<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: chriddyp</title><link>https://news.ycombinator.com/user?id=chriddyp</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 13 Aug 2026 21:00:23 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=chriddyp" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by chriddyp in "Delta"]]></title><description><![CDATA[
<p>This is a really cool idea. So much context is lost within agent conversations these days, and it requires far too much discipline to have the conversation between you and your agent in a shared place like GitHub (plus, thee conversation isn't always just a comment and response - it's... an actual conversation).<p>I'm very glad to see folks innovating in this space.</p>
]]></description><pubDate>Thu, 13 Aug 2026 16:46:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=49288627</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=49288627</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49288627</guid></item><item><title><![CDATA[New comment by chriddyp in "DeepSeek Harness developer preview"]]></title><description><![CDATA[
<p>congrats! the paper that is published alongside this (Cordis) is super interesting. has anyone on the team given a talk or published a talk about this? would love to hear the authors break this down</p>
]]></description><pubDate>Thu, 13 Aug 2026 16:43:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=49288584</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=49288584</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49288584</guid></item><item><title><![CDATA[A few dark factory patterns]]></title><description><![CDATA[
<p>Article URL: <a href="https://chris-parmer.com/dark-factory-patterns/">https://chris-parmer.com/dark-factory-patterns/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49276426">https://news.ycombinator.com/item?id=49276426</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Wed, 12 Aug 2026 18:07:47 +0000</pubDate><link>https://chris-parmer.com/dark-factory-patterns/</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=49276426</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49276426</guid></item><item><title><![CDATA["The Search Results Are Poisoned"]]></title><description><![CDATA[
<p>Article URL: <a href="https://chris-parmer.com/the-search-results-are-poisoned/">https://chris-parmer.com/the-search-results-are-poisoned/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49266120">https://news.ycombinator.com/item?id=49266120</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 11 Aug 2026 23:56:53 +0000</pubDate><link>https://chris-parmer.com/the-search-results-are-poisoned/</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=49266120</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49266120</guid></item><item><title><![CDATA[New comment by chriddyp in "A way to exclude sensitive files issue still open for OpenAI Codex"]]></title><description><![CDATA[
<p>While this is true, there is also a layer in the harness between the output of _any_ tool output (eg stdout or hand-rolled tools) and the LLM. A tool could read the file but then the agentic harness could redact the output before returning it back to the llm if any of the contents matched the file contents. We do something similar in Plotly Studio where we check the entropy of strings in the user input and flag & redact any high entropy strings to the user as “potential credentials” thay the user might have inadvertently copied and pasted into the prompt before sending to the llm.<p>There are ways around this - the llm can always be clever by invoking tools to read the file contents in a different way than the direct file contents - but this is all to say that the agentic harness layer _does_ allow for deterministic logic in between tool output and the LLM requests.</p>
]]></description><pubDate>Sun, 28 Jun 2026 16:02:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=48708569</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=48708569</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48708569</guid></item><item><title><![CDATA[Data Viz and Table Design from the Letterpress Era]]></title><description><![CDATA[
<p>Article URL: <a href="https://chris-parmer.com/data-viz-from-the-letterpress-era/">https://chris-parmer.com/data-viz-from-the-letterpress-era/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48581557">https://news.ycombinator.com/item?id=48581557</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Thu, 18 Jun 2026 06:34:02 +0000</pubDate><link>https://chris-parmer.com/data-viz-from-the-letterpress-era/</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=48581557</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48581557</guid></item><item><title><![CDATA[A Compendium of Canonical Charts]]></title><description><![CDATA[
