<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: vinaigrette</title><link>https://news.ycombinator.com/user?id=vinaigrette</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 02 Jul 2026 04:41:58 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=vinaigrette" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by vinaigrette in "Launch HN: Parsewise (YC P25) – Reason Across Documents with an API"]]></title><description><![CDATA[
<p>I'm surprised at the low rate every model manages considering the (apparent) ease of the benchmarked document. Can your pipeline produce ground truth as a byproduct ? How do you think open-weight ocr models compare to the one showcased ? I've had good results with glm-ocr on complex documents (complex by their handwriting, pretty easy layouts).<p>What I like about your solution is the traceability of the information. A scruffy pipeline I used was gemini-flash 3.0 to pdf to notebook-lm (really amateurish work i know), but it yielded tremendeous time gains to extract info from documents (that could be borderline impossible to read for me). However, to trace back the info was obviously very tedious. But from my experience, notebooklm can now manage ocr/htr without a third party. I wonder how competitive your solution might be compared to messy workflows that work -- albeit with efforts -- but let's the researcher be "in contact" with the material.<p>What I really want is obviously an easy to setup local rag system, with the (very) light model that goes with it ... sweet dream.</p>
]]></description><pubDate>Wed, 01 Jul 2026 17:45:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=48750605</link><dc:creator>vinaigrette</dc:creator><comments>https://news.ycombinator.com/item?id=48750605</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48750605</guid></item><item><title><![CDATA[New comment by vinaigrette in "Launch HN: Parsewise (YC P25) – Reason Across Documents with an API"]]></title><description><![CDATA[
<p>This looks great for digital humanities, specifically archival work. Would love to try it.</p>
]]></description><pubDate>Wed, 01 Jul 2026 17:21:27 +0000</pubDate><link>https://news.ycombinator.com/item?id=48750251</link><dc:creator>vinaigrette</dc:creator><comments>https://news.ycombinator.com/item?id=48750251</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48750251</guid></item><item><title><![CDATA[New comment by vinaigrette in "Peopleless economy? Not technically impossible"]]></title><description><![CDATA[
<p>Some of the points you are making are great. I would however argue that a small group of people trading is still an economy albeit a shrunk one. But what I believe the point the author is making is the following : consumption based economy isn't required to an economy.
In some aspect, I believe that it is false : we are *required* to have a bunch of strictly unnecessary stuff to survival (cellphones and data plans being one example) to be considered as a functionnal human being (pretty much everywhere in the planet at that). I also believe that he points out something that is true however : most people have been unable to partake in consumption society (see Lewis' 1954 article on unlimited supply of labor) for a long time. This, and chronic unemployement that appeared in the 70's, showed that a 2 speed economy is possible, even preferable for the capitalists ; and what was limited to the global south at Lewis' time (though some more recent historical work tend to disagree) is happening in the North. My conclusion would be that the acceleration of inequalities to the point of "uneconomicness" of a large part of the population is what the author points out to be a "peopleless economy".
I personnaly found the article to be stimulating, even with its many shortcuts. What I make of your points are less a contradiction but a optimistic point of view of the situation : secession of the capitalists mean freedom for the rest. But empirical observation show that ressources being limited (although standalone freedom being unacceptable as well for them), auto-organization is confronted with the armed forces mentionned in the article.<p>Also, enterprises trading capital endlessly isn't exactly what is happening with the ai (alledged) bubble ?</p>
]]></description><pubDate>Tue, 16 Jun 2026 16:18:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=48557633</link><dc:creator>vinaigrette</dc:creator><comments>https://news.ycombinator.com/item?id=48557633</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48557633</guid></item><item><title><![CDATA[New comment by vinaigrette in "The Importance of Being Idle"]]></title><description><![CDATA[
<p>Poverty levels are roughly the same between Vietnam and the US from a quick search. Mean standard of living is a poor way to calculate inequality. If you have a link to a median one it would help to compare.</p>
]]></description><pubDate>Thu, 09 Apr 2026 08:45:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=47700892</link><dc:creator>vinaigrette</dc:creator><comments>https://news.ycombinator.com/item?id=47700892</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47700892</guid></item><item><title><![CDATA[New comment by vinaigrette in "Your File System Is Already A Graph Database"]]></title><description><![CDATA[
<p>Isn't text a basic linear structure that can cover sufficiently complex topics ?</p>
]]></description><pubDate>Wed, 08 Apr 2026 16:14:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=47692247</link><dc:creator>vinaigrette</dc:creator><comments>https://news.ycombinator.com/item?id=47692247</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47692247</guid></item></channel></rss>