to bring cars from Hawaii or Alaska if they don't follow along.</p>
]]></description><pubDate>Fri, 01 May 2026 19:24:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=47979032</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47979032</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47979032</guid></item><item><title><![CDATA[New comment by kajman in "How we run iSCSI over the internet"]]></title><description><![CDATA[
<p>I've bounced off learning (hobby-tier) Elixir a few times now. For me, the gradual type system combined with the functional flow has been hardest to wrap my head around when it comes to solving problems.<p>I'm curious about how you approached handling a lower-level protocol in Elixir. When you began implementation, did you start thinking at the level of packets as structs, or maybe first as just maps that flowed through functions?</p>
]]></description><pubDate>Fri, 01 May 2026 18:45:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=47978466</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47978466</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47978466</guid></item><item><title><![CDATA[New comment by kajman in "Alignment whack-a-mole: Finetuning activates recall of copyrighted books in LLMs"]]></title><description><![CDATA[
<p>There's a lot of money that wants this future. I think it has no hope of outrunning commoditization.</p>
]]></description><pubDate>Thu, 30 Apr 2026 20:57:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=47968130</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47968130</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47968130</guid></item><item><title><![CDATA[New comment by kajman in "We need a federation of forges"]]></title><description><![CDATA[
<p>Understatement, probably. Your blog posts are so far the best introduction I've seen to ATProto. Is there any tagging I missed that collects them all in one place?</p>
]]></description><pubDate>Wed, 29 Apr 2026 18:46:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=47952625</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47952625</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47952625</guid></item><item><title><![CDATA[New comment by kajman in "We need a federation of forges"]]></title><description><![CDATA[
<p>Not in expert in either but ATProto services (what they call AppViews) are substantially different from the fediverse because they rely on a shared relay instead of explicit federation.<p>I'm conflicted about the costs of what is currently effectively global discovery, but it's not just another Mastodon.<p>E: I think its funny multiple other people said the same thing in the time it took me to write this</p>
]]></description><pubDate>Wed, 29 Apr 2026 18:33:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=47952464</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47952464</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47952464</guid></item><item><title><![CDATA[New comment by kajman in "A good AGENTS.md is a model upgrade. A bad one is worse than no docs at all"]]></title><description><![CDATA[
<p>I don't have a ton of experience with this, but every attempt I've made to quickly get an LLM to one-shot an AGENTS file has been too verbose in all the wrong areas. I'm not convinced LLMs are actually good at summarizing anything complex. Maybe some "blessed" prompts will bubble up in time that change my mind.</p>
]]></description><pubDate>Tue, 28 Apr 2026 21:25:26 +0000</pubDate><link>https://news.ycombinator.com/item?id=47941028</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47941028</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47941028</guid></item><item><title><![CDATA[New comment by kajman in "Show HN: Browser Harness – Gives LLM freedom to complete any browser task"]]></title><description><![CDATA[
<p>This is an advertisement that looks like a technical blogpost for a moment.</p>
]]></description><pubDate>Fri, 24 Apr 2026 19:12:42 +0000</pubDate><link>https://news.ycombinator.com/item?id=47894578</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47894578</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47894578</guid></item><item><title><![CDATA[New comment by kajman in "Mozilla Used Anthropic's Mythos to Find and Fix 271 Bugs in Firefox"]]></title><description><![CDATA[
<p>So where are they, then? Am I misunderstanding the process and this stuff is kept under wraps even after release?<p>There's three CVEs in today's security advisory that mention Anthropic.<p><a href="https://www.mozilla.org/en-US/security/advisories/mfsa2026-30/" rel="nofollow">https://www.mozilla.org/en-US/security/advisories/mfsa2026-3...</a><p>There's also no write-up I can see that distinguishes to what extent this is the work of the seven people credited alongside Mythos.</p>
]]></description><pubDate>Wed, 22 Apr 2026 02:22:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=47858018</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47858018</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47858018</guid></item><item><title><![CDATA[New comment by kajman in "Anthropic's Claude Mythos isn't a sentient super-hacker, it's a sales pitch"]]></title><description><![CDATA[
<p>> Nearly all big tech code is AI written today. What about that isn't "production-grade?"<p>I do not think the recent reliability and quality of major services would be considered acceptable if it weren't for a domestic tech oligopoly systematically lowering standards.<p>The U.S. auto industry also shows that oligopolies can shut out competition for quite some time and steer their captive market into accepting mediocrity, but innovation continues anyway. It's quite the contrast seeing domestic automakers turn their backs on EVs at the same time the Chinese are advertising 10 minute flash charging.</p>
]]></description><pubDate>Fri, 10 Apr 2026 18:57:28 +0000</pubDate><link>https://news.ycombinator.com/item?id=47722180</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47722180</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47722180</guid></item><item><title><![CDATA[New comment by kajman in "WireGuard VPN developer's Microsoft account locked"]]></title><description><![CDATA[
<p>Yesterday I couldn't get a download link for the Windows install media creation tool for a solid hour due to errors at their CDN(?). I have a friend in IT who tells me "degraded availability" across their many services is a daily thing.<p>The solution, clearly, is even more layoffs.</p>
]]></description><pubDate>Thu, 09 Apr 2026 01:25:53 +0000</pubDate><link>https://news.ycombinator.com/item?id=47698289</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47698289</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47698289</guid></item><item><title><![CDATA[New comment by kajman in "Mercor says it was hit by cyberattack tied to compromise LiteLLM"]]></title><description><![CDATA[
<p>I don't think it's crazy to imagine a misconfigured production environment. I always see these same examples of how "containers aren't really secure" and they're very amateur sins to commit though, as you mention.<p>AFAIK a comprehensive SELinux policy (like Red Hat ships) set to enforce will also prevent quite a few file accesses or modifications from escapes.</p>
]]></description><pubDate>Thu, 02 Apr 2026 15:44:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=47616011</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47616011</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47616011</guid></item><item><title><![CDATA[New comment by kajman in "Swappa.com for GrapheneOS compatible devices – Stay Away"]]></title><description><![CDATA[
<p>You've got a lot of people contesting the "often" part but I'll also add that there's a project tracking banking apps that do work: <a href="https://privsec.dev/posts/android/banking-applications-compatibility-with-grapheneos/" rel="nofollow">https://privsec.dev/posts/android/banking-applications-compa...</a><p>My tiny local credit union app isn't on there, but it worked fine. I miss wallet a bit and it's a shame that there are important apps which still refuse to act reasonably, but I don't think it's really that bad.</p>
]]></description><pubDate>Thu, 02 Apr 2026 03:07:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=47609539</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47609539</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47609539</guid></item><item><title><![CDATA[New comment by kajman in "Swappa.com for GrapheneOS compatible devices – Stay Away"]]></title><description><![CDATA[
<p>I think eBay's larger volume makes it easier to find a seller that will go through the "infeasible" trouble of sorting OEM-unlocked devices. I very recently switched to Graphene and nearly purchased from a listing that had similar "may or may not be OEM locked" language before finding one that promised it was unlocked.</p>
]]></description><pubDate>Thu, 02 Apr 2026 02:59:08 +0000</pubDate><link>https://news.ycombinator.com/item?id=47609486</link><dc:creator>kajman</dc:creator><comments>https://news.ycombinator.com/item?id=47609486</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47609486</guid></item></channel></rss>