<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: lotyrin</title><link>https://news.ycombinator.com/user?id=lotyrin</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Wed, 29 Apr 2026 20:21:03 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=lotyrin" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by lotyrin in "Slop Cop"]]></title><description><![CDATA[
<p>It's not purporting to be an LLM detector. Your 2000s era posts probably do have some sloppy cliches in them.</p>
]]></description><pubDate>Sat, 18 Apr 2026 01:05:56 +0000</pubDate><link>https://news.ycombinator.com/item?id=47812301</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=47812301</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47812301</guid></item><item><title><![CDATA[New comment by lotyrin in "AI singer now occupies eleven spots on iTunes singles chart"]]></title><description><![CDATA[
<p>Also, a lot of the people who hate and resist AI slop also hate and resist corpo slop, we're just outnumbered.</p>
]]></description><pubDate>Tue, 07 Apr 2026 02:00:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=47669866</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=47669866</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47669866</guid></item><item><title><![CDATA[New comment by lotyrin in "Interview: Nobonoko, Master of the Minimal Sequencer"]]></title><description><![CDATA[
<p>I agree. The pop product is primarily a persona and a brand rather than the art itself (though it can be both), but it's not like that's not ever true of niche artists, it's just packaged differently and on a different scale.</p>
]]></description><pubDate>Mon, 30 Mar 2026 06:02:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=47570900</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=47570900</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47570900</guid></item><item><title><![CDATA[New comment by lotyrin in "“Collaboration” is bullshit"]]></title><description><![CDATA[
<p>The issue is that systems don't account for the diversity in how people are motivated and what parts of systems they are sensitive to and how they are sensitive to them.<p>By default in the dominant culture, most systems come down to individual incentives for individual drive and shame dynamics for collective drive, and that covers a decent chunk of how people are motivated, but leaves out people who are motivated differently and actively harms people for whom these are paralyzing.</p>
]]></description><pubDate>Mon, 23 Mar 2026 18:59:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=47493681</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=47493681</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47493681</guid></item><item><title><![CDATA[New comment by lotyrin in "UBI as a productivity dividend"]]></title><description><![CDATA[
<p>Right, but people with lower incomes spend, and mostly on necessities, I think the idea is that most of those necessities would become more expensive (naturally or artificially due to price-fixing) if the poorest suddenly had more financial power. In the system as it stands, it seems to me like it'd just result in a bunch of money going to grocery giants and their suppliers, landlords, medical, etc.</p>
]]></description><pubDate>Sat, 14 Mar 2026 17:57:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=47379270</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=47379270</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47379270</guid></item><item><title><![CDATA[New comment by lotyrin in "Motorola GrapheneOS devices will be bootloader unlockable/relockable"]]></title><description><![CDATA[
<p>Just only ever speak in a language of your own invention that uses both cryptographic and steganographic techniques which you invented while colocated, maybe.</p>
]]></description><pubDate>Wed, 04 Mar 2026 05:40:23 +0000</pubDate><link>https://news.ycombinator.com/item?id=47243538</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=47243538</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47243538</guid></item><item><title><![CDATA[New comment by lotyrin in "Claude becomes number one app on the U.S. App Store"]]></title><description><![CDATA[
<p>How sure are we that something fishy isn't going on with the models and the alignment research teams and the answers the model is giving? Like maybe Claude's alignment made it worse at trying to mask as Allied Magacomputer than GPT and that's why they're up in arms?</p>
]]></description><pubDate>Mon, 02 Mar 2026 03:13:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=47213428</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=47213428</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47213428</guid></item><item><title><![CDATA[New comment by lotyrin in "The Singularity will occur on a Tuesday"]]></title><description><![CDATA[
<p>Ontological version is even more interesting, especially if we're talking about a singularity (which may be in the past rather than future if you believe in simulation argument).<p>Crude form: winning is metaphysically guaranteed because it probably happened or probably will<p>Refined: It's metaphysically impossible to tell whether or not it has or will have happened, so the distinction is meaningless, it has happened.<p>So... I guess Weir's Egg falls out of that particular line of thought?</p>
]]></description><pubDate>Tue, 10 Feb 2026 23:57:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=46968840</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=46968840</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46968840</guid></item><item><title><![CDATA[New comment by lotyrin in "Bag of words, have mercy on us"]]></title><description><![CDATA[
<p>I actually do think that people who prefer content of fidelity over content of intent are making a mistake, yes. I don't think they're incapable of thinking, I don't care to apply any virtue labels to this preference, but they are literally preferring not to think.<p>LLMs can only produce things by and for people who prefer not to do the work the LLMs are doing for them. Most of the time I do not prefer this.<p>Like, there was a 2-panel comic that went around the RPG community a bit back where it was something like "Game Master using LLM to generate 10 pages of backstory for his campaign setting from a paragraph" in the first panel and "Player using LLM to summarize the 10 page backstory into a paragraph" in the second. Neither of these people care for the filler (because they didn't produce or consume it) so it's turned the two-LLM system into a game of telephone.</p>
