<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: floitsch</title><link>https://news.ycombinator.com/user?id=floitsch</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Wed, 18 Nov 2026 19:18:14 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=floitsch" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by floitsch in "Toit v2.0.0"]]></title><description><![CDATA[
<p>After Dart, where we "had to" use familiar syntax (JavaScript), we experimented with less-noisy syntax for Toit. We removed everything that wasn't necessary first. Over time we added some parts (like `extends`, `implements`, ...) back, but we got used to the non-parens calls (similar to shell or Caml) and like it now. To me, every other language feels cluttered and noisy.<p>Wrt performance: depending on the program/benchmark Toit is sometimes 10x faster than MicroPython. For small loops that just do some arithmetic it's less noticeable, but real programs with method-calls benefit a lot from the static class definitions.</p>
]]></description><pubDate>Tue, 03 Nov 2026 15:01:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=50352870</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=50352870</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50352870</guid></item><item><title><![CDATA[New comment by floitsch in "Toit v2.0.0"]]></title><description><![CDATA[
<p>We probably could have made it work without the `--`.  Coming from Posix, the `--` feels natural, and it also allows for code completion.</p>
]]></description><pubDate>Tue, 03 Nov 2026 13:18:18 +0000</pubDate><link>https://news.ycombinator.com/item?id=50351623</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=50351623</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50351623</guid></item><item><title><![CDATA[New comment by floitsch in "Toit v2.0.0"]]></title><description><![CDATA[
<p>It's a good point.<p>LLMs have not been trained on Toit code as much as they have on Python or JS. No question about that. Without the Toit skills (<a href="https://github.com/toitlang/ai-instructions" rel="nofollow">https://github.com/toitlang/ai-instructions</a>) LLMs used to be really bad. (They might be better now, but I never bothered to remove the skill files). From that point of view, Toit isn't a good choice.<p>On the other hand, Toit provides containers and a fast refresh-cycle. Both are godsends for AI agents: the containers make it so that the agent doesn't accidentally bring down the whole system when there is a bug, and the refresh cycle makes iterating on programs much faster. (Something I find agents to be really good at).<p>Fwiw, there isn't any clever metaprogramming in Toit.</p>
]]></description><pubDate>Tue, 03 Nov 2026 13:10:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=50351556</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=50351556</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50351556</guid></item><item><title><![CDATA[New comment by floitsch in "Toit v2.0.0"]]></title><description><![CDATA[
<p>A few more bullet points to encourage readers to have a look at Toit:<p>- Faster than Python. Depending on the benchmark up to x10.<p>- Can be compiled to executables. Currently `-o` just takes an interpreter and attaches the bytecodes, but Kasper also started experimenting with native output.<p>- Cleaner base. Objects aren't just hashtables. This can be seen as a downside as well (making the language less flexible), but in my book it's a win.<p>- The `--named=value` syntax is just a winner :)</p>
]]></description><pubDate>Tue, 03 Nov 2026 12:47:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=50351363</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=50351363</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50351363</guid></item><item><title><![CDATA[New comment by floitsch in "Toit v2.0.0"]]></title><description><![CDATA[
<p>I think "toy language" isn't really fair. Toit has been developed for 7 years, some of which with multiple engineers working on it full time. (The blog post says 4 years, but that's just for the v2.0.0 development).<p>Toit doesn't have a lot of users, but it is used in production (for example by Consibio) and is definitely developed for production environments.</p>
]]></description><pubDate>Tue, 03 Nov 2026 11:21:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=50350611</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=50350611</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50350611</guid></item><item><title><![CDATA[New comment by floitsch in "Toit v2.0.0"]]></title><description><![CDATA[
<p>We looked at these when we started. The problem with them is that they weren't designed with constrained resources in mind and it shows. For example, objects are hash tables and method calls are thus expensive. An obligatory top level execution makes startup slow, ...<p>If you have a bigger machine you can optimize some of this (V8), but these optimizations typically need memory and/or a JIT. (And Python struggles to optimize even on stronger machines).<p>By starting from scratch (with what we learned from V8 and Dart) you can design something that feels similar but is much cleaner and faster to run.</p>
