<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: BenoitEssiambre</title><link>https://news.ycombinator.com/user?id=BenoitEssiambre</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Sun, 24 May 2026 22:34:23 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=BenoitEssiambre" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[The Era of Compounding Language]]></title><description><![CDATA[
<p>Article URL: <a href="https://benoitessiambre.com/compound.html">https://benoitessiambre.com/compound.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47433828">https://news.ycombinator.com/item?id=47433828</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Thu, 19 Mar 2026 01:54:38 +0000</pubDate><link>https://benoitessiambre.com/compound.html</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=47433828</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47433828</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "Nobody gets promoted for simplicity"]]></title><description><![CDATA[
<p>If you're at a place that rewards complexity, you might be able to pitch simplicity as something that to get right, requires deep understanding of complex information theoretic concepts and deep understanding of ontological knowledge about the domain (it helps that this is true <a href="https://benoitessiambre.com/simple.html" rel="nofollow">https://benoitessiambre.com/simple.html</a> ). You can talk about the Bayesian Occam's Razor ( <a href="https://benoitessiambre.com/abstract.html" rel="nofollow">https://benoitessiambre.com/abstract.html</a> ) and about minimizing the entropy of code ( <a href="https://benoitessiambre.com/entropy.html" rel="nofollow">https://benoitessiambre.com/entropy.html</a> ), of tests ( <a href="https://benoitessiambre.com/integration.html" rel="nofollow">https://benoitessiambre.com/integration.html</a> ) and even of infrastructure ( <a href="https://benoitessiambre.com/pgcentrism.html" rel="nofollow">https://benoitessiambre.com/pgcentrism.html</a> ).</p>
]]></description><pubDate>Wed, 04 Mar 2026 14:49:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=47248257</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=47248257</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47248257</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "When does MCP make sense vs CLI?"]]></title><description><![CDATA[
<p>One underrated reason that CLIs are often better than MCP is that Unix tools seem to have close to information theoretically optimal layout to enabled reasoning. They are concise, in the Solomonoff/Kolmogorov sense.<p>This means that the related parts in the inputs and outputs are recursively as close together as possible.<p>There's a reason humans don't type and read http/json on the command line. It's hard to read and reason over that type of syntax. json is made to be easy to parse for simple algorithms, not meant to organise info in an easy to reason about layout.<p>AIs benefit from the easy to reason about layout. It's not just about being able to fit in the context window but about the contents of that context window being organized such that the attention mechanism doesn't have to stretch itself out trying to connect far apart related pieces. It doesn't have to try to match brackets to disambiguate information. CLIs tend to use formats that are obvious at a glance for humans and LLMs alike.<p>It's about the entropy of the formats. <a href="https://benoitessiambre.com/entropy.html" rel="nofollow">https://benoitessiambre.com/entropy.html</a></p>
]]></description><pubDate>Mon, 02 Mar 2026 01:18:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=47212710</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=47212710</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47212710</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "Pretty soon, heat pumps will be able to store and distribute heat as needed"]]></title><description><![CDATA[
<p>Yeah I always wondered if I ever switched to solar panels, would there be a way to accumulate heat to be used in the Canadian cold months that have little sunlight? The closest I found was electric thermal storage based on heating bricks. They can accumulate more energy than water since they can go to higher temperatures. For example these say they go to 1300°F or 700°C <a href="https://steffes.com/ets/roomheater/" rel="nofollow">https://steffes.com/ets/roomheater/</a> . They don't seem to have large models that could heat a house for months however.</p>
]]></description><pubDate>Tue, 03 Feb 2026 04:08:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=46866362</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46866362</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46866362</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "How does misalignment scale with model intelligence and task complexity?"]]></title><description><![CDATA[
<p>This matches my intuition. Systematic misalignment seems like it could be prevented by somewhat simple rules like the hippocratic oath or Asimov's Laws of robotics or rather probabilistic bayesian versions of these rules that take into account error bounds and risk.<p>The probabilistic version of "Do No Harm" is "Do not take excessive risk of harm".<p>This should work as AIs become smarter because intelligence implies becoming better bayesians which implies being great at calibrating confidence intervals of their interpretations and their reasoning and basically gaining a superhuman ability for evaluating the bounds of ambiguity and risk.<p>Now this doesn't mean that AIs won't be misaligned, only that it should be possible to align them. Not every AI maker will necessarily bother to align them properly, especially in adversarial, military applications.</p>