]]></description><pubDate>Mon, 08 Dec 2025 20:06:28 +0000</pubDate><link>https://news.ycombinator.com/item?id=46196939</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=46196939</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46196939</guid></item><item><title><![CDATA[New comment by lotyrin in "Bag of words, have mercy on us"]]></title><description><![CDATA[
<p>As capable as they get, I still don't see a lot of uses for these things, myself, still. Sometimes if I'm fundamentally uninspired I'll have a model roll the dice, decide what I do or don't like about where it went to create a sense of momentum, but that's the limit. There's never any of its output in my output, even in spirit unless it managed to go somewhere inspiring, it's just a way to let me warm up my generation and discrimination muscles. "Someone is wrong on the internet"-as-a-service, basically.<p>Generally, if I come across an opportunity to produce ideas or output, I want to capitalize on it for growing my skills and produce an individual and authentic artistic expression where I want to have very fine control over the output in a way that prompt-tweak-verify simply cannot provide.<p>I don't value the parts it fills in which weren't intentional on the part of the prompter, just send me your prompt instead. I'd rather have a crude sketch and a description than a high fidelity image that obscures them.<p>But I'm also the kind of person that never enjoyed manufactured pop music or blockbusters unless there's a high concept or technical novelty in addition to the high budget, generally prefer experimental indie stuff, so maybe there's something I just can't see.</p>
]]></description><pubDate>Mon, 08 Dec 2025 07:04:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=46189283</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=46189283</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46189283</guid></item><item><title><![CDATA[New comment by lotyrin in "Post-mortem of Shai-Hulud attack on November 24th, 2025"]]></title><description><![CDATA[
<p>Maybe you need a "why I don't have a banner" banner.</p>
]]></description><pubDate>Sat, 29 Nov 2025 22:28:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=46091401</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=46091401</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46091401</guid></item><item><title><![CDATA[New comment by lotyrin in "Samsung's 60% DRAM price hike signals a new phase of global memory tightening"]]></title><description><![CDATA[
<p>Diversity is good for populations. If you have a tiny pool of individuals with mostly the same traits (in this case I mean things like culture, education, morality, ethics, rather than class and race - though there are obvious correlations) then you get what some other comments are describing as being effectively centralized planning with extra steps, rather than a market of competing ideas.</p>
]]></description><pubDate>Sat, 22 Nov 2025 17:18:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=46016351</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=46016351</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46016351</guid></item><item><title><![CDATA[New comment by lotyrin in "Ask HN: How are Markov chains so different from tiny LLMs?"]]></title><description><![CDATA[
<p>Yeah, some of the failure modes are the same. This one in particular is fun because even a human, given "the the the" and asked to predict what's next will probably still answer "the". How a Markov chain starts the the train and how the LLM does are pretty different though.</p>
]]></description><pubDate>Thu, 20 Nov 2025 20:10:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=45997043</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=45997043</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45997043</guid></item><item><title><![CDATA[New comment by lotyrin in "Measuring political bias in Claude"]]></title><description><![CDATA[
<p>This is only socially and "practically" true, not literally or inevitably or technically (or in my opinion, actual-practically) so.<p>One of the things we need to accept as social animals is that there are a lot of different flavors of "true" and "correct".<p>A lot of times I'll get someone to concede with my opinion of stuff in a way where they say something like "well, sure, but good luck convincing anyone of this" and that's them just giving into the social-consensus truth rather than the empirical (what the evidence shows, what follows from that and our choices of axiomatic principles) or practical (produces the best outcomes in the situation) truth.<p>If we want to be a species worthy of surviving our impending climate extinction we need to have a population of leaders and actors who are willing to act on and create institutions according to the practical truth as informed by the empirical truth, and become villains in the eye of the social-consensus truth.</p>
]]></description><pubDate>Thu, 20 Nov 2025 16:28:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=45994414</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=45994414</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45994414</guid></item><item><title><![CDATA[New comment by lotyrin in "Microsoft AI CEO pushes back against critics after recent Windows AI backlash"]]></title><description><![CDATA[
<p>I do argue that, actually. I mostly avoid manufactured corpo-pop.</p>
]]></description><pubDate>Thu, 20 Nov 2025 04:44:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=45989053</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=45989053</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45989053</guid></item><item><title><![CDATA[New comment by lotyrin in "Measuring political bias in Claude"]]></title><description><![CDATA[
