<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: octacat</title><link>https://news.ycombinator.com/user?id=octacat</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Sat, 23 May 2026 08:14:04 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=octacat" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by octacat in "We asked camera companies why their RAW formats are all different and confusing"]]></title><description><![CDATA[
<p>By the way (hello, adobe).
DNG compression compresses worse than 7zip, haha.
Too much for an open standard, too hard to add LZ4 as an option in the times of having 24 core setups :D</p>
]]></description><pubDate>Mon, 07 Apr 2025 10:59:26 +0000</pubDate><link>https://news.ycombinator.com/item?id=43609914</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43609914</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43609914</guid></item><item><title><![CDATA[New comment by octacat in "We asked camera companies why their RAW formats are all different and confusing"]]></title><description><![CDATA[
<p>"Sony’s software for processing ARW RAW files is called Imaging Edge. Like most first-party software from camera manufacturers, it’s terrible and unintuitive to use — and should be saved for situations like a high-resolution multishot mode where it’s the only method to use a camera’s proprietary feature."<p>I think the primarily reason is that they have great hardware developers and terrible software developers. So, having ARWs it is maximum they could provide to photographer, so they could take the files and run away from sony as soon as possible (i.e. do the rest in the better software).<p>Pentax could save DNGs, there are zero reasons for other companies not to do the same.</p>
]]></description><pubDate>Mon, 07 Apr 2025 10:46:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=43609843</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43609843</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43609843</guid></item><item><title><![CDATA[New comment by octacat in "We asked camera companies why their RAW formats are all different and confusing"]]></title><description><![CDATA[
<p>It is always written into a memory buffer first, which could be like 256 megabytes... it tooks time to fill it up, once it is filled, memory card speed becomes a bottleneck. So, actually, writing only jpegs would trigger the slowdown later, so you could take more frames before the buffer fills up</p>
]]></description><pubDate>Mon, 07 Apr 2025 10:34:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=43609787</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43609787</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43609787</guid></item><item><title><![CDATA[New comment by octacat in "Dijkstra On the foolishness of "natural language programming""]]></title><description><![CDATA[
<p>They do way less then that. They just form a final list of locks and download that at the build time. Of course you have to also "recursively" go though all your dep tree and add submodules for each of subdependencies (recommend to add them in the main repo). Then you will have do waste infinite amount of time setting include dirs or something. If you have two libs that require a specific version of a shared lib, no dep manager would help you. Using submodules is questionable practice though. Useful for simple stuff, like 10 deps in total in the final project.</p>
]]></description><pubDate>Thu, 03 Apr 2025 09:34:57 +0000</pubDate><link>https://news.ycombinator.com/item?id=43567141</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43567141</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43567141</guid></item><item><title><![CDATA[New comment by octacat in "Dijkstra On the foolishness of "natural language programming""]]></title><description><![CDATA[
<p>Most popular dependency management systems literally linking to a git sha commit (tag), see locks file that npm/rebar/other tool gives you. Just in a recursive way.</p>
]]></description><pubDate>Thu, 03 Apr 2025 07:20:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=43566001</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43566001</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43566001</guid></item><item><title><![CDATA[New comment by octacat in "Dijkstra On the foolishness of "natural language programming""]]></title><description><![CDATA[
<p>Natural language is pretty good for describing the technical requirements for the complex system, though. I.e. not the current code implementation, but why the current code implementation is selected vs other possible implementations. Not what code do, but what it is expected to do. Basically, most of the missing parts, that live in Jira-s, instead of your repo. It is also good, at allowing better refactoring capabilities, when all your system is described by outside rules, which could be enforced on the whole codebase. We just use programming languages, because it is easier to use in automated/computer context (and was the only way to use, to be honest, before all the LLM stuff). Though, while it gives us non-ambiguity on the local scale, it stops working on the global scale, the first moment person went and copy-pasted part of the code. Are you sure that part follows all the high-level restrictions we should to follow and is correct program? It is program that would run, when compile, but definition of run is pretty loose. In C++ program that corrupts all the memory is also runnable.</p>
]]></description><pubDate>Thu, 03 Apr 2025 07:04:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=43565870</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43565870</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43565870</guid></item><item><title><![CDATA[New comment by octacat in "US Administration announces 34% tariffs on China, 20% on EU"]]></title><description><![CDATA[
<p>It's called US companies now could increase their prices by 30% and just don't worry much, if sales are pretty good already for them.</p>
]]></description><pubDate>Thu, 03 Apr 2025 06:50:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=43565724</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43565724</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43565724</guid></item><item><title><![CDATA[New comment by octacat in "France fines Apple €150M for “excessive” pop-ups that let users reject tracking"]]></title><description><![CDATA[
<p>Apple has its own trackers and ad services. But it treats itself specially. While other apps need to ask. Maybe I don't want some apps from apple to have the same permissions, as other apps from apple. It is basically bundling. I agreed for one app from apple to share data, not for all apps from apple. Oh, and don't let me start about wording they use on these pop-ups... Like when I download some app from the internet, which is signed, notarized and so on, it still asks: oh, you downloaded this app from the internet (surprise!)... Instead of: "ok, you downloaded from internet, it has sign, notarification, developer is already delivering software for 10 years and have 20 million users, so looks safe to run...". Basically, fear inducing messages instead of actually useful from security sense.</p>
]]></description><pubDate>Wed, 02 Apr 2025 23:01:05 +0000</pubDate><link>https://news.ycombinator.com/item?id=43562780</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43562780</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43562780</guid></item><item><title><![CDATA[New comment by octacat in "Everyone knows all the apps on your phone"]]></title><description><![CDATA[
<p>+ working notifications
- adblocker is more of a minus for publishers though<p>But mainly don't expect any good web app integration on mobile, because it would hit the store 30% tax.</p>
]]></description><pubDate>Mon, 31 Mar 2025 01:40:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=43529798</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43529798</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43529798</guid></item><item><title><![CDATA[New comment by octacat in "Upcoming Windows 11 builds cannot install without internet and Microsoft Account"]]></title><description><![CDATA[
<p>Would not fly.