]]></description><pubDate>Tue, 03 Feb 2026 03:36:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=46866143</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46866143</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46866143</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "AGENTS.md outperforms skills in our agent evals"]]></title><description><![CDATA[
<p>Wouldn't this have been more readable with a \n newline instead of a pipe operator as a seperator? This wouldn't have made the prompt longer.</p>
]]></description><pubDate>Thu, 29 Jan 2026 22:55:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=46818024</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46818024</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46818024</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "France Aiming to Replace Zoom, Google Meet, Microsoft Teams, etc."]]></title><description><![CDATA[
<p>Countries are waking up to the danger of having the US in a position to take control of most of their computers and phones via software updates.<p>Open source solutions like  <a href="https://en.wikipedia.org/wiki/GendBuntu" rel="nofollow">https://en.wikipedia.org/wiki/GendBuntu</a> could become more prominent. There's even interesting non us hardware options like <a href="https://starlabs.systems/" rel="nofollow">https://starlabs.systems/</a><p>The US has had an unfair advantage in tech, defense, science and finance because it hosted the global hubs of the free world. This attracted eye-watering amounts of money to places like SF and NY. With the newfound isolationism, tariffs, threats etc. reducing the viability of hosting the global hubs, there's massive opportunities opening in europe and elsewhere, especially if governments can help bootstrap these sectors with efforts like these.</p>
]]></description><pubDate>Mon, 26 Jan 2026 20:12:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=46770858</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46770858</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46770858</guid></item><item><title><![CDATA[The Unreasonable Effectiveness of Failure]]></title><description><![CDATA[
<p>Article URL: <a href="https://benoitessiambre.com/fail.html">https://benoitessiambre.com/fail.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46747596">https://news.ycombinator.com/item?id=46747596</a></p>
<p>Points: 4</p>
<p># Comments: 0</p>
]]></description><pubDate>Sat, 24 Jan 2026 20:58:10 +0000</pubDate><link>https://benoitessiambre.com/fail.html</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46747596</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46747596</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "Proof of Corn"]]></title><description><![CDATA[
<p>It could be just an owner or a board or directors at the top. It's possible the CEO will be automated for some companies.</p>
]]></description><pubDate>Fri, 23 Jan 2026 20:33:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=46737504</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46737504</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46737504</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "European Alternatives"]]></title><description><![CDATA[
<p>I also mean Hacker News</p>
]]></description><pubDate>Fri, 23 Jan 2026 18:47:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=46736163</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46736163</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46736163</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "Proof of Corn"]]></title><description><![CDATA[
<p>AI CEOs are coming.</p>
]]></description><pubDate>Fri, 23 Jan 2026 18:33:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=46735967</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46735967</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46735967</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "European Alternatives"]]></title><description><![CDATA[
<p>Its founder lives in europe so there's that.</p>
]]></description><pubDate>Fri, 23 Jan 2026 18:23:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=46735834</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46735834</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46735834</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "European Alternatives"]]></title><description><![CDATA[
<p>It is a sad reality. The US has recently threatened to annex Denmark and Canada. Some of us are suddenly keenly aware that the US is in a position to take control of most of our computers and phones via software updates.<p>Open source is the global alternative you're looking for. There's even interesting hardware options like <a href="https://starlabs.systems/" rel="nofollow">https://starlabs.systems/</a><p>The US also has had an unfair advantage in tech/defense and finance because it hosted the global hubs of the free world. This attracted eye-watering amounts of money to places like SF and NY. With this newfound isolationism, tariffs etc. reducing the viability of hosting the global hubs, there's massive opportunities opening in europe and elsewhere.</p>
]]></description><pubDate>Fri, 23 Jan 2026 18:06:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=46735623</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46735623</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46735623</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "European Alternatives"]]></title><description><![CDATA[
<p>For consumers, these computers look interesting: <a href="https://starlabs.systems/" rel="nofollow">https://starlabs.systems/</a></p>
]]></description><pubDate>Fri, 23 Jan 2026 17:50:28 +0000</pubDate><link>https://news.ycombinator.com/item?id=46735440</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46735440</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46735440</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "A battle over Canada’s mystery brain disease"]]></title><description><![CDATA[
