<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: Jaxan</title><link>https://news.ycombinator.com/user?id=Jaxan</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 17 Sep 2026 17:39:03 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=Jaxan" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by Jaxan in "Rate limits on GitLab.com are changing"]]></title><description><![CDATA[
<p>Yes. A lot is not accessible if not logged in.</p>
]]></description><pubDate>Thu, 17 Sep 2026 16:50:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=49743415</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49743415</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49743415</guid></item><item><title><![CDATA[New comment by Jaxan in "Rate limits on GitLab.com are changing"]]></title><description><![CDATA[
<p>Doesn’t help with scrapers though. They use a unique IP for each query.</p>
]]></description><pubDate>Thu, 17 Sep 2026 16:49:57 +0000</pubDate><link>https://news.ycombinator.com/item?id=49743402</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49743402</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49743402</guid></item><item><title><![CDATA[New comment by Jaxan in "Online Z3 Guide"]]></title><description><![CDATA[
<p>Sometimes you can use SMT for “theorem proving”. It is a rather broad term. I don’t think they added something much different than what they already had.</p>
]]></description><pubDate>Thu, 17 Sep 2026 11:18:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=49739160</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49739160</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49739160</guid></item><item><title><![CDATA[New comment by Jaxan in "A beginning for mathematics"]]></title><description><![CDATA[
<p>With human proofs, I have some trust in process behind it.</p>
]]></description><pubDate>Tue, 15 Sep 2026 08:41:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=49709552</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49709552</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49709552</guid></item><item><title><![CDATA[New comment by Jaxan in "A beginning for mathematics"]]></title><description><![CDATA[
<p>The recent proof of Fermats Last Theorem is interesting: it is (iirc) 13 million lines of lean code. And type-checking takes 5 hours or so on a pretty beefy machine. <i>I cannot</i> independently verify the proof, and I have to take Anthropics word for it that it actually type-checks.</p>
]]></description><pubDate>Mon, 14 Sep 2026 20:23:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=49703389</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49703389</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49703389</guid></item><item><title><![CDATA[New comment by Jaxan in "Show HN: Art – draw one stroke, let symmetry complete it"]]></title><description><![CDATA[
<p>I looked in the AppStore for “symmetry draw”, and there are at least 10 such apps.</p>
]]></description><pubDate>Thu, 10 Sep 2026 16:44:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=49646668</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49646668</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49646668</guid></item><item><title><![CDATA[New comment by Jaxan in "Formalizing Fermat's Last Theorem"]]></title><description><![CDATA[
<p>Wouldn’t a lot already be in leans mathlib?</p>
]]></description><pubDate>Sat, 05 Sep 2026 05:48:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=49573443</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49573443</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49573443</guid></item><item><title><![CDATA[New comment by Jaxan in "Formalizing Fermat's Last Theorem"]]></title><description><![CDATA[
<p>Most systems i have seen are way beyond a 100 lines. And their GitHub repository contain many issues, often soundness bugs. (Granted, many get fixed very fast.)</p>
]]></description><pubDate>Fri, 04 Sep 2026 19:41:57 +0000</pubDate><link>https://news.ycombinator.com/item?id=49569291</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49569291</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49569291</guid></item><item><title><![CDATA[New comment by Jaxan in "Formalizing Fermat's Last Theorem"]]></title><description><![CDATA[
<p>This is a crucial point. There have been many bugs in Lean (and in other proof assistants for that matter). Proof assistants work well on human input, because it was created with a certain intent.<p>We simply don’t know what those 13M contain and whether it “makes sense” and doesn’t trigger Lean bugs. (There are “independent” lean verifiers, but historically they contained the same, or similar, bugs.)</p>
]]></description><pubDate>Fri, 04 Sep 2026 19:40:27 +0000</pubDate><link>https://news.ycombinator.com/item?id=49569268</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49569268</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49569268</guid></item><item><title><![CDATA[New comment by Jaxan in "We found a division by zero bug in FFmpeg with a vibecoded fuzzer"]]></title><description><![CDATA[
<p>I guess it’s submitted for the method rather than the result.</p>
]]></description><pubDate>Fri, 28 Aug 2026 06:22:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=49475005</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49475005</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49475005</guid></item><item><title><![CDATA[New comment by Jaxan in "Show HN: Make your logo extra bright on HDR screens"]]></title><description><![CDATA[
