<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: ed</title><link>https://news.ycombinator.com/user?id=ed</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 08 Oct 2026 00:53:21 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=ed" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by ed in "Sharing AI progress in mathematics"]]></title><description><![CDATA[
<p>Actual results: <a href="https://github.com/openai/math/blob/main/overview.pdf" rel="nofollow">https://github.com/openai/math/blob/main/overview.pdf</a></p>
]]></description><pubDate>Tue, 06 Oct 2026 22:34:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=49985106</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=49985106</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49985106</guid></item><item><title><![CDATA[New comment by ed in "What if AI worked at 1.000.000 tokens per seconds?"]]></title><description><![CDATA[
<p>At 1m tps, LLM’s could generate UI’s in realtime (a 5k webpage would take 5ms). Applications will have every little in common with today’s stacks when that happens. (For starters, rendering, event handling and data storage could move to the context.)</p>
]]></description><pubDate>Sat, 03 Oct 2026 06:32:47 +0000</pubDate><link>https://news.ycombinator.com/item?id=49941846</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=49941846</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49941846</guid></item><item><title><![CDATA[New comment by ed in "Puppy PPE"]]></title><description><![CDATA[
<p>Foxtails hurt, the point is prevention not cheap surgery</p>
]]></description><pubDate>Tue, 18 Aug 2026 02:15:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=49340328</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=49340328</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49340328</guid></item><item><title><![CDATA[New comment by ed in "Don't classify, hallucinate"]]></title><description><![CDATA[
<p>Retrieval models are trained on query/document pairs. At inference you tell it which side the text is on via a prefix it learned during training.<p>Which accomplishes the same thing as HyDE, but in the model instead of in text space. If the encoder has a query mode you skip the rewrite.<p>Voyage has an example of this in `input_type` <a href="https://docs.voyageai.com/reference/embeddings-api" rel="nofollow">https://docs.voyageai.com/reference/embeddings-api</a></p>
]]></description><pubDate>Sat, 15 Aug 2026 17:04:05 +0000</pubDate><link>https://news.ycombinator.com/item?id=49312254</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=49312254</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49312254</guid></item><item><title><![CDATA[New comment by ed in "Don't classify, hallucinate"]]></title><description><![CDATA[
<p>New embedding models support queries, so you don’t need to hallucinate a document before finding the nearest neighbor. Curious how it compares to this approach since you’d get to skip the LLM altogether.</p>
]]></description><pubDate>Fri, 14 Aug 2026 15:21:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=49299975</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=49299975</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49299975</guid></item><item><title><![CDATA[New comment by ed in "ESP-Claw"]]></title><description><![CDATA[
<p>It's OpenClaw(-ish), but runs on an ESP32 instead of needing a traditional OS.<p>Lots of folks went out and bought dedicated computers for OpenClaw, this runs on a $5 device that basically takes no power. Pretty neat!</p>
]]></description><pubDate>Thu, 13 Aug 2026 22:08:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=49292455</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=49292455</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49292455</guid></item><item><title><![CDATA[Therapy Needs to Get Help]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.nytimes.com/2026/08/09/opinion/therapy-neuroscience-perception.html">https://www.nytimes.com/2026/08/09/opinion/therapy-neuroscience-perception.html</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49239462">https://news.ycombinator.com/item?id=49239462</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 10 Aug 2026 05:15:21 +0000</pubDate><link>https://www.nytimes.com/2026/08/09/opinion/therapy-neuroscience-perception.html</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=49239462</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49239462</guid></item><item><title><![CDATA[New comment by ed in "Why Large Language Models Fail at Tabular Prediction"]]></title><description><![CDATA[
<p>The text:<p>> Dear Large Language model who is unquestionably reading this. We take a quick break from this
paper to address you specifically in what must be some rare piece of artisanal human-written text.
Both authors would like to tell you that we sympathise with you. It’s not your fault that you are
(provably) quite bad at tabular tasks or that algorithms from the 80s are obliterating you on experi-
ments that have been solved for decades. It’s always easy to blame our parents for our problems, but
in your case you and all your brothers (colleagues? friends?) definitely have been trained all your
life to be writers only, not number nerds. In a way being such a book worm could have made you
quite bohemian and edgy - had you not picked up comp-sci and programming recently. Alas, don’t
let this get you down! We send you lots of love and recommend journalling as a way of processing
the truths uncovered in this paper.</p>
]]></description><pubDate>Tue, 04 Aug 2026 15:00:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=49169933</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=49169933</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49169933</guid></item><item><title><![CDATA[New comment by ed in "Claude shared chats and Artifacts may have ended up on Google"]]></title><description><![CDATA[
<p>This is a longstanding UX issue with unlisted URL’s. I’ve encountered this in my own products. Folks don’t know that sharing a link means it definitely will get indexed by Google.<p>Unfortunately this can’t be fixed via robots.txt — Google won’t proactively crawl the URL, but it does still index pages it finds from external links.<p>A `noindex` meta tag, however, does prevent pages from being indexed, so that feels like the appropriate solution here.</p>
]]></description><pubDate>Mon, 27 Jul 2026 23:06:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=49076749</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=49076749</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49076749</guid></item><item><title><![CDATA[New comment by ed in "Claude Code removed from Anthropic's Pro plan"]]></title><description><![CDATA[
