<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: woodglyst</title><link>https://news.ycombinator.com/user?id=woodglyst</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 14 Sep 2026 23:42:28 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=woodglyst" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[Entropy of a Large Language Model output]]></title><description><![CDATA[
<p>Article URL: <a href="https://nikkin.dev/blog/llm-entropy.html">https://nikkin.dev/blog/llm-entropy.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=42649315">https://news.ycombinator.com/item?id=42649315</a></p>
<p>Points: 161</p>
<p># Comments: 63</p>
]]></description><pubDate>Thu, 09 Jan 2025 20:00:47 +0000</pubDate><link>https://nikkin.dev/blog/llm-entropy.html</link><dc:creator>woodglyst</dc:creator><comments>https://news.ycombinator.com/item?id=42649315</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42649315</guid></item><item><title><![CDATA[New comment by woodglyst in "M4 MacBook Pro"]]></title><description><![CDATA[
<p>Thanks to your comment. I persuaded my friend who purchased an M3 Air 24GB recently and we got 200$ back (Remuneration for price drop valid for 14 days after the date of DELIVERY) where we live</p>
]]></description><pubDate>Thu, 31 Oct 2024 07:17:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=42004214</link><dc:creator>woodglyst</dc:creator><comments>https://news.ycombinator.com/item?id=42004214</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42004214</guid></item><item><title><![CDATA[New comment by woodglyst in "Show HN: I built a task manager that separates "do" and "due" dates"]]></title><description><![CDATA[
<p>I was looking for something like Noteplan as well. The subscription model and the price was a deterrent to me and I went with Agenda [0]<p>[0] <a href="https://agenda.com/" rel="nofollow">https://agenda.com/</a></p>
]]></description><pubDate>Sun, 27 Oct 2024 17:50:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=41964273</link><dc:creator>woodglyst</dc:creator><comments>https://news.ycombinator.com/item?id=41964273</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41964273</guid></item><item><title><![CDATA[New comment by woodglyst in "Data Version Control"]]></title><description><![CDATA[
<p>This sounds a lot like the experimental project Jacquard [0] from Ink & Switch.<p>[0] <a href="https://www.inkandswitch.com/jacquard/notebook/" rel="nofollow">https://www.inkandswitch.com/jacquard/notebook/</a></p>
]]></description><pubDate>Sun, 20 Oct 2024 12:20:36 +0000</pubDate><link>https://news.ycombinator.com/item?id=41894861</link><dc:creator>woodglyst</dc:creator><comments>https://news.ycombinator.com/item?id=41894861</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41894861</guid></item><item><title><![CDATA[New comment by woodglyst in "Show HN: A journaling service that runs over WhatsApp"]]></title><description><![CDATA[
<p>End to end encryption is not really a pledge. That is expected of companies like such. Nevertheless, their promise to not sell any data is interesting. If they don’t sell data (which cannot be sold anyways for an E2EE system) I wonder why they collect so much data related to one’s identity as disclosed by them in the App Store Page? Is the behaviour of journaling then becomes a data point to be sold by these companies? Makes you wonder. And as mentioned in their privacy policy page, they are also not except from disclosing information the the US Govt if mandated by a warrant.</p>
]]></description><pubDate>Tue, 24 Sep 2024 09:17:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=41634666</link><dc:creator>woodglyst</dc:creator><comments>https://news.ycombinator.com/item?id=41634666</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41634666</guid></item><item><title><![CDATA[New comment by woodglyst in "I won't be renewing my Pinboard subscription"]]></title><description><![CDATA[
<p>I got Anybox[0] with the lifetime subscription (40$) and have been happy with it (Only for Apple devices unfortunately)<p>I can choose to automatically download a web archive when I bookmark. Also has a trial version. Can be a bit overwhelming to set things up. But works seamlessly once done.<p>[0] <a href="https://anybox.app/" rel="nofollow">https://anybox.app/</a></p>
]]></description><pubDate>Fri, 13 Sep 2024 20:59:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=41535065</link><dc:creator>woodglyst</dc:creator><comments>https://news.ycombinator.com/item?id=41535065</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41535065</guid></item><item><title><![CDATA[New comment by woodglyst in "Show HN: Optigraph – optimum graph network generator"]]></title><description><![CDATA[
<p>The statement that this can be implemented with a quantum algorithm is a bit ambiguous. If you look in detail, the problem is only formulated on the quantum computer while the optimization routine which essential solves the problem is left to a classical computer. There are some notions of quantum gradients. But I wouldn’t know how it applies to such problems</p>
]]></description><pubDate>Tue, 21 May 2024 16:37:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=40430663</link><dc:creator>woodglyst</dc:creator><comments>https://news.ycombinator.com/item?id=40430663</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40430663</guid></item><item><title><![CDATA[New comment by woodglyst in "New design of the OpenAI blog page"]]></title><description><![CDATA[
<p>Still missing RSS feed</p>
]]></description><pubDate>Tue, 14 May 2024 14:55:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=40355855</link><dc:creator>woodglyst</dc:creator><comments>https://news.ycombinator.com/item?id=40355855</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40355855</guid></item><item><title><![CDATA[New comment by woodglyst in "RSS Feed Organization Strategies and New Feed Cost"]]></title><description><![CDATA[
<p>The problem with this approach is determining the what k is for the k-means. But again, we could use the “elbow” technique to determine what’s the optimal k and then start grouping them together. I wonder if there are any automatic sophisticated clustering algorithms?</p>
]]></description><pubDate>Tue, 02 Apr 2024 15:07:57 +0000</pubDate><link>https://news.ycombinator.com/item?id=39906593</link><dc:creator>woodglyst</dc:creator><comments>https://news.ycombinator.com/item?id=39906593</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39906593</guid></item></channel></rss>