<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: how_gauche</title><link>https://news.ycombinator.com/user?id=how_gauche</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Sat, 27 Jun 2026 06:14:10 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=how_gauche" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by how_gauche in "Anthropic says Alibaba illicitly extracted Claude AI model capabilities"]]></title><description><![CDATA[
<p>Is leveraging an enormous capital advantage to strip-mine the Internet and sell it back to us cannibalism or not? Confused on this point. I think they are exploiting a loophole in copyright law (and kind of redefining the meaning of "derivative work" in my opinion, but hey I'm not a lawyer) that collectively we tolerate because the end result is so useful</p>
]]></description><pubDate>Thu, 25 Jun 2026 13:59:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=48673542</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=48673542</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48673542</guid></item><item><title><![CDATA[New comment by how_gauche in "MiniMax M2.7 Is Now Open Source"]]></title><description><![CDATA[
<p>I love it. It's not quite as good as Sonnet but it's quick, and Minimax 2.5 is like 1/4 the cost of Haiku. With enough of a harness around it, almost any breed of monkey can be coerced into producing excellent typewriter work. GLM 5 and 5.1 are other really competitive options on the price/performance curve</p>
]]></description><pubDate>Sun, 12 Apr 2026 11:34:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=47738487</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=47738487</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47738487</guid></item><item><title><![CDATA[New comment by how_gauche in "John Carmack on mutable variables"]]></title><description><![CDATA[
<p>Side effecting computations that depend on the "real world" go into an IO monad. The game in Haskell is shifting as much of the codebase as possible into pure functions/non-side-effecting code, because it's easier to reason about and prove correct.</p>
]]></description><pubDate>Fri, 31 Oct 2025 12:58:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=45771504</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=45771504</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45771504</guid></item><item><title><![CDATA[New comment by how_gauche in "Why are we templating YAML? (2019)"]]></title><description><![CDATA[
<p>We do the same thing but commit to a second git repo that we treat like the "k8s yaml release database".</p>
]]></description><pubDate>Tue, 23 Jan 2024 16:27:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=39105523</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=39105523</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=39105523</guid></item><item><title><![CDATA[New comment by how_gauche in "House Judiciary panel plans contempt proceedings for Mark Zuckerberg"]]></title><description><![CDATA[
<p>I have an extremely similar profile (chess, PBS space time!) and I get non stop Jordan Peterson and Andrew Tate in my recommendations. YouTube Shorts are the worst for this, I can't scroll through ten videos without seeing this guy, no matter how many times I click "do not want!"</p>
]]></description><pubDate>Thu, 27 Jul 2023 18:16:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=36897924</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=36897924</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=36897924</guid></item><item><title><![CDATA[New comment by how_gauche in "Hetzner launches three new dedicated servers"]]></title><description><![CDATA[
<p>This was one of my Q4 projects at work last year. We moved CI to 3x hetzner machines, each running four copies of the self-hosted github runner, and drove our build/test times from >20min down to 3-4 min on average. It's ridiculous how big a difference running on a capable bare metal box makes. We run a thousand or so builds daily and pay about 300 euro a month for the setup; our overage fees from github actions were often higher than that. Reliability has been "ok": one of the machines started throwing errors that smell like bad RAM/CPU (bus errors, random reboots, etc), we raised a support ticket, they nuked it and gave us a fresh one.<p>We provision them with ~200 lines of shell script, which we get away with because they are not running a "prod" workload. Don't forget to run "docker system prune" on a timer! Overall these machines have been mostly unobtrusive and reliable, and the engineers greatly appreciate the order of magnitude reduction in github actions time. I've also noticed that they are writing more automation tooling now since budget anxiety is no longer a factor and the infrastructure is so much faster.</p>
]]></description><pubDate>Wed, 15 Mar 2023 13:38:05 +0000</pubDate><link>https://news.ycombinator.com/item?id=35168135</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=35168135</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=35168135</guid></item><item><title><![CDATA[New comment by how_gauche in "Experiment: The costs of slow build times"]]></title><description><![CDATA[
<p>Developer velocity has been my project for the last quarter at work: we just switched from GitHub runners to Hetzner machines, and saw similar improvements to average build times. Between self hosting our runners and switching to bazel which caches aggressively, we've driven down average CI test time from 23 min to 2m40.</p>
