<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: toaste_</title><link>https://news.ycombinator.com/user?id=toaste_</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Wed, 10 Jun 2026 06:52:03 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=toaste_" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by toaste_ in "Bun support is now limited and deprecated"]]></title><description><![CDATA[
<p>Reading and understanding code is more difficult than writing code.<p>It is significantly easier to modify code that you personally wrote, or code that you have read and understood to fix an issue in previously. This is why the maintainers of a project change slowly over time and it takes a long time for new ones to get up to speed.<p>All of Bun has been rewritten by a tool. In a different language that maintainers may not be fully proficient in.<p>Even though the rewrite was done well, and even if we assume it's functionally equivalent to the old Zig code, there will still be future issues. And ALl of the maintainers are essentially now new hires who have never seen that code in their lives.<p>It's not "politics" to have an ounce of sense to foresee problems in such a project as a dependency.</p>
]]></description><pubDate>Fri, 22 May 2026 23:25:28 +0000</pubDate><link>https://news.ycombinator.com/item?id=48242850</link><dc:creator>toaste_</dc:creator><comments>https://news.ycombinator.com/item?id=48242850</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48242850</guid></item><item><title><![CDATA[New comment by toaste_ in "Original GrapheneOS responses to WIRED fact checker"]]></title><description><![CDATA[
<p>When Louis Rossmann thinks your communication has a problem with going on rants, it must be pretty out there.</p>
]]></description><pubDate>Tue, 21 Apr 2026 16:33:57 +0000</pubDate><link>https://news.ycombinator.com/item?id=47851107</link><dc:creator>toaste_</dc:creator><comments>https://news.ycombinator.com/item?id=47851107</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47851107</guid></item><item><title><![CDATA[New comment by toaste_ in "Tom7: No one can force me to have a secure website [video]"]]></title><description><![CDATA[
<p>Tom appears to have totally missed SSLStrip.<p>Before browsers screamed bloody murder over http, a MITM could defeat SSL by acting as the SSL endpoint and forwarding everything as plain http. And back then, the only indication was lack of a 16px lock icon and a missing "s" in "https".<p>It's additionally daft to think that just because the page is public knowledge, a specific person reading the page is never sensitive information. As a blunt example, Wikipedia is obviously public knowledge. If you are a Chinese national reading <a href="https://en.wikipedia.org/wiki/1989_Tiananmen_Square_protests_and_massacre" rel="nofollow">https://en.wikipedia.org/wiki/1989_Tiananmen_Square_protests...</a> then the CCP might like to know your location.</p>
]]></description><pubDate>Tue, 14 Apr 2026 02:47:38 +0000</pubDate><link>https://news.ycombinator.com/item?id=47760657</link><dc:creator>toaste_</dc:creator><comments>https://news.ycombinator.com/item?id=47760657</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47760657</guid></item><item><title><![CDATA[New comment by toaste_ in "Case study: recovery of a corrupted 12 TB multi-device pool"]]></title><description><![CDATA[
<p>I mean, the only other option was bcachefs, which might have been funny if this LLM-generated blogpost were written by the OpenClaw instance the developer has decided is sentient:<p><a href="https://www.reddit.com/r/bcachefs/comments/1rblll1/the_blog_of_an_llm_saying_its_owned_by_kent_and/" rel="nofollow">https://www.reddit.com/r/bcachefs/comments/1rblll1/the_blog_...</a><p>But no. It was btrfs.<p>As a side note, it's somewhat impressive that an LLM agent was able to produce a suite of custom tools that were apparently successfully used to recover some data from a corrupted btrfs array, even ad-hoc.</p>
]]></description><pubDate>Mon, 06 Apr 2026 07:19:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=47657858</link><dc:creator>toaste_</dc:creator><comments>https://news.ycombinator.com/item?id=47657858</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47657858</guid></item><item><title><![CDATA[New comment by toaste_ in "The Inference Shift – How Cheap Chips Could Put Frontier AI in Everyone's Hands"]]></title><description><![CDATA[
<p>>Research and fact-checking assistance from Claude (Anthropic).<p>What paper? This is slop.<p>No, BitNet not requiring multiplication will not put a foundational model in your pocket. It would be nice for power if tinary models had scaled, but since it requires roughly 3x the parameters of a similarly capable model, the memory bandwidth does not scale down nearly as well.<p>The real trick is that a classic LLM is not useful in the scenarios the author proposes. The hypothetical livestock vet is far better served by her books and a phone call to a university ag extension to confer with colleagues than an LLM disconnected from the internet that will hallucinate nonsense.</p>
]]></description><pubDate>Tue, 31 Mar 2026 08:48:28 +0000</pubDate><link>https://news.ycombinator.com/item?id=47584451</link><dc:creator>toaste_</dc:creator><comments>https://news.ycombinator.com/item?id=47584451</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47584451</guid></item></channel></rss>