<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: dave1010uk</title><link>https://news.ycombinator.com/user?id=dave1010uk</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 10 Sep 2026 06:05:46 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=dave1010uk" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by dave1010uk in "Emergency alerts sent to millions across England over threat of wildfires"]]></title><description><![CDATA[
<p>Here's the alert: <a href="https://www.gov.uk/alerts/14-aug-2026-2" rel="nofollow">https://www.gov.uk/alerts/14-aug-2026-2</a><p>What's interesting is that it doesn't seem to meet the normal requirements of an urgent action.<p>Previous alerts, for context: <a href="https://www.gov.uk/alerts/past-alerts" rel="nofollow">https://www.gov.uk/alerts/past-alerts</a><p>Does anyone know where this came from and what the procedure is for sending?  Presumably Cabinet Office / COBR, or NFCC.</p>
]]></description><pubDate>Fri, 14 Aug 2026 18:27:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=49302728</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=49302728</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49302728</guid></item><item><title><![CDATA[New comment by dave1010uk in "AMD acquires Taalas to boost inference performance by etching models in silicon"]]></title><description><![CDATA[
<p>I hope this doesn't come to BCI.</p>
]]></description><pubDate>Fri, 07 Aug 2026 10:53:19 +0000</pubDate><link>https://news.ycombinator.com/item?id=49208467</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=49208467</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49208467</guid></item><item><title><![CDATA[New comment by dave1010uk in "AMD acquires Taalas to boost inference performance by etching models in silicon"]]></title><description><![CDATA[
<p>I'm surprised there's not more discussion about potential inflection points here. When technology gets faster, it opens up whole new classes of UX that were hard to predict<p>For example, faster internet didn't mean being able to view 100x as many HTML4 web pages. It brought SaaS, streaming media and interactivity.<p>I'm not good at predicting, but some ideas:<p>1. All information gets augmented in real time with personalised context.<p>2. AI interaction seems more like find-as-you-type than a back and forth.<p>3. AI produces many outputs to pick from. Either the human, AI, or another system can do the deciding.<p>Even if it's last year's model, speeding up LLM inference could open up all sorts of opportunities.</p>
]]></description><pubDate>Fri, 07 Aug 2026 08:04:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=49207278</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=49207278</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49207278</guid></item><item><title><![CDATA[New comment by dave1010uk in "Cowboys, Frontiersmen, Settlers, Townspeople, Cityfolk"]]></title><description><![CDATA[
<p>Wardley's original post on this [0] is worth a read. OP adds cowboys and city folk, and makes it a bit more personal.<p>5 interesting things that are worth making explicit:<p>1. These teams/roles are as much about appetite and attitude, as they are about someone's skills and capabilities. Some people just aren't comfortable or happy when operating in these environments.<p>2. Innovation isn't just at the early "0 to 1" stages. Townspeople need to innovate to scale from a million users to a billion.<p>3. Successful orgs have teams at all stages, working together. Each team evolves what the previous stage built. And cowboys / pioneers are only successful if they can build on industrialised components.<p>5. As a product evolves, it moves through the stages. Teams either need to let go, or change their way of working.<p>[0] <a href="https://blog.gardeviance.org/2015/03/on-pioneers-settlers-town-planners-and.html" rel="nofollow">https://blog.gardeviance.org/2015/03/on-pioneers-settlers-to...</a></p>
]]></description><pubDate>Fri, 03 Jul 2026 07:23:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=48771949</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=48771949</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48771949</guid></item><item><title><![CDATA[New comment by dave1010uk in "A Eureka machine that thinks like nature and explores what AI cannot"]]></title><description><![CDATA[
<p>> Explores what AI cannot<p>In other words, gradient descent isn't good at combinatorial optimisation. I'm sure the research is better but the hype in the blog post leaves a bad taste.<p>There must be a version of Rich Sutton’s Bitter Lesson that applies to alternative computing like this, along with all the other exciting specialised hardware we've seen come and go over the years, like expert systems, optical computing, neuromorphic computing, etc.<p>Something like:<p><pre><code>    General purpose commodity silicon with rapidly evolving software generally beats specialised hardware.
