<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: DigitalNoumena</title><link>https://news.ycombinator.com/user?id=DigitalNoumena</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Sun, 03 May 2026 08:38:39 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=DigitalNoumena" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by DigitalNoumena in "How to Build a Medieval Castle"]]></title><description><![CDATA[
<p>Coincidentally I recently watched these videos about this topic. Both great channels!<p>3D Guide - How to Build the Perfect Medieval Castle
<a href="https://youtu.be/Syjg6PHYFBo?si=JceRfeOks3hOVqWu" rel="nofollow">https://youtu.be/Syjg6PHYFBo?si=JceRfeOks3hOVqWu</a><p>How to Lay Siege to a Medieval Fortress (1000-1300)
<a href="https://www.youtube.com/watch?v=aQ7hTNoK-OA&t=1900s" rel="nofollow">https://www.youtube.com/watch?v=aQ7hTNoK-OA&t=1900s</a></p>
]]></description><pubDate>Tue, 19 Aug 2025 09:02:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=44949660</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=44949660</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44949660</guid></item><item><title><![CDATA[New comment by DigitalNoumena in "Ask HN: What are you working on? (May 2025)"]]></title><description><![CDATA[
<p>An app to estimate the risk of your job being automated and how to hedge against it professionally and financially</p>
]]></description><pubDate>Mon, 26 May 2025 08:49:19 +0000</pubDate><link>https://news.ycombinator.com/item?id=44095372</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=44095372</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44095372</guid></item><item><title><![CDATA[Dendrites can compute XOR]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.quantamagazine.org/neural-dendrites-reveal-their-computational-power-20200114/">https://www.quantamagazine.org/neural-dendrites-reveal-their-computational-power-20200114/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=42772261">https://news.ycombinator.com/item?id=42772261</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 20 Jan 2025 19:40:12 +0000</pubDate><link>https://www.quantamagazine.org/neural-dendrites-reveal-their-computational-power-20200114/</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=42772261</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42772261</guid></item><item><title><![CDATA[New comment by DigitalNoumena in "Leaving a permanent record of humanity on the moon – in 100B pixels"]]></title><description><![CDATA[
<p>You can look at one of the discs here: <a href="https://sanctuaryonthemoon.com/discs" rel="nofollow">https://sanctuaryonthemoon.com/discs</a></p>
]]></description><pubDate>Thu, 16 Jan 2025 09:37:41 +0000</pubDate><link>https://news.ycombinator.com/item?id=42723289</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=42723289</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42723289</guid></item><item><title><![CDATA[New comment by DigitalNoumena in "Don't use cosine similarity carelessly"]]></title><description><![CDATA[
<p>HyDE is the way to go! Just ask the model to generate a bunch of hypothetical answers to the question in different formats and do similarity on those.<p>Or even better, as the OP suggests standardise the format of the chunks and generate a hypothetical answer in the same format.</p>
]]></description><pubDate>Wed, 15 Jan 2025 10:48:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=42709511</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=42709511</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42709511</guid></item><item><title><![CDATA[New comment by DigitalNoumena in "Snyk security researcher deploys malicious NPM packages targeting cursor.com"]]></title><description><![CDATA[
<p>It may interest you that Guy Podjarny, one of the Snyk founders, now has an AI coding company (<a href="https://www.tessl.io/about" rel="nofollow">https://www.tessl.io/about</a>) that looks like a competitor of yours</p>
]]></description><pubDate>Tue, 14 Jan 2025 10:29:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=42695727</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=42695727</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42695727</guid></item><item><title><![CDATA[New comment by DigitalNoumena in "Wikipedia searches reveal differing styles of curiosity"]]></title><description><![CDATA[
<p>There seems to be a deliberate, implicit value judgement about "busybodies" that would explain the negative connotations:<p>> Bassett hypothesizes that “in countries that have more structures of oppression or patriarchal forces, there may be a constraining of knowledge production that pushes people more toward this hyperfocus.”</p>
]]></description><pubDate>Fri, 10 Jan 2025 10:03:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=42654278</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=42654278</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42654278</guid></item><item><title><![CDATA[New comment by DigitalNoumena in "Fast LLM Inference From Scratch (using CUDA)"]]></title><description><![CDATA[
<p>Great post! I've been looking to get into the guts of large scale model training (I'm half-way between the design and application layer of LLMs, mostly in python, sometimes a bit of c++) and this will be a great reference to have.<p>PS. appreciate it if anyone can recommend more material like this</p>
]]></description><pubDate>Mon, 16 Dec 2024 14:14:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=42431115</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=42431115</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42431115</guid></item><item><title><![CDATA[Show HN: A Raycast script for the HN bookmarklet]]></title><description><![CDATA[
<p>Since Arc doesn't have bookmarks, this became the best way for me to post.<p>```
  #!/bin/bash<p><pre><code>  # Required parameters:
  # @raycast.schemaVersion 1
  # @raycast.title Submit to Hacker News
  # @raycast.mode fullOutput

