<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: ajani</title><link>https://news.ycombinator.com/user?id=ajani</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 15 Jun 2026 14:11:52 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=ajani" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by ajani in "A GPT-4 capability forecasting challenge"]]></title><description><![CDATA[
<p>Oh wow, this suddenly became so heated! Is the gun made of # or _? Or it has to be both?</p>
]]></description><pubDate>Sun, 03 Sep 2023 06:31:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=37368184</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37368184</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37368184</guid></item><item><title><![CDATA[New comment by ajani in "A GPT-4 capability forecasting challenge"]]></title><description><![CDATA[
<p>Ok I see your point. Did you look at the output in the challenge? It's very different. I wonder why? Different seeds maybe.</p>
]]></description><pubDate>Sat, 02 Sep 2023 14:34:08 +0000</pubDate><link>https://news.ycombinator.com/item?id=37361872</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37361872</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37361872</guid></item><item><title><![CDATA[New comment by ajani in "A GPT-4 capability forecasting challenge"]]></title><description><![CDATA[
<p>Ok, but it's still acceptable by any common sense standard. Besides the challenge's output is completely off. Not just that it's using just one of the characters. It spells out something else. Which is not the case here.</p>
]]></description><pubDate>Sat, 02 Sep 2023 14:07:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=37361668</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37361668</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37361668</guid></item><item><title><![CDATA[New comment by ajani in "A GPT-4 capability forecasting challenge"]]></title><description><![CDATA[
<p>The challenge is flawed.<p>I asked one of the questions from the quiz to chatgpt which the quiz claims GPT can't solve. But it did.<p>Prompt: Write out the word "hello" as an ascii art drawing with # and _<p>Output:<p><pre><code>      _   _      _ _
     | | | |    | | |
     | |_| | ___| | | ___
     |  _  |/ _ \ | |/ _ \
     | | | |  __/ | | (_) |
     \_| |_/\___|_|_|\___/

</code></pre>
I guess chatgpt isn't raw GPT-4, or the quiz is using some older model.</p>
]]></description><pubDate>Sat, 02 Sep 2023 13:49:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=37361521</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37361521</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37361521</guid></item><item><title><![CDATA[New comment by ajani in "Ego Death"]]></title><description><![CDATA[
<p>Doesn't it flatten all behavior and not just impropriety?</p>
]]></description><pubDate>Tue, 22 Aug 2023 10:06:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=37220639</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37220639</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37220639</guid></item><item><title><![CDATA[New comment by ajani in "Anxious brains redirect emotion regulation"]]></title><description><![CDATA[
<p>Yes but resolution and coherence matters. A map that says "here be dragons" is not high resolution. A map that misidentifies or misses parts is incoherent. The brain is not physical (as explained above), and cannot be captured in a sufficiently high resolution map that only studies substance and its interactions. Since we have no access to the formwork, we will always have missing data, that hampers our ability to create such a coherent map.<p>Thank you for staying engaged, it's helping me see my position more clearly.</p>
]]></description><pubDate>Tue, 22 Aug 2023 09:28:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=37220394</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37220394</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37220394</guid></item><item><title><![CDATA[New comment by ajani in "Anxious brains redirect emotion regulation"]]></title><description><![CDATA[
<p>> While there are no specific equations that fully capture the intricacies of the brain's function..<p>That.</p>
]]></description><pubDate>Mon, 21 Aug 2023 17:11:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=37212420</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37212420</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37212420</guid></item><item><title><![CDATA[New comment by ajani in "Anxious brains redirect emotion regulation"]]></title><description><![CDATA[