</code></pre>
Software is just so much faster to iterate and improve than hardware. AI is also improving it too (eg AlphaEvolve).<p>Specialized hardware may give a single, significant improvement that grabs headlines but in the long term, compounding small improvements win.</p>
]]></description><pubDate>Thu, 28 May 2026 08:44:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=48306346</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=48306346</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48306346</guid></item><item><title><![CDATA[New comment by dave1010uk in "Greg Brockman: Inside the 72 Hours That Almost Killed OpenAI"]]></title><description><![CDATA[
<p>1. Solve reinforcement learning.<p>2. solve unsupervised learning.<p>3. gradually tackle more complicated things.<p>> what was the "real reason" they couldn't achieve their original goals?<p>I assume this is referring to why they gave up being a non-profit. The answer is that they needed more money.</p>
]]></description><pubDate>Sun, 24 May 2026 10:57:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=48256244</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=48256244</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48256244</guid></item><item><title><![CDATA[New comment by dave1010uk in "Show HN: Freenet, a peer-to-peer platform for decentralized apps"]]></title><description><![CDATA[
<p>It looks a lot like a CvRDT (i.e. a state-based CRDT).<p>They describe it as a commutative monoid, which means it has associativity and commutativity. CvRDTs also need idempotence, so they can handle duplicate data. Either they are idempotent too (which would make it semilattice-like), or the network protocol handles the deduplication outside of the data itself.<p>Letting the payload/application define the merge operation is clever. I assume it would mean contracts could opt in to idempotency if it doesn't already exist.<p>The other bit Freenet has added is doing all this with DHT routing and subscriptions, rather than a more basic peer mesh. This is very different to a blockchain and means it probably isn't suited for anything transactional.</p>
]]></description><pubDate>Thu, 21 May 2026 21:58:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=48229345</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=48229345</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48229345</guid></item><item><title><![CDATA[Biological Weapons Convention]]></title><description><![CDATA[
<p>Article URL: <a href="https://en.wikipedia.org/wiki/Biological_Weapons_Convention">https://en.wikipedia.org/wiki/Biological_Weapons_Convention</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47968070">https://news.ycombinator.com/item?id=47968070</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Thu, 30 Apr 2026 20:51:47 +0000</pubDate><link>https://en.wikipedia.org/wiki/Biological_Weapons_Convention</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=47968070</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47968070</guid></item><item><title><![CDATA[New comment by dave1010uk in "The path to ubiquitous AI (17k tokens/sec)"]]></title><description><![CDATA[
<p>You could also parse prompts into an AST, run inference, run evals, then optimise the prompts with something like a genetic algorithm.</p>
]]></description><pubDate>Fri, 20 Feb 2026 19:06:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=47092369</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=47092369</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47092369</guid></item><item><title><![CDATA[New comment by dave1010uk in "Show HN: Bible translated using LLMs from source Greek and Hebrew"]]></title><description><![CDATA[
<p>> Heck, there's others who were wealthy in scripture, even kings are they all doomed?<p>This is a great question. In the next verses, the disciples ask pretty much the same thing: "Who then can be saved?" and then Jesus explains to them:<p><pre><code>    With men this is impossible, but with God all things are possible.