]]></description><pubDate>Tue, 03 Nov 2026 11:14:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=50350542</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=50350542</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50350542</guid></item><item><title><![CDATA[New comment by floitsch in "Toit v2.0.0"]]></title><description><![CDATA[
<p>It's from Austin Powers.<p>"toit like a tiger"<p><a href="https://youtu.be/gu31VyXlTzo" rel="nofollow">https://youtu.be/gu31VyXlTzo</a></p>
]]></description><pubDate>Tue, 03 Nov 2026 09:18:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=50349774</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=50349774</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50349774</guid></item><item><title><![CDATA[New comment by floitsch in "Toit v2.0.0"]]></title><description><![CDATA[
<p>Kasper and I are still working on it.</p>
]]></description><pubDate>Tue, 03 Nov 2026 08:28:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=50349484</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=50349484</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50349484</guid></item><item><title><![CDATA[Toit v2.0.0]]></title><description><![CDATA[
<p>Article URL: <a href="https://toit.io/blog/toit-v2-0-0/">https://toit.io/blog/toit-v2-0-0/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=50321605">https://news.ycombinator.com/item?id=50321605</a></p>
<p>Points: 238</p>
<p># Comments: 72</p>
]]></description><pubDate>Sun, 01 Nov 2026 00:25:04 +0000</pubDate><link>https://toit.io/blog/toit-v2-0-0/</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=50321605</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50321605</guid></item><item><title><![CDATA[Eight-year study links more childhood screen time to better cognitive processing]]></title><description><![CDATA[
<p>Article URL: <a href="https://spacedaily.com/b-researchers-in-finland-tracked-childrens-screen-time-for-eight-years-expecting-to-document-the-damage-the-children-who-spent-more-hours-on-screens-showed-better-cognitive-processing-as-teenagers/">https://spacedaily.com/b-researchers-in-finland-tracked-childrens-screen-time-for-eight-years-expecting-to-document-the-damage-the-children-who-spent-more-hours-on-screens-showed-better-cognitive-processing-as-teenagers/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49399101">https://news.ycombinator.com/item?id=49399101</a></p>
<p>Points: 4</p>
<p># Comments: 0</p>
]]></description><pubDate>Sat, 22 Aug 2026 12:33:17 +0000</pubDate><link>https://spacedaily.com/b-researchers-in-finland-tracked-childrens-screen-time-for-eight-years-expecting-to-document-the-damage-the-children-who-spent-more-hours-on-screens-showed-better-cognitive-processing-as-teenagers/</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=49399101</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49399101</guid></item><item><title><![CDATA[Day-ahead electricity prices hit –€500/MWh in France and Germany]]></title><description><![CDATA[
<p>Article URL: <a href="https://data.nordpoolgroup.com/auction/day-ahead/prices?deliveryDate=latest&currency=EUR&aggregation=DeliveryPeriod&deliveryAreas=AT,FR,GER">https://data.nordpoolgroup.com/auction/day-ahead/prices?deliveryDate=latest&currency=EUR&aggregation=DeliveryPeriod&deliveryAreas=AT,FR,GER</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47960854">https://news.ycombinator.com/item?id=47960854</a></p>
<p>Points: 9</p>
<p># Comments: 0</p>
]]></description><pubDate>Thu, 30 Apr 2026 11:19:57 +0000</pubDate><link>https://data.nordpoolgroup.com/auction/day-ahead/prices?deliveryDate=latest&amp;currency=EUR&amp;aggregation=DeliveryPeriod&amp;deliveryAreas=AT,FR,GER</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=47960854</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47960854</guid></item><item><title><![CDATA[New comment by floitsch in "Prism"]]></title><description><![CDATA[
<p>I felt the same, but then thought of experts in their field. For example, my PhD advisor would already know all these papers. For him the prompt would actually be similar to what was shown in the video.</p>
]]></description><pubDate>Wed, 28 Jan 2026 12:57:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=46794713</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=46794713</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46794713</guid></item><item><title><![CDATA[New comment by floitsch in "A glimpse into V8 development for RISC-V"]]></title><description><![CDATA[
<p>You are absolutely right, and a follow-up CL fixed the non-Zba code path. At the time of writing of the blog post, the generated code was however still using the 4 instructions instead of just two.<p>V8 is a huge project and ports almost have to start from backends of existing architectures. Over time we (I recently worked on the RISC-V port) improve the situation.<p>Note that it's also crucial to stay somehow similar to the x86 and ARM backends as the V8 team frequently makes changes to these and keeping up with them would be a nightmare if the RISC-V backend was too different.<p>"Does no one at Google know RISC-V"
This work wasn't done by Google. Also, you don't hire RISC-V experts, but V8 experts (or other similar VMs) for such a port. I, for example, had never worked with RISC-V before. As such, it can happen that some code of the RISC-V backend is written by engineers that aren't yet experts in RISC-V.</p>