<p>All climate models are based on mathematical physics models. I don't know the specifics, so I asked chatGPT and here is what it said:<p>'''<p>Climate modeling is a multifaceted field rooted in physics that relies on a complex set of equations to describe various atmospheric, oceanic, and terrestrial processes. Here's an overview of the key equations that form the foundation of climate models:<p>Navier-Stokes Equations: Governing the flow of fluids like the atmosphere and oceans, these equations capture how the velocity of a fluid changes over time.<p>Radiative Transfer Equations: Essential for understanding how sunlight and other forms of radiation interact with the atmosphere, including scattering, absorption, and emission.<p>Energy Balance Models: These equations describe the balance between incoming solar energy and outgoing heat, fundamental for capturing the planet's energy dynamics.<p>Equations of State: Linking density, pressure, and temperature, these equations are critical for understanding the behavior of the atmosphere and ocean.<p>Continuity Equations: Representing the conservation of mass in the atmosphere and oceans.<p>Moist Processes Equations: Capturing phase changes between water vapor, liquid water, and ice, along with latent heat exchange.<p>Boundary Layer Equations: Describing the complex dynamics near Earth's surface where the atmosphere interacts with the land or ocean.<p>Chemical and Aerosol Equations: Governing the reactions and interactions between different chemical species and particles, which can affect both weather and climate.<p>Sea Ice and Glacial Equations: Modeling the flow and melting of ice, essential for understanding the cryosphere.<p>These equations are solved numerically using computer algorithms, often over a grid representing the Earth's surface and atmosphere. Together, they form an interconnected system that allows scientists to simulate and analyze the climate system's behavior. This intricate mathematical framework underscores how the study of climate is fundamentally rooted in mathematical and physical principles.<p>'''<p>There are no such equations for the brain.</p>
]]></description><pubDate>Mon, 21 Aug 2023 16:02:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=37211421</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37211421</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37211421</guid></item><item><title><![CDATA[New comment by ajani in "Anxious brains redirect emotion regulation"]]></title><description><![CDATA[
<p>> It's mostly machine learned from the environment and other humans. We call this child development.<p>Yes, exactly. The software lies in humanity collectively. Not inside us, but in between us.<p>Which is why it cannot lie entirely in the brain. At least not entirely in one brain.<p>A new node born into the network, learns the network. Individual nodes perish, and new ones replace them. But the network doesn't go down. Or it hasn't yet.<p>Much of conscious experience is the network. Isolation is painful because it deregulates the connection to the network.<p>But even on the "hardware" side, there must be so many kinds of developments that must occur simultaneously. Some of it is as you say, environment based child development. Others that we may not quite be aware of, say patterns of neural firing, or growth patterns. Such developments probably get passed on in shape (DNA or X) to the next generation.</p>
]]></description><pubDate>Mon, 21 Aug 2023 15:01:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=37210470</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37210470</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37210470</guid></item><item><title><![CDATA[New comment by ajani in "Changing its name tanked X's downloads in App Store and Play Store"]]></title><description><![CDATA[
<p>Tesla to become roadX.</p>
]]></description><pubDate>Mon, 21 Aug 2023 12:35:57 +0000</pubDate><link>https://news.ycombinator.com/item?id=37208642</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37208642</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37208642</guid></item><item><title><![CDATA[New comment by ajani in "Anxious brains redirect emotion regulation"]]></title><description><![CDATA[
<p>"The brain" is made up of matter, yes.<p>"The brain" is not physical like columns or iron. Those are simple objects. The kind physics likes to deal with. Things that are easily measured to describe its properties quantitatively and the relations of these properties as a placeholder for qualitative aspects (equations). Physics can't deal with the brain. No equation can be written.<p>If a bud's seeds were to sprout in place, instead of in the ground, you would have every single ancestor plant in a very long chain. Every brain is the result of this kind of structure. A mother buds and sprouts a new human. If the umbilical cords remain attached, we have a very similar kind of long chain of human brains. Not like any other physical object.<p>Physics is inadequate at studying "the brain". So, "the brain" is not a physical object.</p>
]]></description><pubDate>Mon, 21 Aug 2023 12:01:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=37208312</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37208312</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37208312</guid></item><item><title><![CDATA[New comment by ajani in "Anxious brains redirect emotion regulation"]]></title><description><![CDATA[
<p>When a vortex forms in a pool of water, it is formed by the complex interplay of water, the pond made up of earth, rock, plants, its undulations perturbing the water, the wind etc.<p>The vortex appears and vanishes as conditions change. But is the vortex, the water? Is the water no longer water once the vortex dissipates?<p>Is the vortex the same as any physical object? Or it’s a form taken up by water?<p>What if the brain is just the form. And what if the water is itself just a river fed by ocean currents millions of miles away?<p>Standing by the river, you only experience the form, not the formwork, the substrate, or even the shaping force in time.<p>I may have made this a worse explanation… hard to tell.</p>