</code></pre>
Whether it's a camel or a rope (and whether it's a literal needle or a small city gate, as some people argue), I think is less important (though still interesting). Either way, after the rich young ruler walks away, Jesus turns to his disciples and paints a picture something that's completely impossible without God, no matter how hard we might try by ourselves.</p>
]]></description><pubDate>Fri, 23 Jan 2026 00:30:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=46726915</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=46726915</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46726915</guid></item><item><title><![CDATA[New comment by dave1010uk in "Show HN: Bible translated using LLMs from source Greek and Hebrew"]]></title><description><![CDATA[
<p>Genesis is a theological narrative, which is very different to most things we read these days, especially as a software engineer.<p>1. The general consensus is that there were more people. This is assumed in Genesis and it (annoyingly!) doesn't bother to explain it, as the audience at the time already assumed it. Also, the authors weren't interested in all the logistics and technicalities that we are today.<p>2. Cities referenced in Genesis were likely fortified settlements, rather than like modern cities.<p>The idea that people in Africa could only build simple huts is a myth that came from the colonial era. Africa had large cities, architecture and metallurgy while parts of Europe were still tribal.<p>If you're keen to learn more, there are some good books that explain this much better than a comment can, such as "How to Read the Bible for All Its Worth" by Fee & Stuart and "Genesis for Normal People" by Pete Enns. I haven't read it but "African Civilizations" by Graham Connah is probably the go-to book on how African cities and technologies were so much further ahead than traditional European/US narratives place them.<p>The best resource for these kinds of questions is probably "The Bible Project". They have a load of YouTube videos and podcasts that cover these kinds of questions.</p>
]]></description><pubDate>Fri, 23 Jan 2026 00:10:57 +0000</pubDate><link>https://news.ycombinator.com/item?id=46726784</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=46726784</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46726784</guid></item><item><title><![CDATA[New comment by dave1010uk in "Show HN: Agent Skills Leaderboard"]]></title><description><![CDATA[
<p>The install is very opaque. It's not clear where these skills are installed, how to upgrade them or remove them.<p>Here's the `skills` package on NPM: <a href="https://www.npmjs.com/package/skills" rel="nofollow">https://www.npmjs.com/package/skills</a> - it's MIT licensed but I can't find it on Github.<p>`skills` looks to be a wrapper around `add-skill`: <a href="https://github.com/vercel-labs/add-skill" rel="nofollow">https://github.com/vercel-labs/add-skill</a><p>From the docs, `add-skill` auto detects from 16 different potential paths to copy skills to in a repository (.claude/, .codex/, .Gemini/, etc).<p>`add-skill` also let's you install skills globally (~/). From the code, `skills` looks like it doesn't support global installs but under the hood it passes all args to add-skill, so you should be able to install skills globally or install multiple skills (even if the wrapper doesn't expect it).<p>Aside: although lots of agents have adopted SKILLS.md conventions, they're currently all using their own paths. There doesn't seem to be a consensus yet, like there is with AGENTS.md. There are even 3 generic paths: .agent/skills/, .agents/skills/ and just skills/</p>
]]></description><pubDate>Wed, 21 Jan 2026 08:34:08 +0000</pubDate><link>https://news.ycombinator.com/item?id=46702711</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=46702711</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46702711</guid></item><item><title><![CDATA[New comment by dave1010uk in "Ask HN: Share your personal website"]]></title><description><![CDATA[
<p><a href="https://dave.engineer/" rel="nofollow">https://dave.engineer/</a></p>
]]></description><pubDate>Thu, 15 Jan 2026 00:08:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=46625963</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=46625963</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46625963</guid></item><item><title><![CDATA[New comment by dave1010uk in "Giving coding agents situational awareness (from shell prompts to agent prompts)"]]></title><description><![CDATA[
<p>tl;dr: like Starship's PS1 combined with Claude Code's skills.</p>
]]></description><pubDate>Tue, 13 Jan 2026 21:50:52 +0000</pubDate><link>https://news.ycombinator.com/item?id=46608741</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=46608741</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46608741</guid></item><item><title><![CDATA[Giving coding agents situational awareness (from shell prompts to agent prompts)]]></title><description><![CDATA[
<p>Article URL: <a href="https://dave.engineer/blog/2026/01/agent-situations/">https://dave.engineer/blog/2026/01/agent-situations/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46608740">https://news.ycombinator.com/item?id=46608740</a></p>
<p>Points: 2</p>
<p># Comments: 1</p>
