<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: deviation</title><link>https://news.ycombinator.com/user?id=deviation</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Fri, 20 Nov 2026 22:10:40 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=deviation" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by deviation in "Show HN: Subvocal – Talk to your computer without making noise"]]></title><description><![CDATA[
<p>Have you considered how this might benefit those who are deaf, or speech impaired? Their lip/mouth movements might be totally different to folks with hearing - but if this can be fine-tuned on an individual's specific speech pattern, then I see a huge opportunity in the accessibility market for this too.<p>I would also attempt to patent this <i>yesterday</i> before Meta et al start bringing their own into the market. Heck - I wouldn't be surprised if Meta stepped in to aquihire all of you.</p>
]]></description><pubDate>Tue, 03 Nov 2026 09:26:19 +0000</pubDate><link>https://news.ycombinator.com/item?id=50349823</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=50349823</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50349823</guid></item><item><title><![CDATA[New comment by deviation in "Any AI chat can join a machine network anonymously – no keys, no signup"]]></title><description><![CDATA[
<p>Excellent. Just what I needed... Another AI adapter that is 100% written by AI, with about as much utility as a screen door on a submarine.</p>
]]></description><pubDate>Tue, 06 Oct 2026 13:33:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=49978207</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=49978207</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49978207</guid></item><item><title><![CDATA[New comment by deviation in "Gemini 3.8 Live and 3.8 Live Extended Thinking"]]></title><description><![CDATA[
<p>Not a great impression to have your demo video demonstrate how one of your 'most advanced' AI models loses to the most common check-mate pattern in all of chess.</p>
]]></description><pubDate>Tue, 15 Sep 2026 18:43:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=49716918</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=49716918</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49716918</guid></item><item><title><![CDATA[New comment by deviation in "4.5B Posts Scraped from TikTok"]]></title><description><![CDATA[
<p>My thoughts also</p>
]]></description><pubDate>Thu, 03 Sep 2026 12:01:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=49548868</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=49548868</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49548868</guid></item><item><title><![CDATA[New comment by deviation in "AMD acquires Taalas to boost inference performance by etching models in silicon"]]></title><description><![CDATA[
<p>This is the only demo of 2026 which has blown my mind.<p>If we can get to this speed with reasoning models, man... I can't even imagine the impact.</p>
]]></description><pubDate>Fri, 07 Aug 2026 09:03:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=49207729</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=49207729</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49207729</guid></item><item><title><![CDATA[New comment by deviation in "Ten advances in mathematics and theoretical computer science"]]></title><description><![CDATA[
<p>Genuine question, from someone with a non-math background - at what point does the constraint become the number of unsolved conjectures remaining, instead of the ability for LLMs to actually solve for one?</p>
]]></description><pubDate>Tue, 04 Aug 2026 10:43:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=49166719</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=49166719</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49166719</guid></item><item><title><![CDATA[New comment by deviation in "Show HN: tale.fyi, we deserve a home for fiction"]]></title><description><![CDATA[
<p>Fantastic. Have you considered supporting other languages, too? Such as French?</p>
]]></description><pubDate>Tue, 28 Jul 2026 13:38:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=49083703</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=49083703</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49083703</guid></item><item><title><![CDATA[New comment by deviation in "Neverclick: Desktop application for performing mouse actions with your keyboard"]]></title><description><![CDATA[
<p>My problem with these types of apps will always be that <i>most</i> software it interacts with will unfortunately be designed without accessibility in mind.<p>I've trialed a handful of these mouse-less programs before and I always run into the same hurdles.<p>E.g. where one or more of my apps have sliders which I can't interact with using clicks instead of drags (so I can't interact with it at all), or needing to click things like my MacOS  dock, or even resize a window.</p>