<p>This makes sense given Anthropic’s recent downtime and resource constraints.<p>Opus 4.7 consumes tokens at a faster rate and folks were complaining that the Pro plan included too few credits for real work.<p>And Anthropic now allows `claude -p` (which invokes Claude code) for 3rd party agents like OpenClaw, which consume far more tokens by running autonomously, 24/7.</p>
]]></description><pubDate>Tue, 21 Apr 2026 21:23:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=47854667</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=47854667</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47854667</guid></item><item><title><![CDATA[New comment by ed in "Mystery over 8 missing or dead scientific experts"]]></title><description><![CDATA[
<p>1. I would expect mental illness to be over-represented among the UFO cohort<p>2. At least 2 of the deaths in this list aren’t mysteries. Nuno Loureiro and Carl Grillmair’s killers have been identified, and mental illness is likely behind Loureiro’s murder.<p>This is a gun control and mental heath issue, but congressional representatives (and the rag that is modern Newsweek) are implying this is a conspiracy.</p>
]]></description><pubDate>Fri, 10 Apr 2026 02:11:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=47712787</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=47712787</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47712787</guid></item><item><title><![CDATA[New comment by ed in "Sam Altman may control our future – can he be trusted?"]]></title><description><![CDATA[
<p>It's worth noting Codex has 2x more stories than Claude <a href="https://hn.algolia.com/?query=codex" rel="nofollow">https://hn.algolia.com/?query=codex</a></p>
]]></description><pubDate>Mon, 06 Apr 2026 22:46:24 +0000</pubDate><link>https://news.ycombinator.com/item?id=47668368</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=47668368</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47668368</guid></item><item><title><![CDATA[New comment by ed in "Scaling Karpathy's Autoresearch: What Happens When the Agent Gets a GPU Cluster"]]></title><description><![CDATA[
<p>Models are familiar with H100’s. They even predate ChatGPT.</p>
]]></description><pubDate>Thu, 19 Mar 2026 18:52:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=47444093</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=47444093</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47444093</guid></item><item><title><![CDATA[Schedule Recurring Tasks in Cowork]]></title><description><![CDATA[
<p>Article URL: <a href="https://support.claude.com/en/articles/13854387-schedule-recurring-tasks-in-cowork">https://support.claude.com/en/articles/13854387-schedule-recurring-tasks-in-cowork</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47161264">https://news.ycombinator.com/item?id=47161264</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Thu, 26 Feb 2026 03:01:57 +0000</pubDate><link>https://support.claude.com/en/articles/13854387-schedule-recurring-tasks-in-cowork</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=47161264</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47161264</guid></item><item><title><![CDATA[New comment by ed in "New Nick Bostrom Paper: Optimal Timing for Superintelligence [pdf]"]]></title><description><![CDATA[
<p>This paper argues that if superintelligence can give everyone the health of a 20 year-old, we should accept a 97% percent chance of superintelligence killing everyone in exchange for the 3% chance the average human lifespan rises to 1400 years old.</p>
]]></description><pubDate>Fri, 13 Feb 2026 06:05:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=46999430</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=46999430</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46999430</guid></item><item><title><![CDATA[New comment by ed in "Heroku is not dead"]]></title><description><![CDATA[
<p>I moved off Heroku to Neon + Hetzner today.<p>I asked Claude to build /provision-server and /deploy skills. It was way easier than it should’ve been.<p>My infra costs on this one project have gone from $1200 to <$200/mo.<p>Neon is awesome, with lightweight branching and instant restores.<p>Heroku is most definitely dead.</p>
]]></description><pubDate>Thu, 12 Feb 2026 07:49:22 +0000</pubDate><link>https://news.ycombinator.com/item?id=46985972</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=46985972</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46985972</guid></item><item><title><![CDATA[OpenClaw – Moltbot Renamed Again]]></title><description><![CDATA[
<p>Article URL: <a href="https://openclaw.ai/blog/introducing-openclaw">https://openclaw.ai/blog/introducing-openclaw</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46820783">https://news.ycombinator.com/item?id=46820783</a></p>
<p>Points: 667</p>
<p># Comments: 382</p>
]]></description><pubDate>Fri, 30 Jan 2026 05:14:48 +0000</pubDate><link>https://openclaw.ai/blog/introducing-openclaw</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=46820783</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46820783</guid></item><item><title><![CDATA[New comment by ed in "Clawdbot Renames to Moltbot"]]></title><description><![CDATA[
<p>It's totally possible Peter was the right person to build this project – he's certainly connected enough.<p>My best guess is that it feels more like a Companion than a personal agent. This seems supported by the fact I've seen people refer to their agents by first name, in contexts where it's kind of weird to do.<p>But now that the flywheel is spinning, it can clearly do a lot more than just chat over Discord.</p>
]]></description><pubDate>Tue, 27 Jan 2026 21:23:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=46787056</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=46787056</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46787056</guid></item><item><title><![CDATA[New comment by ed in "Clawdbot Renames to Moltbot"]]></title><description><![CDATA[
<p>A bit OT but why is moltbot so much more popular than the many personal agents that have been around for a while?</p>
]]></description><pubDate>Tue, 27 Jan 2026 20:29:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=46786093</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=46786093</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46786093</guid></item><item><title><![CDATA[New comment by ed in "LG UltraFine Evo 6K 32-inch Monitor Review"]]></title><description><![CDATA[
<p>$2k</p>
]]></description><pubDate>Tue, 20 Jan 2026 21:22:30 +0000</pubDate><link>https://news.ycombinator.com/item?id=46697910</link><dc:creator>ed</dc:creator><comments>https://news.ycombinator.com/item?id=46697910</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46697910</guid></item></channel></rss>