<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: razemio</title><link>https://news.ycombinator.com/user?id=razemio</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Mon, 17 Aug 2026 07:16:58 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=razemio" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by razemio in "What happens when an LLM never sees material beyond fifth grade?"]]></title><description><![CDATA[
<p>Exactly, I have the luxury to be able to use all top tier models without limitation. Fable and OPUS most definitely say that you are wrong and try to proove it most of the time with links to sources or math. I even had an argument with Fable where I was 100% sure it was wrong and tried to explain the issue. Turns out I was wrong. Fable did NOT give an inch. Always said, you are wrong, let me try to explain it like this. It even made a graphic when I did not get it. 
The times of LLMs only saying yes is over since 3-6 months. Where it still lags are decisions for infrastructure. It makes a plan. I say "Why not this?" and it responds with "that is much better" in 90% of the cases. However, I am not sure how to solve this. I also do not want an LLM to say: "I wont implement this."</p>
]]></description><pubDate>Sun, 16 Aug 2026 16:57:42 +0000</pubDate><link>https://news.ycombinator.com/item?id=49321692</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=49321692</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49321692</guid></item><item><title><![CDATA[New comment by razemio in "Mistral OCR 4.1"]]></title><description><![CDATA[
<p>Ah okay. Ofc handwriting is not copyrighted. Makes sense, that we do not have these issues.</p>
]]></description><pubDate>Fri, 14 Aug 2026 05:42:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=49295093</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=49295093</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49295093</guid></item><item><title><![CDATA[New comment by razemio in "Mistral OCR 4.1"]]></title><description><![CDATA[
<p>I made a benchmark for handwriting recognition for a project while keeping line breaks and errors (grammar, spelling). Sonnet absolutely dominates it since a good half a year. 5.6 did not change that for me. This should also translate to better ocr.</p>
]]></description><pubDate>Thu, 13 Aug 2026 22:41:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=49292695</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=49292695</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49292695</guid></item><item><title><![CDATA[New comment by razemio in "The primary purpose of code review is to find code that will be hard to maintain"]]></title><description><![CDATA[
<p>You are right. Let me rephrase and my argument still stands: „To even believe there is a primary purpose…“</p>
]]></description><pubDate>Mon, 06 Jul 2026 16:43:19 +0000</pubDate><link>https://news.ycombinator.com/item?id=48807146</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=48807146</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48807146</guid></item><item><title><![CDATA[New comment by razemio in "The primary purpose of code review is to find code that will be hard to maintain"]]></title><description><![CDATA[
<p>Thank you! I find it somewhat ignorant to say, most people do not understand the reason behind a peer review, while obviously being missinformed. To even believe that it only serves a single purpose somewhat tells me, that the author is missing experiences with other teams / people.</p>
]]></description><pubDate>Thu, 02 Jul 2026 15:17:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=48762941</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=48762941</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48762941</guid></item><item><title><![CDATA[New comment by razemio in "Previewing GPT‑5.6 Sol: a next-generation model"]]></title><description><![CDATA[
<p>Not sure if you mean something else, but the model itself is not able to send anything.</p>
]]></description><pubDate>Sat, 27 Jun 2026 17:56:46 +0000</pubDate><link>https://news.ycombinator.com/item?id=48700285</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=48700285</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48700285</guid></item><item><title><![CDATA[New comment by razemio in "I love the work of the ArchWiki maintainers"]]></title><description><![CDATA[
<p>I think with arch wiki it is even more than that. Before I switched to arch back then, you would consult the arch wiki for an unrelated distro, because it was (is) that good. Even the aur repository helps you alot, by checking the raw scripts, how to compile stuff. I can't make a good example but it feeled like reading vi specific wiki that helped you with plugin development for emacs.</p>
]]></description><pubDate>Sun, 15 Feb 2026 07:38:21 +0000</pubDate><link>https://news.ycombinator.com/item?id=47021783</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=47021783</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47021783</guid></item><item><title><![CDATA[New comment by razemio in "Pocketbase – open-source realtime back end in 1 file"]]></title><description><![CDATA[