Developers would not go into the store. Because everyone knows, that 30% tax would be next (basically it is the current situation on the Mac, apple could pull the switch on gatekeeper any time). And, because a lot of modern apps are just electron wrappers, people would just move to the web versions for everything. Which means killing their own platform.</p>
]]></description><pubDate>Mon, 31 Mar 2025 01:09:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=43529599</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43529599</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43529599</guid></item><item><title><![CDATA[New comment by octacat in "Upcoming Windows 11 builds cannot install without internet and Microsoft Account"]]></title><description><![CDATA[
<p>The moment they roll it as mandatory - people would stop updating.</p>
]]></description><pubDate>Mon, 31 Mar 2025 00:53:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=43529466</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43529466</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43529466</guid></item><item><title><![CDATA[New comment by octacat in "Upcoming Windows 11 builds cannot install without internet and Microsoft Account"]]></title><description><![CDATA[
<p>Similar why gatekeeper exists on Mac.
This would mean that old software would stop to work, though. Which is a bit anti-microsoft politics. Without running all the old unsigned software Win is not Win anymore.</p>
]]></description><pubDate>Mon, 31 Mar 2025 00:49:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=43529439</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43529439</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43529439</guid></item><item><title><![CDATA[New comment by octacat in "I genuinely don't understand why some people are still bullish about LLMs"]]></title><description><![CDATA[
<p>Just make a benchmark which would measure quality of included references ;)</p>
]]></description><pubDate>Mon, 31 Mar 2025 00:38:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=43529357</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43529357</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43529357</guid></item><item><title><![CDATA[New comment by octacat in "Konva.js - Declarative 2D Canvas for React, Vue, and Svelte"]]></title><description><![CDATA[
<p>By hand or regl library (almost by hand). I tried THREE, but it feels it gives me more pain than benefits (and you still end up tweaking their shaders sometimes). AI is pretty good at small tasks like tweaking shaders. I needed to draw a spectroscope, like a line connecting all pixels on the image one by one on the colorspace projection. Ended up 100 times more efficient in webgl comparing to browser API. But it is pretty low level. On another hand once you dig deeper into shaders, they are actually more powerful. Let's just say, API is just painful too :). Also, there is some fun with how transparency works and order of elements displayed (because of parallel processing).  Oh, ideally I want web canvas API, but which allows to pass a buffer of lines to draw instead of one ;)</p>
]]></description><pubDate>Mon, 24 Mar 2025 15:33:43 +0000</pubDate><link>https://news.ycombinator.com/item?id=43462227</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43462227</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43462227</guid></item><item><title><![CDATA[New comment by octacat in "Amazon wants a product safety regulator declared unconstitutional"]]></title><description><![CDATA[
<p>If it is not a store, just don't buy from them, problem solved.</p>
]]></description><pubDate>Mon, 24 Mar 2025 12:18:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=43460117</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43460117</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43460117</guid></item><item><title><![CDATA[New comment by octacat in "Tencent's 'Hunyuan-T1'–The First Mamba-Powered Ultra-Large Model"]]></title><description><![CDATA[
<p>older docs are forever there. what it needs is more training data with new APIs.
Actually, because older docs are there, you can ask to update some old code to newer versions automatically.<p>Point is that it needs enough examples with a newer version.
Also, reasoning models are pretty good at spotting which version they are using.<p>(tested not with tailwind, but some other JS libs).</p>
]]></description><pubDate>Mon, 24 Mar 2025 11:40:36 +0000</pubDate><link>https://news.ycombinator.com/item?id=43459780</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43459780</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43459780</guid></item><item><title><![CDATA[New comment by octacat in "Konva.js - Declarative 2D Canvas for React, Vue, and Svelte"]]></title><description><![CDATA[
<p>true, also declarative approach is amazing for wegbl, you need to draw many lines - pass start/end points in a buffer at once.</p>
]]></description><pubDate>Thu, 20 Mar 2025 11:23:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=43421756</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43421756</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43421756</guid></item><item><title><![CDATA[New comment by octacat in "Nvidia's RTX Pro 6000 has 96GB of VRAM and 600W of power"]]></title><description><![CDATA[
<p>Wiring also, chips for ddr7 look like they have a lot of pads.
And you need pads on the graphics chip for them too.</p>
]]></description><pubDate>Wed, 19 Mar 2025 09:26:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=43409798</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43409798</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43409798</guid></item><item><title><![CDATA[New comment by octacat in "Speedrunners are vulnerability researchers, they just don't know it yet"]]></title><description><![CDATA[
<p>they pretty much know it.
Also, many speedrunners just run, while research is done and "published" by people before, pretty much using all the hacking technics.</p>
]]></description><pubDate>Tue, 04 Mar 2025 07:11:05 +0000</pubDate><link>https://news.ycombinator.com/item?id=43251321</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43251321</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43251321</guid></item><item><title><![CDATA[New comment by octacat in "California homeowners to fund half of high-risk insurer's $1B 'bailout'"]]></title><description><![CDATA[
<p>A lot of stuff is simple, like cleaning brushes around the house, replacing wooden fences with metal/stone. More harder - changing ventilation for the attics. Except it is still not popular, so nobody would do it, if not mandated.</p>
]]></description><pubDate>Sat, 15 Feb 2025 22:34:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=43063130</link><dc:creator>octacat</dc:creator><comments>https://news.ycombinator.com/item?id=43063130</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43063130</guid></item></channel></rss>