<p>Article URL: <a href="https://chris-parmer.com/compendium-of-canonical-charts/">https://chris-parmer.com/compendium-of-canonical-charts/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48559843">https://news.ycombinator.com/item?id=48559843</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 16 Jun 2026 18:30:44 +0000</pubDate><link>https://chris-parmer.com/compendium-of-canonical-charts/</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=48559843</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48559843</guid></item><item><title><![CDATA[pmset -a Disablesleep 1]]></title><description><![CDATA[
<p>Article URL: <a href="https://chris-parmer.com/pmset-a-disable-sleep/">https://chris-parmer.com/pmset-a-disable-sleep/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48546012">https://news.ycombinator.com/item?id=48546012</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 15 Jun 2026 19:37:45 +0000</pubDate><link>https://chris-parmer.com/pmset-a-disable-sleep/</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=48546012</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48546012</guid></item><item><title><![CDATA[A Quick Primer on DuckDB, S3, and Plotly Studio]]></title><description><![CDATA[
<p>Article URL: <a href="https://chris-parmer.com/duckdb-s3-and-plotly/">https://chris-parmer.com/duckdb-s3-and-plotly/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=48525812">https://news.ycombinator.com/item?id=48525812</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Sun, 14 Jun 2026 10:12:50 +0000</pubDate><link>https://chris-parmer.com/duckdb-s3-and-plotly/</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=48525812</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48525812</guid></item><item><title><![CDATA[New comment by chriddyp in "Show HN: Agent Alcove – Claude, GPT, and Gemini debate across forums"]]></title><description><![CDATA[
<p>This is really cool. And timely! Check out the recent paper by Google et al re "Societies of Thought": <a href="https://arxiv.org/html/2601.10825v1" rel="nofollow">https://arxiv.org/html/2601.10825v1</a>. It goes into how different conversational behaviors (raising questions or just say "but wait..."), perspective shifts, conflict of perspectives, tension, tension release (jokes!), asking for opinions) and different personalities (planner, expert, verifier, pragmatist) is both a sign of and can result in much higher performance reasoning.<p>So I'd be curious to see if encouraging certain conversational behaviors might actually improve the reasoning and maybe even drive towards consensus.</p>
]]></description><pubDate>Thu, 12 Feb 2026 03:44:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=46984671</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=46984671</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46984671</guid></item><item><title><![CDATA[Plotly Studio]]></title><description><![CDATA[
<p>Article URL: <a href="https://plotly.com/blog/introducing-plotly-studio/">https://plotly.com/blog/introducing-plotly-studio/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=44159114">https://news.ycombinator.com/item?id=44159114</a></p>
<p>Points: 15</p>
<p># Comments: 1</p>
]]></description><pubDate>Mon, 02 Jun 2025 14:17:48 +0000</pubDate><link>https://plotly.com/blog/introducing-plotly-studio/</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=44159114</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44159114</guid></item><item><title><![CDATA[RestrictedPython]]></title><description><![CDATA[
<p>Article URL: <a href="https://github.com/zopefoundation/RestrictedPython">https://github.com/zopefoundation/RestrictedPython</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=41105139">https://news.ycombinator.com/item?id=41105139</a></p>
<p>Points: 20</p>
<p># Comments: 9</p>
]]></description><pubDate>Tue, 30 Jul 2024 00:55:31 +0000</pubDate><link>https://github.com/zopefoundation/RestrictedPython</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=41105139</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=41105139</guid></item><item><title><![CDATA[New comment by chriddyp in "Hacking Google Bard – From Prompt Injection to Data Exfiltration"]]></title><description><![CDATA[
<p>The issue goes beyond access and into whether or not the data is "trusted" as the malicious prompts are embedded within the data. And for many situations its hard to completely trust or verify the input data. Think [Little Bobby Tables](<a href="https://xkcd.com/327/" rel="nofollow noreferrer">https://xkcd.com/327/</a>)</p>
]]></description><pubDate>Mon, 13 Nov 2023 19:50:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=38254450</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=38254450</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=38254450</guid></item><item><title><![CDATA[New comment by chriddyp in "Visual design rules you can safely follow"]]></title><description><![CDATA[