]]></description><pubDate>Wed, 07 Jan 2026 22:58:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=46534449</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=46534449</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46534449</guid></item><item><title><![CDATA[A glimpse into V8 development for RISC-V]]></title><description><![CDATA[
<p>Article URL: <a href="https://riseproject.dev/2025/12/09/a-glimpse-into-v8-development-for-risc-v/">https://riseproject.dev/2025/12/09/a-glimpse-into-v8-development-for-risc-v/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=46522686">https://news.ycombinator.com/item?id=46522686</a></p>
<p>Points: 65</p>
<p># Comments: 8</p>
]]></description><pubDate>Wed, 07 Jan 2026 04:48:50 +0000</pubDate><link>https://riseproject.dev/2025/12/09/a-glimpse-into-v8-development-for-risc-v/</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=46522686</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46522686</guid></item><item><title><![CDATA[New comment by floitsch in "KDE onboarding is good now"]]></title><description><![CDATA[
<p>When Plasma crashes, all programs keep running. The only noticable thing (to me) is that the order of my apps in the task bar is different.</p>
]]></description><pubDate>Sun, 04 Jan 2026 10:43:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=46486781</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=46486781</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46486781</guid></item><item><title><![CDATA[New comment by floitsch in "Zmij: Faster floating point double-to-string conversion"]]></title><description><![CDATA[
<p>Pretty impressive.<p>When I published Grisu (Google double-conversion), it was multiple times faster than the existing algorithms. I knew that there was still room for improvement, but I was at most expecting a factor 2 or so. Six times faster is really impressive.</p>
]]></description><pubDate>Wed, 17 Dec 2025 20:45:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=46305277</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=46305277</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46305277</guid></item><item><title><![CDATA[Make the most of compiled C loops on the 68000]]></title><description><![CDATA[
<p>Article URL: <a href="https://dciabrin.net/posts/make-the-most-of-compiled-c-loops-on-the-68000/make-the-most-of-compiled-c-loops-on-the-68000/">https://dciabrin.net/posts/make-the-most-of-compiled-c-loops-on-the-68000/make-the-most-of-compiled-c-loops-on-the-68000/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=45406879">https://news.ycombinator.com/item?id=45406879</a></p>
<p>Points: 69</p>
<p># Comments: 19</p>
]]></description><pubDate>Sun, 28 Sep 2025 18:55:49 +0000</pubDate><link>https://dciabrin.net/posts/make-the-most-of-compiled-c-loops-on-the-68000/make-the-most-of-compiled-c-loops-on-the-68000/</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=45406879</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45406879</guid></item><item><title><![CDATA[New comment by floitsch in "Regarding the Monetization of Project Maintenance"]]></title><description><![CDATA[
<p>Sleef is a SIMD library that is often used (transitively) when working with LLMs.<p>Its maintainer now requests $1000 for approval of PRs.<p>Here is an example where they reject a small PR to add loongarch64 support from individual contributor: <a href="https://github.com/shibatch/sleef/pull/672">https://github.com/shibatch/sleef/pull/672</a></p>
]]></description><pubDate>Sun, 03 Aug 2025 18:57:45 +0000</pubDate><link>https://news.ycombinator.com/item?id=44778797</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=44778797</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44778797</guid></item><item><title><![CDATA[Regarding the Monetization of Project Maintenance]]></title><description><![CDATA[
<p>Article URL: <a href="https://github.com/shibatch/sleef/discussions/632">https://github.com/shibatch/sleef/discussions/632</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=44778796">https://news.ycombinator.com/item?id=44778796</a></p>
<p>Points: 1</p>
<p># Comments: 2</p>
]]></description><pubDate>Sun, 03 Aug 2025 18:57:45 +0000</pubDate><link>https://github.com/shibatch/sleef/discussions/632</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=44778796</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44778796</guid></item><item><title><![CDATA[New comment by floitsch in "Ring introducing new feature to allow police to live-stream access to cameras"]]></title><description><![CDATA[
<p>If they can get the data without a user's consent, then it's independent of this new feature and thus unrelated. If you believe that the government has unlimited access, then it was most likely already possible before this feature.
Now, there is at least a "proper" way to give law enforcement access.</p>
]]></description><pubDate>Sun, 20 Jul 2025 08:57:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=44623200</link><dc:creator>floitsch</dc:creator><comments>https://news.ycombinator.com/item?id=44623200</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44623200</guid></item></channel></rss>