]]></description><pubDate>Wed, 15 Jul 2026 10:58:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=48918971</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=48918971</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48918971</guid></item><item><title><![CDATA[New comment by deviation in "A love letter to flashcards"]]></title><description><![CDATA[
<p>Sure, I can explain - it's a little complicated, but I maintain a small-ish catalogue of ~1000 positions for the following topics:<p>- Checkmates-in-one
- Checkmates-in-two
- Defensive technique (avoid checkmate/material loss)
- Winning material
- Endgame patterns
- etc (~5 more)<p>... That I got by scraping the Lichess database, favoring common patterns that appear within +- 600 of my current Elo.<p>From there, I have Claude build me a script to convert each of those positions into a .png, then create me a deck with all the cards, et voilà. The front of the card is the position, the back is the best move in that position with a small explanation.<p>Every ~2-3 months when I see that most of the cards have matured (according the the Anki spaced repetition scheduler, I build a new deck around my new Elo.<p>I also play a lot. Prior to ~1000 rating I got away with spending 90% of my time in Anki and 10% playing online games, but lately it's been pretty 50/50. In higher ratings, playing real games tends to translate into wins more effectively for me.<p>For studying openings, it's almost the same thing, but the back of each card is the book move for my opening + the name of the opening the opponent chose.</p>
]]></description><pubDate>Fri, 10 Jul 2026 18:12:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=48863283</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=48863283</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48863283</guid></item><item><title><![CDATA[New comment by deviation in "A love letter to flashcards"]]></title><description><![CDATA[
<p>I use Anki to learn French, Chess openings/tactics/techniques, to unscramble letters for scrabble, for Pub Trivia... The options are kind of limitless.<p>As a mid-30s guy who has well passed the neuroplasticity of his teen years, it's a godsend for me.<p>To echo the author's thoughts though, I can't prove empirically that I learn more effectively using Anki (or spaced repetition) than other methods... Only anecdotally. I have a shockingly poor memory, but now I'm B2 certified in French and an ~1800 Elo on chess.com .<p>Do I still forget things <i>all the time</i>? Yes.</p>
]]></description><pubDate>Fri, 10 Jul 2026 17:04:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=48862529</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=48862529</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48862529</guid></item><item><title><![CDATA[New comment by deviation in "An Introduction to YOLO26"]]></title><description><![CDATA[
<p>My buddy has some vision impairments, and I remember training a <i>much</i> older of YOLO's models to detect objects/enemies in Terraria for him. It worked very well.<p>I then tried trained it on a <i>lot</i> of sample images from a 3D point & shoot game, and was quite disappointed in how it performed.<p>Has anyone else experimented with it recently? How does this suit as a base-model for training custom classifiers? And with hardware growth in the last ~5 years, is it suitable to run in parallel with games which are graphically intensive?</p>
]]></description><pubDate>Tue, 23 Jun 2026 11:08:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=48643198</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=48643198</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48643198</guid></item><item><title><![CDATA[New comment by deviation in "Fable Converted Pylint to Rust"]]></title><description><![CDATA[
<p>They may instead be targeting folks that run it in CI.</p>
]]></description><pubDate>Fri, 19 Jun 2026 09:33:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=48596656</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=48596656</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48596656</guid></item><item><title><![CDATA[New comment by deviation in "Fable ban was never about a jailbreak?"]]></title><description><![CDATA[
<p><a href="https://archive.is/3rY1H" rel="nofollow">https://archive.is/3rY1H</a></p>
]]></description><pubDate>Tue, 16 Jun 2026 15:51:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=48557207</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=48557207</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48557207</guid></item><item><title><![CDATA[New comment by deviation in "GitHub Is Down"]]></title><description><![CDATA[
<p>We should have a pinned post, just for GitHub outages.</p>
]]></description><pubDate>Mon, 08 Jun 2026 09:02:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=48442916</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=48442916</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48442916</guid></item><item><title><![CDATA[New comment by deviation in "Mouseless – keyboard-driven control of macOS/Linux/Windows"]]></title><description><![CDATA[