<p>I know doctors who've worked with Marrero and they seem to be split in their opinion. They seem to agree he tends to be "excessively thorough", frequently sending tests to labs across the world. This makes him liked by desperate patients with potentially incurable diseases who want someone to "do something".<p>They are split on whether his thoroughness is just fueling false hopes and sending patients down unnecessary rabbit holes or if he could have potentially identified a real issue.</p>
]]></description><pubDate>Sun, 11 Jan 2026 15:47:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=46576711</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46576711</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46576711</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "Jensen: 'We've done our country a great disservice' by offshoring"]]></title><description><![CDATA[
<p>Right but the majority of people holding significant amounts of capital is retirees or people saving for retirement. There is a small minority of people wealthy for other reasons. It doesn't really make sense to strongly associate these people to "capital" since they are a small minority of capital holders.</p>
]]></description><pubDate>Tue, 06 Jan 2026 14:25:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=46512635</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46512635</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46512635</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "Jensen: 'We've done our country a great disservice' by offshoring"]]></title><description><![CDATA[
<p>Do you have a link? What I've  seen in most discussions is obscuring of the fact that the majority of "capital" is directly or indirectly retirees or people saving for retirement. Those in the top 5% wealthiest often need to survive on that wealth for decades so it's not as if they have per year spending power that is that high. You too will be at your top percentile wealthiest of your life when you are nearing retirement.</p>
]]></description><pubDate>Tue, 06 Jan 2026 14:19:24 +0000</pubDate><link>https://news.ycombinator.com/item?id=46512557</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46512557</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46512557</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "Jensen: 'We've done our country a great disservice' by offshoring"]]></title><description><![CDATA[
<p>Note that the "Labor vs Capital" distinction mostly means "workers vs retirees". The reason more money goes to capital these days is not necessarily that each retiree is getting more but that in an aging population, there's more retirees so it takes more resources diverted from workers to support this larger non working population. This problem can be solved with more babies 20 years ago or more immigration of workers now to share the burden (unless AI makes everything weird).</p>
]]></description><pubDate>Mon, 05 Jan 2026 18:22:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=46502567</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46502567</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46502567</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "Trillions spent and big software projects are still failing"]]></title><description><![CDATA[
<p>If you click on the link, I mention other competing attempts and architectures, like Multics, Hurd, MacOS and even early Windows that either failed or started adopting Unix patterns.</p>
]]></description><pubDate>Wed, 26 Nov 2025 14:40:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=46057794</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46057794</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46057794</guid></item><item><title><![CDATA[New comment by BenoitEssiambre in "Trillions spent and big software projects are still failing"]]></title><description><![CDATA[
<p>Unix and Linux would be your quintessential examples.<p>Unix was an effort to take Multics, an operating system that had gotten too modular, and integrate the good parts into a more unified whole (book recommendation: <a href="https://www.amazon.com/UNIX-History-Memoir-Brian-Kernighan/dp/1695978552/" rel="nofollow">https://www.amazon.com/UNIX-History-Memoir-Brian-Kernighan/d...</a>).<p>Even though there were some benefits to the modularity of Multics (apparently you could unload and replace hardware in Multics servers without reboot, which was unheard of at the time), it was also its downfall. Multics was eventually deemed over-engineered and too difficult to work with. It couldn't evolve fast enough with the changing technological landscape. Bell Labs' conclusion after the project was shelved was that OSs were too costly and too difficult to design. They told engineers that no one should work on OSs.<p>Ken Thompson wanted a modern OS so he disregarded these instructions. He used some of the expertise he gained while working on Multics and wrote Unix for himself (in three weeks, in assembly). People started looking over Thompson's shoulder being like "Hey what OS are you using there, can I get a copy?" and the rest is history.<p>Brian Kernighan described Unix as "one of" whatever Multics was "multiple of". Linux eventually adopted a similar architecture.<p>More here: <a href="https://benoitessiambre.com/integration.html" rel="nofollow">https://benoitessiambre.com/integration.html</a></p>
]]></description><pubDate>Tue, 25 Nov 2025 17:26:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=46048172</link><dc:creator>BenoitEssiambre</dc:creator><comments>https://news.ycombinator.com/item?id=46048172</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46048172</guid></item></channel></rss>