]]></description><pubDate>Fri, 09 Dec 2022 12:18:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=33920444</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=33920444</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=33920444</guid></item><item><title><![CDATA[New comment by how_gauche in "Systemd 252"]]></title><description><![CDATA[
<p>Pipewire broke me lots of times during its development and rollout, and that's okay, because new software has bugs and in some of my other use cases "hey it's a lot better than Jack". What I'm not going to do is carry a grudge against the maintainer for decades?<p>I've been using Linux since Slackware installed from a pile of floppy disks. I'm extremely happy with systemd (desktop and server) and view it as one of the most important UX advancements ever made in the Linux sysadmin space. It replaced a smorgasbord of broken nonsense with a unified and thoughtful system that is objectively superior to what it replaced. I suspect a lot of the extreme reaction to systemd is not just resistance to change, but based on an emotional attachment to the rag-tag heterogeneity it made obsolete.</p>
]]></description><pubDate>Tue, 01 Nov 2022 13:48:52 +0000</pubDate><link>https://news.ycombinator.com/item?id=33421197</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=33421197</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=33421197</guid></item><item><title><![CDATA[New comment by how_gauche in "Joining the Church of Emacs"]]></title><description><![CDATA[
<p>On GitHub you can add ?w=1 to diff URLs to get the same thing (there's a button for it too?)</p>
]]></description><pubDate>Mon, 03 Oct 2022 12:20:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=33066450</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=33066450</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=33066450</guid></item><item><title><![CDATA[New comment by how_gauche in "macOS Setup after 15 Years of Linux"]]></title><description><![CDATA[
<p>My partner also works in tech. Between the two of us, and between home and work, we've bought or used probably 25-30 of these over the years. I gave her an M1 air this holiday season! We've had pros, airs, and iMacs in four or five different colors occupying our living spaces for fifteen or twenty years now. We've used them for work and for play, for software development and video editing and graphic design and media consumption.<p>My verdict is: the hardware is usually GREAT but MacOS isn't. It bothered me when I quit the platform ages ago, and it is still bothering me now: little I care about has changed or improved. New annoyances though!<p>In aggregate I'd say I've spent ca. 15 hours staring at the beachball. Intrusive admin software has been a problem in the past for sure -- especially at Google -- but I've bought dozens of these for home use as well. Either 90% of their units are defective or we're using our computers very differently.<p>Terrible keyboards are another thing entirely, Lenovo went through a bout of that with the X1. My gripes about Macs are almost entirely about OSX performance/reliability and command-line habitability: version over version, it almost never improves. <i>For my workflows</i></p>
]]></description><pubDate>Fri, 31 Dec 2021 16:52:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=29749795</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=29749795</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=29749795</guid></item><item><title><![CDATA[New comment by how_gauche in "macOS Setup after 15 Years of Linux"]]></title><description><![CDATA[
<p>In my case: your employer only gives out MacBooks and expects you to use them.</p>
]]></description><pubDate>Fri, 31 Dec 2021 13:50:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=29748013</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=29748013</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=29748013</guid></item><item><title><![CDATA[New comment by how_gauche in "macOS Setup after 15 Years of Linux"]]></title><description><![CDATA[
<p>I hate my work-provided MacBook SO. MUCH. I begged them to give me a ThinkPad with Linux instead, and it's been a litany of problems in the three months I've been using it. I did all my dev work on MacBooks between 2007-2013, but my experience this year has completely vindicated my decision to ditch the Mac at home and at work back then.<p>I see people in this thread complaining that Linux distros are unreliable, but my Linux machines don't do things like beachball for 20s at a time or have complete audio system crashes requiring reboot twice a week. (Always right before a videoconference! I missed fifteen minutes of a call last week scrambling to fetch another device because the audio subsystem died and the Mac decided it <i>had</i> to install updates for a half hour when it rebooted.)<p>It's not uncommon to see uptimes on my ThinkPads measured in months. The Mac? At least it boots fast. I have to restart it once or twice a week.<p>Reliability aside -- and I realize that everyone values ergonomic factors differently -- Linux is just a better choice in every regard for me and the kind of work that I do. Interactive performance is much better, the docker-based workflows everyone's using are so much faster when they're native, and there's no mismatch between Darwin's BSD userland and the Ubuntu/CentOS you're probably using in prod.</p>
]]></description><pubDate>Fri, 31 Dec 2021 13:36:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=29747882</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=29747882</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=29747882</guid></item><item><title><![CDATA[New comment by how_gauche in "PostgREST: REST API for any Postgres database"]]></title><description><![CDATA[