<p>No, simple answer. Try it, link to the repo and I will tell you all the placed it failed. Vibe coding has its places. For something like pocketbase it most definitely is not ready yet.</p>
]]></description><pubDate>Fri, 28 Nov 2025 06:47:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=46076169</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=46076169</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46076169</guid></item><item><title><![CDATA[New comment by razemio in "German government comes out against Chat Control"]]></title><description><![CDATA[
<p>AfD has f'd up views on many things. I wouldn't even be surprised if they would vote for it, as soon as they are in a position of power. However in its current state, they would hurt themselfs with a chat control law. So yes, your a right. I would also not expect them to vote for it.</p>
]]></description><pubDate>Wed, 08 Oct 2025 04:05:27 +0000</pubDate><link>https://news.ycombinator.com/item?id=45511946</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=45511946</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45511946</guid></item><item><title><![CDATA[New comment by razemio in "Tesla said it didn't have key data in a fatal crash, then a hacker found it"]]></title><description><![CDATA[
<p>How would you name a system which can do everything a Level 5 system can, but with Level 2/3 supervision? A name, which a PR team would choose without the missleading stuff as you are saying.</p>
]]></description><pubDate>Fri, 29 Aug 2025 13:38:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=45063985</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=45063985</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45063985</guid></item><item><title><![CDATA[New comment by razemio in "Tesla said it didn't have key data in a fatal crash, then a hacker found it"]]></title><description><![CDATA[
<p>That's something different. The problem with the level is, that it only focuses on the attention the human driver needs to give to the automation. In this sense my Kia EV6 is also Level 2/3, same as FSD. However FSD can do so much more than my Kia EV6. That's a fact. Still the same level. Where did Tesla say FSD is SAE Level 5 approved? They would be responsible everytime FSD is active during a crash. Tesla is full self driving with Level 2/3 supervision and in my opinion this is not missleading.<p>Also "All this demonstrates is the term “full self driving” is meaningless." prooves my point that it is not missleading.</p>
]]></description><pubDate>Fri, 29 Aug 2025 13:16:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=45063723</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=45063723</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45063723</guid></item><item><title><![CDATA[New comment by razemio in "Tesla said it didn't have key data in a fatal crash, then a hacker found it"]]></title><description><![CDATA[
<p>Everytime this comes up, I am on the opposite site of this. It is clearly full self driving. It can stop at red lights, cross intersections, make turns, park, drive, change lanes, break and navigate on its own. There are various videos online where FSD managed to drive a route start to finish without a single human override. That's full self driving. It can also crash like humans "can" and that why it needs supervision. In this sense, we as humans are also "full self driving" with a much (?) lower risk of crashing.<p>Like also everytime let the downvotes rain. If you downvote, it would be nice, if you could tell me where I am wrong. It might change my view on things.</p>
]]></description><pubDate>Fri, 29 Aug 2025 13:03:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=45063570</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=45063570</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45063570</guid></item><item><title><![CDATA[New comment by razemio in "Turning Claude Code into my best design partner"]]></title><description><![CDATA[
<p>That is exactly my issue. I am more districted while being more productive. It feels just wrong, but works for now. In the long run, I need to find a solution for this. What works best for now, is to let multiple agents run on multiple repos of the same project solving different tasks. This way, I stay somewhat focused, since I constantly need to approve things. Just like a Projekt Manager with a big team... Indeed curious times.</p>
]]></description><pubDate>Sun, 24 Aug 2025 09:17:36 +0000</pubDate><link>https://news.ycombinator.com/item?id=45002713</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=45002713</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45002713</guid></item><item><title><![CDATA[New comment by razemio in "AnduinOS"]]></title><description><![CDATA[
<p>Yes, this makes sense, since most distros use a "standard release" system. The intention behind it is to keep your OS stable and only apply security patches at some point. 