  # Optional parameters:
  # @raycast.icon ./images/yc.png
  # @raycast.packageName Web Tools
  # @raycast.argument1 { "type": "text", "placeholder": "URL to submit", "percentEncoded": false }

  # Documentation:
  # @raycast.description Submit a URL to Hacker News with the current page title.
  # @raycast.author DigitalNoumena

  # Extract the <title> tag
  page_title=$(curl -sL "$1" | sed -n 's:.*<title>\(.*\)</title>.*:\1:p')

  # If no title is found:
  if [ -z "$page_title" ]; then
    page_title="No title found"
  fi

  # URL encode the title and the URL
  encoded_url=$(echo -n "$1" | jq -sRr @uri)
  encoded_title=$(echo -n "$page_title" | jq -sRr @uri)

  # Open the Hacker News submission page with the encoded URL and title
  open "https://news.ycombinator.com/submitlink?u=$encoded_url&t=$encoded_title"
</code></pre>
```</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=41519420">https://news.ycombinator.com/item?id=41519420</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Thu, 12 Sep 2024 10:58:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=41519420</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=41519420</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41519420</guid></item><item><title><![CDATA[The Platonic Representation Hypothesis]]></title><description><![CDATA[
<p>Article URL: <a href="https://phillipi.github.io/prh/">https://phillipi.github.io/prh/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=40406419">https://news.ycombinator.com/item?id=40406419</a></p>
<p>Points: 3</p>
<p># Comments: 3</p>
]]></description><pubDate>Sun, 19 May 2024 12:16:05 +0000</pubDate><link>https://phillipi.github.io/prh/</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=40406419</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40406419</guid></item><item><title><![CDATA[Violation of Expectation via Metacognitive Prompting Reduces]]></title><description><![CDATA[
<p>Article URL: <a href="https://arxiv.org/abs/2310.06983">https://arxiv.org/abs/2310.06983</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=39124134">https://news.ycombinator.com/item?id=39124134</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Wed, 24 Jan 2024 23:35:05 +0000</pubDate><link>https://arxiv.org/abs/2310.06983</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=39124134</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39124134</guid></item><item><title><![CDATA[The first AI model based on Yann LeCun's vision for more human-like AI]]></title><description><![CDATA[
<p>Article URL: <a href="https://ai.meta.com/blog/yann-lecun-ai-model-i-jepa/">https://ai.meta.com/blog/yann-lecun-ai-model-i-jepa/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=38652708">https://news.ycombinator.com/item?id=38652708</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Fri, 15 Dec 2023 10:13:38 +0000</pubDate><link>https://ai.meta.com/blog/yann-lecun-ai-model-i-jepa/</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=38652708</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=38652708</guid></item><item><title><![CDATA[Mimicking natural selection in chemical systems]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.nature.com/articles/s41570-019-0155-6">https://www.nature.com/articles/s41570-019-0155-6</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=37319461">https://news.ycombinator.com/item?id=37319461</a></p>
<p>Points: 42</p>
<p># Comments: 3</p>
]]></description><pubDate>Wed, 30 Aug 2023 09:30:00 +0000</pubDate><link>https://www.nature.com/articles/s41570-019-0155-6</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=37319461</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37319461</guid></item><item><title><![CDATA[New comment by DigitalNoumena in "Beating GPT-4 on HumanEval with a fine-tuned CodeLlama-34B"]]></title><description><![CDATA[
<p>I think the issue of test set contamination is important, but it’s academic - when a model contains a good enough distilled representation of arguably all the code out there, does it really matter whether it can generalise OOD?<p>Realistically how many of the practical use cases where it’ll be applied will be OOD? If you can take GPT4 there then you are either a genius or working on something extremely novel so why use GPT4 in the first place?<p>I understand the goal is for LLMs to get there, but the majority of practical applications just don’t need that.</p>
]]></description><pubDate>Fri, 25 Aug 2023 22:59:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=37268084</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=37268084</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37268084</guid></item><item><title><![CDATA[New comment by DigitalNoumena in "A New Physics Theory of Life (2014)"]]></title><description><![CDATA[
<p>> <i>Its the outcome when agents don't co-operate</i><p>Cooperation is a choice, as you very well say immediately below the above statement.</p>