<p>You can also use NOTIFY in your postgrest stored procedure to wake up a hypothetical backend processor with low latency. I really love this project! I recommend it to people all the time.</p>
]]></description><pubDate>Sat, 21 Nov 2020 22:13:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=25173139</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=25173139</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=25173139</guid></item><item><title><![CDATA[New comment by how_gauche in "How to properly manage SSH keys for server access"]]></title><description><![CDATA[
<p>It's even better if you force a hardware two-factor authentication before you grant the ssh cert.</p>
]]></description><pubDate>Sat, 26 Sep 2020 22:21:45 +0000</pubDate><link>https://news.ycombinator.com/item?id=24602639</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=24602639</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=24602639</guid></item><item><title><![CDATA[New comment by how_gauche in "Creating a Home IPv6 Network"]]></title><description><![CDATA[
<p>It's mainly so I can "road warrior" to my internal resources from my laptop transparently. IPv6 is a good choice for this since it won't conflict with any NAT address space you're likely to be on.</p>
]]></description><pubDate>Sat, 26 Sep 2020 17:24:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=24600492</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=24600492</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=24600492</guid></item><item><title><![CDATA[New comment by how_gauche in "Toward a “modern” Emacs"]]></title><description><![CDATA[
<p>Many of us switched to evil mode because we got RSI from incorrectly chording all the time! C-a & C-e I'm looking at you</p>
]]></description><pubDate>Sat, 26 Sep 2020 13:39:57 +0000</pubDate><link>https://news.ycombinator.com/item?id=24598764</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=24598764</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=24598764</guid></item><item><title><![CDATA[New comment by how_gauche in "Toward a “modern” Emacs"]]></title><description><![CDATA[
<p>Huge skew depending on what platforms you like to work on. Java or Windows people are significantly less likely to use trad Unix tools. Backend or Linux people, it's 30/30/* in my experience between vim/Emacs/everything else -- this was certainly the case at Google when I was there.</p>
]]></description><pubDate>Sat, 26 Sep 2020 13:33:19 +0000</pubDate><link>https://news.ycombinator.com/item?id=24598697</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=24598697</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=24598697</guid></item><item><title><![CDATA[New comment by how_gauche in "Toward a “modern” Emacs"]]></title><description><![CDATA[
<p>I don't know dude, I've had new grads recently wholesale switch to Emacs just after watching me work for five minutes. "Holy crap, why are you editing these programs so much faster than me?"<p>Completely agree that out-of-the-box Emacs experience is NOT where it's at. I have twenty years invested in tweaking it to minimize keystrokes and operate faster for my use cases. Just remember that Emacs is not actually an editor, it's a Lisp Machine OS you can use to build an editor out of parts<p>Also keep in mind that alternate distros like Spacemacs are the way Emacs <i>has</i> to evolve, because making drastic structural changes to defaults would break so many user scripts and packages -- the things that provide most of the long-term value proposition of using Emacs in the first place.</p>
]]></description><pubDate>Sat, 26 Sep 2020 13:22:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=24598622</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=24598622</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=24598622</guid></item><item><title><![CDATA[New comment by how_gauche in "Creating a Home IPv6 Network"]]></title><description><![CDATA[
<p>You're supposed to use DHCPv6 or Neighbor Discovery -- like everything else in IPv6-land, it's significantly more complicated than it is over IPv4.<p>I don't run the whole network IPv6 -- for hosts I care about having an IPv6 egress for, I use a Wireguard tunnel in IPv6 private address space to a bastion host. If I want to expose a port, I forward it from the other side. It's a sad state of affairs :-(</p>
]]></description><pubDate>Mon, 21 Sep 2020 13:53:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=24543451</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=24543451</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=24543451</guid></item><item><title><![CDATA[New comment by how_gauche in "PostgreSQL beginner guide"]]></title><description><![CDATA[
<p>I've done this before (postgREST) for internal apps and it's <i>amazing</i>. You can POST a JSON file to the front end from a shell script with curl and have it show up on a Grafana dashboard immediately. I've used this in the past for setting up automated quality tests for machine learning pipelines -- the pipeline runs automatically and you get the results in a central place, so you can see how your quality metrics are trending over time.<p>Given that postgREST exists, I don't see any reason to talk to postgres directly from a JS app.</p>
]]></description><pubDate>Mon, 03 Aug 2020 13:17:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=24037377</link><dc:creator>how_gauche</dc:creator><comments>https://news.ycombinator.com/item?id=24037377</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=24037377</guid></item></channel></rss>