If you want to have a system with always the newest software, you need to use a "rolling release " system. I think Arch Linux is the most popular (arch user btw). This is much more fun, if you know what you are doing. Otherwise you will end up with a broken system pretty fast. Ofc you can fix it, but depending on experience and skill something like Ubuntu is the better choice.</p>
]]></description><pubDate>Wed, 20 Aug 2025 05:44:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=44958966</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=44958966</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44958966</guid></item><item><title><![CDATA[New comment by razemio in "OpenMower – An open source lawn mower"]]></title><description><![CDATA[
<p>I have a Mammotion Yuba and trust me, the grass looks awesome as a grid or in lines. It can even do logos. So far nicer looking grass and much faster then random.</p>
]]></description><pubDate>Tue, 19 Aug 2025 14:08:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=44951737</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=44951737</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44951737</guid></item><item><title><![CDATA[New comment by razemio in "Open hardware desktop 3D printing is dead?"]]></title><description><![CDATA[
<p>A genuine Question. Is open hardware even possible at some point? The advances in quality and speed are nothing short but impressive. I started 3d printing stuff in my basement one year ago (Ender V3 Plus). With the quality and speed improvements, comes technology which gets more complex every year. Companies spend millions to archive this. Why would they share it? I remember building drones in my basement (still on my wall) with open source software on the flight controllers. Now I can get a drone from DJI for less money with more features, in a smaller from factor, longer flight time, pre build and under 249g. Ofc this comes at the cost of repairability, control and trust. However I can still buy the hardware I used years ago. If I wanted to, I can build a drone by myself. I guess the same will happen to 3d printers.</p>
]]></description><pubDate>Fri, 15 Aug 2025 12:52:45 +0000</pubDate><link>https://news.ycombinator.com/item?id=44911770</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=44911770</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44911770</guid></item><item><title><![CDATA[New comment by razemio in "Claude Sonnet 4 now supports 1M tokens of context"]]></title><description><![CDATA[
<p>I think it is something else. If you think about it, humans often write about correcting errors done by others. Refactoring code, fixing bugs and write code more efficient. I guess it triggers other paths in the model, if we write that someone else did it. It is not about pleasing but our constant desire to improve things.</p>
]]></description><pubDate>Wed, 13 Aug 2025 06:13:19 +0000</pubDate><link>https://news.ycombinator.com/item?id=44885133</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=44885133</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44885133</guid></item><item><title><![CDATA[New comment by razemio in "The rise of AI as a threat to the S&P 500 [pdf]"]]></title><description><![CDATA[
<p>How would you ban AI? It is unstoppable now.</p>
]]></description><pubDate>Thu, 17 Jul 2025 15:42:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=44594572</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=44594572</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44594572</guid></item><item><title><![CDATA[New comment by razemio in "Are we the baddies?"]]></title><description><![CDATA[
<p>Can it, if the technology is there todo exactly that? Capitalism (for me) means doing the maximum allowed in a legal framework to maximize profit.</p>
]]></description><pubDate>Sun, 06 Jul 2025 09:43:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=44479226</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=44479226</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44479226</guid></item><item><title><![CDATA[New comment by razemio in "Sleeping beauty Bitcoin wallets wake up after 14 years to the tune of $2B"]]></title><description><![CDATA[
<p>I do not think that Elon would not claim he is the inventor by now. 
The team theory makes this entirely unbelievable. Something like this can only be pulled of by 1-2 person's whith exceptional self-restraint.</p>
]]></description><pubDate>Sat, 05 Jul 2025 05:10:57 +0000</pubDate><link>https://news.ycombinator.com/item?id=44470221</link><dc:creator>razemio</dc:creator><comments>https://news.ycombinator.com/item?id=44470221</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=44470221</guid></item></channel></rss>