<p>It also depends on your screen brightness. In my case, if it’s set to maximum, there is no difference between SDR and HDR.</p>
]]></description><pubDate>Sat, 22 Aug 2026 20:19:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=49403463</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49403463</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49403463</guid></item><item><title><![CDATA[New comment by Jaxan in "Cave of the Crystals"]]></title><description><![CDATA[
<p>Absolutely amazing. I recently searched this and google images showed so much ai slop. I don’t get that, why do people prefer generic Disney style crystal caves above the actual photographs of this amazing cave.</p>
]]></description><pubDate>Fri, 14 Aug 2026 09:12:19 +0000</pubDate><link>https://news.ycombinator.com/item?id=49296333</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49296333</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49296333</guid></item><item><title><![CDATA[New comment by Jaxan in "Dithered QR Codes"]]></title><description><![CDATA[
<p>The specification of qr codes says you should look only at the centres, iirc.</p>
]]></description><pubDate>Sun, 09 Aug 2026 16:29:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=49232903</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49232903</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49232903</guid></item><item><title><![CDATA[New comment by Jaxan in "Show HN: Reverse Minesweeper"]]></title><description><![CDATA[
<p>I know it as binairo, which I have seen in paper magazines. On Simon Tatham’s puzzle collection it is known as unruly: <a href="https://www.chiark.greenend.org.uk/~sgtatham/puzzles/js/unruly.html" rel="nofollow">https://www.chiark.greenend.org.uk/~sgtatham/puzzles/js/unru...</a></p>
]]></description><pubDate>Mon, 27 Jul 2026 06:54:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=49065973</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49065973</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49065973</guid></item><item><title><![CDATA[New comment by Jaxan in "Show HN: Reverse Minesweeper"]]></title><description><![CDATA[
<p>This is very similar to “Mosaic” on Simon Tatham’s puzzle webpage. There it is noted that: “This game is variously known in other locations as: ArtMosaico, Count and Darken, Cuenta Y Sombrea, Fill-a-Pix, Fill-In, Komsu Karala, Magipic, Majipiku, Mosaico, Mosaik, Mozaiek, Nampre Puzzle, Nurie-Puzzle, Oekaki-Pix, Voisimage.”<p><a href="https://www.chiark.greenend.org.uk/~sgtatham/puzzles/js/mosaic.html" rel="nofollow">https://www.chiark.greenend.org.uk/~sgtatham/puzzles/js/mosa...</a></p>
]]></description><pubDate>Mon, 27 Jul 2026 06:50:57 +0000</pubDate><link>https://news.ycombinator.com/item?id=49065946</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49065946</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49065946</guid></item><item><title><![CDATA[New comment by Jaxan in "Stolen Buttons"]]></title><description><![CDATA[
<p>Oh this reminds me that we used to create buttons for navigation in flash!</p>
]]></description><pubDate>Sat, 25 Jul 2026 17:59:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=49049948</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=49049948</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49049948</guid></item><item><title><![CDATA[New comment by Jaxan in "Blender 5.2 LTS"]]></title><description><![CDATA[
<p>Market size is different: everyone needs a mail client, but not everyone needs 3d modelling.</p>
]]></description><pubDate>Sun, 19 Jul 2026 14:01:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=48968311</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=48968311</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48968311</guid></item><item><title><![CDATA[New comment by Jaxan in "Thunderbird Desktop settings research: what we learned from your feedback"]]></title><description><![CDATA[
<p>Hmm strange, I have never had a thunderbird folder in my home dir. I use thunderbird on Mac, Windows and Linux (Ubuntu).</p>
]]></description><pubDate>Mon, 13 Jul 2026 20:16:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=48898185</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=48898185</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48898185</guid></item><item><title><![CDATA[New comment by Jaxan in "Counting ArXiv Delays"]]></title><description><![CDATA[
<p>You can also just put a paper on your website (or google drive). If arXiv isn’t working for you, why submit there?<p>Personally, i see no problem with delays, research takes longer than a few days. Reviews take a few weeks or months even.</p>
]]></description><pubDate>Mon, 13 Jul 2026 20:12:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=48898144</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=48898144</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48898144</guid></item><item><title><![CDATA[New comment by Jaxan in "Vint Cerf, a “father of the Internet”, is retiring"]]></title><description><![CDATA[
<p>Which also shows there isn’t “one father”, multiple things (and people) had to come together.</p>
]]></description><pubDate>Sun, 12 Jul 2026 08:57:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=48879527</link><dc:creator>Jaxan</dc:creator><comments>https://news.ycombinator.com/item?id=48879527</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48879527</guid></item></channel></rss>