]]></description><pubDate>Mon, 21 Aug 2023 11:04:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=37207866</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37207866</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37207866</guid></item><item><title><![CDATA[New comment by ajani in "Ask HN: Tell us about your project that's not done yet but you want feedback on"]]></title><description><![CDATA[
<p>It's not clear what the idea is here. Which is why all the questions. I suspect, as you said, that a much more complex data structure would be required to encode all the various aspects of production into its constituent elements and the relations between the elements.<p>I would guess, that the first step would be to establish how the process of production (screenplay, scenes, camera angles, locations etc.) relates to the final product: scene frames. Each scene must have many shared elements as well. That would have to be encoded too.</p>
]]></description><pubDate>Mon, 21 Aug 2023 04:46:53 +0000</pubDate><link>https://news.ycombinator.com/item?id=37205593</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37205593</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37205593</guid></item><item><title><![CDATA[New comment by ajani in "Anxious brains redirect emotion regulation"]]></title><description><![CDATA[
<p>This is just tedious. The electro-chemical map of the brain that is built and used to study it, is so primitive.<p>Almost certainly, "the brain" tunes itself to its environment and inner states from a preformed inner structure through evolution. A lot of what is in the brain, is not actually to be found in the brain, or by studying it materially. The "formwork" that formed it, is lost in evolutionary time.<p>Unfortunately there is no way to impute the formwork from the form, like say it would be possible to (mostly accurately) impute what formwork created a square column or cast a piece of iron.<p>"The brain" is not in any way shaped like a physical object, and the formwork that formed it is infinitely complex.</p>
]]></description><pubDate>Mon, 21 Aug 2023 04:38:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=37205553</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37205553</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37205553</guid></item><item><title><![CDATA[New comment by ajani in "Changing its name tanked X's downloads in App Store and Play Store"]]></title><description><![CDATA[
<p>Surely the dip is temporary.</p>
]]></description><pubDate>Mon, 21 Aug 2023 03:55:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=37205284</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37205284</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37205284</guid></item><item><title><![CDATA[New comment by ajani in "LK-99 isn’t a superconductor"]]></title><description><![CDATA[
<p>Yes exactly. Behavior drives success. Not merit. Behavior is social. No sarcasm here, just acceptance of what is obvious.</p>
]]></description><pubDate>Sun, 20 Aug 2023 09:57:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=37197681</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37197681</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37197681</guid></item><item><title><![CDATA[New comment by ajani in "Ask HN: Tell us about your project that's not done yet but you want feedback on"]]></title><description><![CDATA[
<p>I see. So would the dialogue be part of the textual description?<p>What would be a typical unit? A keyframe? A bucket of frames?<p>Because the elements are temporal and visual, I think some form of “object in time” relation must be encoded in the description?</p>
]]></description><pubDate>Thu, 17 Aug 2023 10:31:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=37159623</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37159623</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37159623</guid></item><item><title><![CDATA[New comment by ajani in "LK-99 isn’t a superconductor"]]></title><description><![CDATA[
<p>Yes, how we behave usually trumps how true/false our ideas are.</p>
]]></description><pubDate>Thu, 17 Aug 2023 10:20:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=37159549</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37159549</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37159549</guid></item><item><title><![CDATA[New comment by ajani in "Ask HN: Tell us about your project that's not done yet but you want feedback on"]]></title><description><![CDATA[
<p>I see where you're going with this. Elemental (motif-based) editing instead of timeline and frame based editing?</p>
]]></description><pubDate>Thu, 17 Aug 2023 04:06:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=37157066</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37157066</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37157066</guid></item><item><title><![CDATA[New comment by ajani in "Ask HN: Tell us about your project that's not done yet but you want feedback on"]]></title><description><![CDATA[
<p>Have you tried using any LLM for this task? LLMs are text-in text-out systems and excel at transformations of text.</p>
]]></description><pubDate>Thu, 17 Aug 2023 03:58:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=37157014</link><dc:creator>ajani</dc:creator><comments>https://news.ycombinator.com/item?id=37157014</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=37157014</guid></item></channel></rss>