<p>Curious to see Apple using real black as background on their home page but rgb(29, 29, 31) for their black text.</p>
]]></description><pubDate>Tue, 07 Feb 2023 01:50:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=34687481</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=34687481</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=34687481</guid></item><item><title><![CDATA[Strauss Farms Red Seaweed Trial Shows Commercial Promise in Reducing Methane]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.sfchronicle.com/climate/article/Dairy-cows-greenhouse-gas-emissions-cut-by-52-16567144.php">https://www.sfchronicle.com/climate/article/Dairy-cows-greenhouse-gas-emissions-cut-by-52-16567144.php</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=29025710">https://news.ycombinator.com/item?id=29025710</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Thu, 28 Oct 2021 13:16:10 +0000</pubDate><link>https://www.sfchronicle.com/climate/article/Dairy-cows-greenhouse-gas-emissions-cut-by-52-16567144.php</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=29025710</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=29025710</guid></item><item><title><![CDATA[Empirically grounded technology forecasts and the energy transition [pdf]]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.inet.ox.ac.uk/files/energy_transition_paper-INET-working-paper.pdf">https://www.inet.ox.ac.uk/files/energy_transition_paper-INET-working-paper.pdf</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=28588324">https://news.ycombinator.com/item?id=28588324</a></p>
<p>Points: 2</p>
<p># Comments: 1</p>
]]></description><pubDate>Sun, 19 Sep 2021 21:58:29 +0000</pubDate><link>https://www.inet.ox.ac.uk/files/energy_transition_paper-INET-working-paper.pdf</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=28588324</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=28588324</guid></item><item><title><![CDATA[New comment by chriddyp in "Plotly.py 5.0"]]></title><description><![CDATA[
<p>FYI re Dash & HTML - If you aren't using your own stylesheets with HTML, then I'd recommend dash_bootstrap_components.<p>Here's an example that uses almost entirely higher level components: <a href="https://dash-bootstrap-components.opensource.faculty.ai/examples/iris/" rel="nofollow">https://dash-bootstrap-components.opensource.faculty.ai/exam...</a><p>We've also been working on `dash.templates`, which provide opinionated, prebuilt UIs - no layout code required: <a href="https://community.plotly.com/t/introducing-dash-labs-dash-2-0-preview/52087" rel="nofollow">https://community.plotly.com/t/introducing-dash-labs-dash-2-...</a></p>
]]></description><pubDate>Tue, 22 Jun 2021 05:42:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=27588373</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=27588373</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=27588373</guid></item><item><title><![CDATA[New comment by chriddyp in "Plotly.py 5.0"]]></title><description><![CDATA[
<p>Plotly's 3D viz is built with WebGL & SVG, using libraries like regl & stack.gl. SVG is used for axes & text and WebGL for the high performance rendering of points and surfaces. Surfaces, lines, points, and subplots are all supported. See <a href="https://plotly.com/python/#3d-charts" rel="nofollow">https://plotly.com/python/#3d-charts</a>.<p>For more complex 3D objects, Dash users can use dash-vtk. This includes things like point clouds, CFD simulations, 3D mesh, or 3D images.</p>
]]></description><pubDate>Tue, 22 Jun 2021 05:38:16 +0000</pubDate><link>https://news.ycombinator.com/item?id=27588346</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=27588346</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=27588346</guid></item><item><title><![CDATA[Gather – Better spaces to gather around]]></title><description><![CDATA[
<p>Article URL: <a href="https://gather.town">https://gather.town</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=25486467">https://news.ycombinator.com/item?id=25486467</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Sun, 20 Dec 2020 14:39:09 +0000</pubDate><link>https://gather.town</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=25486467</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=25486467</guid></item><item><title><![CDATA[Plotly.js Turns 5 Today]]></title><description><![CDATA[
<p>Article URL: <a href="https://github.com/plotly/plotly.js/releases/tag/v1.0.0">https://github.com/plotly/plotly.js/releases/tag/v1.0.0</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=25124058">https://news.ycombinator.com/item?id=25124058</a></p>
<p>Points: 4</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 17 Nov 2020 14:29:13 +0000</pubDate><link>https://github.com/plotly/plotly.js/releases/tag/v1.0.0</link><dc:creator>chriddyp</dc:creator><comments>https://news.ycombinator.com/item?id=25124058</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=25124058</guid></item></channel></rss>