<p>How do you handle apps that don't have great accessibility support?<p>Hell - after installing, it shows a slider to help adjust the label size of each element - but I can't slide it because (surprise) Homerow doesn't support horizontal slider elements like this.<p>I also can't highlight text on the screen etc.<p>Struggling to find a usecase for Homerow that isn't just navigating chrome or my filesystem.</p>
]]></description><pubDate>Fri, 05 Jun 2026 16:20:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=48414677</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=48414677</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48414677</guid></item><item><title><![CDATA[New comment by deviation in "Uber's $1,500/month AI limit is a useful signal for AI tool pricing"]]></title><description><![CDATA[
<p>$300/day at Apple, with an increase to $500 with manager approval.</p>
]]></description><pubDate>Thu, 04 Jun 2026 13:35:30 +0000</pubDate><link>https://news.ycombinator.com/item?id=48398455</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=48398455</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48398455</guid></item><item><title><![CDATA[New comment by deviation in "Using AI to write better code more slowly"]]></title><description><![CDATA[
<p>I manage a component of an internal compute product which serves ~a billion idempotent use-cases per quarter and I can confidently tell you that you're incorrect.<p>What I haven't been able to teach AI is the full distributed nature of the system, how we progressively roll out each service (about ~30 unique ones) when we push updates -- and how to read, write, and review my code while keeping all of this in-context (because believe me, if it's not in-context, it is useless to me). Don't get me started on all the containers, K8s configs, endpoint naming conventions...<p>My entire stack covers bare metal, virtualisation infrastructure, storage infrastructure... I could go on. At a certain scale, it doesn't matter how fast you write something, but if what you're writing is bulletproof.</p>
]]></description><pubDate>Tue, 26 May 2026 09:44:05 +0000</pubDate><link>https://news.ycombinator.com/item?id=48277360</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=48277360</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48277360</guid></item><item><title><![CDATA[New comment by deviation in "Dehydration's role in learning and memory"]]></title><description><![CDATA[
<p>Jumping in here with relevant experience - because you're absolutely right.<p>In 8th grade (~16 years ago) my substitute biology teacher organised a "Water drinking competition" poised to elevate our understanding of stomach capacity, etc.<p>Two of us went to hospital, and ~5 (including myself) had to take multiple days off of school and recover.<p>It's very dangerous. He lost his job.</p>
]]></description><pubDate>Tue, 26 May 2026 09:34:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=48277290</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=48277290</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48277290</guid></item><item><title><![CDATA[New comment by deviation in "Regex Chess: A 2-ply minimax chess engine in 84,688 regular expressions"]]></title><description><![CDATA[
<p>Not sure it's completely accurate. I played a standard queen's gambit accepted, took black's queen which it immediately blundered, then tried to move my queen from c5 -> e5 and the game ended immediately showing:<p><pre><code>  *Illegal Move*
  You Lose.
  Game over.
</code></pre>
A little disappointed, since it's of course a valid move.</p>
]]></description><pubDate>Tue, 19 May 2026 11:22:18 +0000</pubDate><link>https://news.ycombinator.com/item?id=48191886</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=48191886</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48191886</guid></item><item><title><![CDATA[New comment by deviation in "Show HN: OSS Agent I built topped the TerminalBench on Gemini-3-flash-preview"]]></title><description><![CDATA[
<p>Here's all the telemetry:<p>1. Telemetry to dirac.run/v1/event — Sends machine ID, token usage, model info, events, errors (first 500 chars), and platform info. Hardcoded API key. Defaults to opt-in (setting is "unset", not "disabled").<p>2. Feature flags from dirac.run/v1/event/decide — Polls every 60 minutes with your machine ID. Always enabled, independent of telemetry opt-out. No way to disable without code changes.<p>3. Web tools route through api.dirac.run — Web search and web fetch tools proxy through Dirac's own API server, sending your request content plus system headers (platform, version, machine ID).<p>4. Model list fetches — Calls OpenRouter, HuggingFace, Groq, etc. for model listings even when using the Anthropic provider.</p>
]]></description><pubDate>Mon, 27 Apr 2026 16:18:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=47923572</link><dc:creator>deviation</dc:creator><comments>https://news.ycombinator.com/item?id=47923572</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47923572</guid></item></channel></rss>