]]></description><pubDate>Tue, 13 Jan 2026 21:50:52 +0000</pubDate><link>https://dave.engineer/blog/2026/01/agent-situations/</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=46608740</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46608740</guid></item><item><title><![CDATA[New comment by dave1010uk in "How to code Claude Code in 200 lines of code"]]></title><description><![CDATA[
<p>Impressive!<p>Here's an agent in 24 lines of PHP, written in 2023. But it relies on `llm` to do HTTP and JSON.<p><a href="https://github.com/dave1010/hubcap" rel="nofollow">https://github.com/dave1010/hubcap</a></p>
]]></description><pubDate>Fri, 09 Jan 2026 13:03:19 +0000</pubDate><link>https://news.ycombinator.com/item?id=46553407</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=46553407</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46553407</guid></item><item><title><![CDATA[New comment by dave1010uk in "'Three norths' alignment about to end"]]></title><description><![CDATA[
<p>Langton Matravers is about 2km north of the coast. The 3 norths will have met land at a place called Dancing Ledge, at about SZ 00000 76833 (50.59121, -2.00000).<p><a href="https://en.wikipedia.org/wiki/Dancing_Ledge" rel="nofollow">https://en.wikipedia.org/wiki/Dancing_Ledge</a><p>The quarry caves at Winspit are worth an explore if you're in the area - they've been used in the set for Dr Who, Blake's 7 and Andor.<p><a href="https://en.wikipedia.org/wiki/Winspit" rel="nofollow">https://en.wikipedia.org/wiki/Winspit</a></p>
]]></description><pubDate>Wed, 31 Dec 2025 12:11:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=46443482</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=46443482</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46443482</guid></item><item><title><![CDATA[New comment by dave1010uk in "UK House of Lords attempting to ban use of VPNs by anyone under 16"]]></title><description><![CDATA[
<p>I know what you're thinking: these restrictions are easy to work around. But don't worry, we can just layer more restrictions on top. Eventually the children will be safe! The government just needs to...<p>- require proof of age (ID) to install apps from unofficial sources on your phone or PC. Probably best to block this at both the OS and also popular VPN downloading sites like github.com and debian.org.<p>- require proof of age (ID) to unblock DNS provider IP addresses like 8.8.8.8 and 1.1.1.1 at your ISP.<p>- make sure children aren't using any other "privacy" tools that might be a slippery slope to installing a VPN.<p>This makes it so much easier for the parents too! The internet will be so safe that they won't even need to talk to their children about internet safety.</p>
]]></description><pubDate>Thu, 11 Dec 2025 22:37:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=46238275</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=46238275</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46238275</guid></item><item><title><![CDATA[New comment by dave1010uk in "Useful patterns for building HTML tools"]]></title><description><![CDATA[
<p>Thanks Simon!<p>My tool collection [0] is inspired by yours, with a handful of differences. I'm only at 53 tools at the moment.<p>What I did differently:<p>Hosted on Cloudflare Pages. This gives you preview URLs for pull requests out the box. This might be possible with Github Pages but I haven't checked. I've used Vercel for similar projects in the past. Cloudflare seems to have the odd failed build that needs a kick from their dashboard.<p>Some tools can make use of Workers/Functions for backend processing and secrets. I try to keep these to a minimum but they're occasionally useful.<p>I have an AGENTS.md that's updated with a Github action to automatically pull in Claude-style Skills from the .skills directory. I blogged about this pattern and am still waiting for a standard to evolve [2].<p>I have a base stylesheet that I instruct agents to pull in. This gives a bit of consistency and also let's them use Tailwind, which they'd seem to love.<p>[0] <a href="https://tools.dave.engineer/" rel="nofollow">https://tools.dave.engineer/</a><p>[1] <a href="https://github.com/dave1010/tools/tree/main/functions" rel="nofollow">https://github.com/dave1010/tools/tree/main/functions</a><p>[2] <a href="https://dave.engineer/blog/2025/11/skills-to-agents/" rel="nofollow">https://dave.engineer/blog/2025/11/skills-to-agents/</a></p>
]]></description><pubDate>Thu, 11 Dec 2025 08:49:57 +0000</pubDate><link>https://news.ycombinator.com/item?id=46229072</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=46229072</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46229072</guid></item><item><title><![CDATA[The Poison Pill in Anthropic's 'Soul Document' for Claude Opus 4.5]]></title><description><![CDATA[
<p>Article URL: <a href="https://schrodingerschatbot.substack.com/p/this-doesnt-look-like-anything-to">https://schrodingerschatbot.substack.com/p/this-doesnt-look-like-anything-to</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46154138">https://news.ycombinator.com/item?id=46154138</a></p>
<p>Points: 5</p>
<p># Comments: 2</p>
]]></description><pubDate>Thu, 04 Dec 2025 22:36:34 +0000</pubDate><link>https://schrodingerschatbot.substack.com/p/this-doesnt-look-like-anything-to</link><dc:creator>dave1010uk</dc:creator><comments>https://news.ycombinator.com/item?id=46154138</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46154138</guid></item></channel></rss>