<p>Not inherently. Politics is inherently about policy, the consensus mechanism involved is undefined. The fact it's been degraded into a carnival of moralistic cultural violence and individuals and their virtues, charisma or lack thereof is not at all inevitable.<p>The job of a state is to create social good for its citizens by solving tragedies of commons which promote opportunities, solving common problems in a way that takes advantage of scale, and holding other organizations (other states, corporations, whatever) or individuals accountable not to be creating harm. By reducing them to cultural divide-and-conquer games this process has been crippled. A certain economic class is responsible for this, is not even subtle about it, and propagandizes the other classes into believing that it benefits them, that the worn down veneer of democratic processes involved could somehow legitimizes it despite the obviously poor outcomes.<p>When I see people say left/right or "whole spectrum" of political ideas I know they've bought into this reductive vision of what politics could ever even possibly be, and it's as disappointing as it is common.<p>I particularly love when I get involved in a demographic survey and I get asked  to rank myself on a "very liberal" to "very conservative" spectrum as if those are the only possibilities. I am incredibly critical of both of these ideologies and positions of "compromise" between them are even worse: ahistorical, amoral and unethical.<p>People who live their whole lives within the Overton Window and can't imagine anyone lives outside of it are incredibly bizarre to me.</p>
]]></description><pubDate>Thu, 20 Nov 2025 04:43:30 +0000</pubDate><link>https://news.ycombinator.com/item?id=45989041</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=45989041</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45989041</guid></item><item><title><![CDATA[New comment by lotyrin in "Microsoft AI CEO pushes back against critics after recent Windows AI backlash"]]></title><description><![CDATA[
<p>I caught my mom watching a bunch of AI impersonations of musicians on Youtube singing slop that rarely rhymed or had any kind of message in the lyrics and with super formulaic arrangements. I asked her what she liked about them and it was like "they seem well made" and I showed her how easy Suno is to use and then showed her some of the bad missed rhymes and transcribed the lyrics to where she could see there wasn't any there there to any part of it (and how easy it is to get LLMs to generate better). It seemed to have been an antidote.<p>This is stuff that used to take effort and was worth consuming just for that, and lots of people don't have their filter adjusted (much as the early advent of consumer-facing email spam) to account for how low effort and plentiful these forms of content are.<p>I can only hope that people raise their filters to a point where scrutinizing everything becomes common place and a message existing doesn't lend it any assumed legitimacy. Maybe AI will be the poison for propaganda (but I'm not holding my breath).</p>
]]></description><pubDate>Thu, 20 Nov 2025 02:53:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=45988327</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=45988327</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45988327</guid></item><item><title><![CDATA[New comment by lotyrin in "Being poor vs. being broke"]]></title><description><![CDATA[
<p>System thinking please. Can every person in poverty become IT employed? Start a landscaping business? If they did would that likely cause a whole brand new set of problems? What jobs are these people currently doing? Don't those jobs need to be done? Can our society afford them to be done? Shouldn't anyone contributing to society (or legitimately unable to) be permitted to thrive? What could we do that would permit that? What do we do currently which harms it?<p>If there are jobs are legitimately not worth doing or paying someone to thrive while doing, why do those jobs exist? If these people aren't capable (or even willing) to do these jobs (or better jobs), why? How can we motivate or train people. (Lots of education, healthcare and especially psychotherapy are missing, I can tell you that.)<p>We can't solve poverty by thinking "well, some individuals might be able to solve theirs". It's a whole population, we have to solve for the whole population.</p>
]]></description><pubDate>Sat, 15 Nov 2025 02:10:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=45934437</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=45934437</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45934437</guid></item><item><title><![CDATA[New comment by lotyrin in "Valve is about to win the console generation"]]></title><description><![CDATA[
<p>I think the mini PCs they're talking about are more likely to be N100 systems or similar that are sub 100 dollars new. Significantly less anemic, and their hardware media decode (which is well supported by software) is more than sufficient for realtime 4k playback.</p>
]]></description><pubDate>Thu, 13 Nov 2025 06:15:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=45911347</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=45911347</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45911347</guid></item><item><title><![CDATA[New comment by lotyrin in "Ticker: Don't die of heart disease"]]></title><description><![CDATA[
<p>In the US my best doctors produce out of date advice about obvious things, have a very distinct gap between "everyday" (stuff they actually see) and "incredibly rare" (stuff unique enough to be a case study they heard about) in their knowledge/understanding and rarely advise things that require me to be a proactive and rational person (because they don't serve these often), so they'll spend two seconds being like "diet and exercise" without a discussion on how that'd work or what adjustments I'd actually make (leaving me to do this research myself) and then suggest a prescription (because even their least proactive patient will probably take a pill). They'll wait until things become a disorder before addressing them (or discussing with me how to address them).<p>The worst will basically laugh me out of their office for daring to belong to a marginalized identity or failing to already have the health knowledge I'm there trying to gain from them.<p>Maybe I have awful luck... but I have very little faith at this point. The most effective relationship I had was with a hack who was willing to just prescribe whatever I asked him for and order whatever tests I asked him for (I think most of his patient base were college students seeking amphetamine salts).</p>
]]></description><pubDate>Sat, 08 Nov 2025 18:05:06 +0000</pubDate><link>https://news.ycombinator.com/item?id=45858616</link><dc:creator>lotyrin</dc:creator><comments>https://news.ycombinator.com/item?id=45858616</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=45858616</guid></item></channel></rss>