]]></description><pubDate>Fri, 25 Aug 2023 16:56:41 +0000</pubDate><link>https://news.ycombinator.com/item?id=37264066</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=37264066</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37264066</guid></item><item><title><![CDATA[Self Replication and Nanotechnology]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.zyvex.com/nanotech/selfRep.html">https://www.zyvex.com/nanotech/selfRep.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=37260570">https://news.ycombinator.com/item?id=37260570</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Fri, 25 Aug 2023 11:40:16 +0000</pubDate><link>https://www.zyvex.com/nanotech/selfRep.html</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=37260570</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37260570</guid></item><item><title><![CDATA[New comment by DigitalNoumena in "Logic Lab"]]></title><description><![CDATA[
<p>Haha fair enough but I think we digress. I did get the warning when I clicked.<p>The point is I see no evidence why HTTP is better for viewability than HTTPS. I showed there are cases where it’s not. You showed there are cases where it’s irrelevant. But the point remains… when is it worse and is it enough to sacrifice the security benefits?</p>
]]></description><pubDate>Fri, 25 Aug 2023 00:01:22 +0000</pubDate><link>https://news.ycombinator.com/item?id=37256109</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=37256109</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37256109</guid></item><item><title><![CDATA[New comment by DigitalNoumena in "A New Physics Theory of Life (2014)"]]></title><description><![CDATA[
<p>Ah I see, sorry to not address that - got carried away on entropic musings. It seems that is not an assumption he’s making, but it’s what he’s proven:<p>> <i>when a group of atoms is driven by an external source of energy (…) and surrounded by a heat bath (…), it will often gradually restructure itself in order to dissipate increasingly more energy.</i><p>To be honest I cannot go any deeper without reading the paper.</p>
]]></description><pubDate>Thu, 24 Aug 2023 23:50:26 +0000</pubDate><link>https://news.ycombinator.com/item?id=37256045</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=37256045</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37256045</guid></item><item><title><![CDATA[New comment by DigitalNoumena in "A New Physics Theory of Life (2014)"]]></title><description><![CDATA[
<p>Agree. I cannot comment on how probable or how widespread, but only that it is possible, and in fact it is arguably a certainty based on the premises above. Now if entropy was the only driver then self-replicating systems would be the norm. But alas, the Fermi paradox. I think we are asking too much.<p>Could it be falsified? I think if you could show that natural selection led to the survival of less dissipative life forms then you could claim falsification.</p>
]]></description><pubDate>Thu, 24 Aug 2023 23:38:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=37255969</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=37255969</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37255969</guid></item><item><title><![CDATA[New comment by DigitalNoumena in "A New Physics Theory of Life (2014)"]]></title><description><![CDATA[
<p>How I understand it:<p>Entropy (according to Boltzmann) is proportional to the number of microstates that can give rise to a particular macrostate of a system. The macrostate with the highest number of possible microstates is the uniform one, where all the accessible microstates are equally likely. So if by the 2nd law entropy must increase, the system will tend towards the uniform configuration.<p>In other words, if a force is pushing water, the configuration with the highest number of states is that where all the particles of water are also moving i.e. indistinguishable microstates, doesn't matter how you rearrange the moleculles, it will look the same as they are all moving uniformly. It would be extremely unlikely that there were a pocket of water that is mysteriously still in the middle of the stream no? Clearly there are not a lot of rearrangements of the mollecules that will keep the macrostate the same, so low entropy, which tends not to be the case.<p>(Edit) How this connects to the "desire to dissipate energy efficiently":<p>Essentially, when we talk about "energy" we really mean "free energy". This is the amount of work that we can extract from a system. This is nothing but a measure of how far a current system is to it's maximum entropy state. So dissipating energy = increasing entropy.<p>The mindblowing part for me is the connection that the process of extracting free energy is the same process that moves a system to its most likely state, uniformity, high entropy. So somehow, the ability to _accelerate_ an already inevitable process lets us reconfigure other systems _away_ from their most likely state!! So if we imagine the arrow of time to progress at the average rate of entropic decay, we are essentially reversing it for some systems by accelerating it for others!!!<p>Man...</p>
]]></description><pubDate>Thu, 24 Aug 2023 22:35:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=37255437</link><dc:creator>DigitalNoumena</dc:creator><comments>https://news.ycombinator.com/item?id=37255437</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37255437</guid></item></channel></rss>