<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: juliusgeo</title><link>https://news.ycombinator.com/user?id=juliusgeo</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Fri, 24 Apr 2026 10:12:47 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=juliusgeo" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[Show HN: Fastest(?) SIMD CSV Parser in Rust]]></title><description><![CDATA[
<p>There are already a quite a few [0][1] CSV parsers that use SIMD, some in Rust, with a variety of approaches. I found simd-csv[1] to have a very interesting approach that leverages memchr to essentially "seek" for the next delimiter, reducing a lot of the overhead that a byte-by-byte CSV parser would have. However, as noted in the README, the creators of simd-csv explicitly chose not to use the classic pclmulqdq trick[2] that other libraries like simdjson use due to portability concerns. I set out to beat to simd-csv's implementation by building a parser more similar to Geoff Langdale's, using the pclmulqdq trick as well as optimized intrinsic usage for aarch64 platforms[3]. If anyone has feedback on the Rust code, or my usage of intrinsics, I would greatly appreciate it.<p>[0] <a href="https://github.com/geofflangdale/simdcsv" rel="nofollow">https://github.com/geofflangdale/simdcsv</a>
[1] <a href="https://github.com/medialab/simd-csv" rel="nofollow">https://github.com/medialab/simd-csv</a>
[2] <a href="https://branchfree.org/2019/03/06/code-fragment-finding-quote-pairs-with-carry-less-multiply-pclmulqdq/" rel="nofollow">https://branchfree.org/2019/03/06/code-fragment-finding-quot...</a>
[3] <a href="https://developer.arm.com/community/arm-community-blogs/b/servers-and-cloud-computing-blog/posts/porting-x86-vector-bitmask-optimizations-to-arm-neon" rel="nofollow">https://developer.arm.com/community/arm-community-blogs/b/se...</a></p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=47029020">https://news.ycombinator.com/item?id=47029020</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Sun, 15 Feb 2026 23:45:19 +0000</pubDate><link>https://github.com/juliusgeo/csimdv-rs</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=47029020</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=47029020</guid></item><item><title><![CDATA[New comment by juliusgeo in "Show HN: I built a CSV parser to try Go 1.26's new SIMD package"]]></title><description><![CDATA[
<p>This is super cool! If I'm understanding your implementation correctly, you do perform bit by bit state machine logic to check whether quotes should be escaped etc. You can do that in a single pass by using carry-less polynomial multiplication instructions (_mm_clmulepi64_si128 on AVX-512 I believe), or by just computing the carryless xor directly on the quote mask and then &ing the inverse with the bitmask for quotes. Simdjson uses this trick, and I use it as well in my Rust simd csv parser:<p><a href="https://github.com/juliusgeo/csimdv-rs/blob/681df3b036f30c5a08d75c8b482f73fb2aff2ab2/src/lib.rs#L141" rel="nofollow">https://github.com/juliusgeo/csimdv-rs/blob/681df3b036f30c5a...</a><p>This is a good write-up on how the approach works: <a href="https://nullprogram.com/blog/2021/12/04/" rel="nofollow">https://nullprogram.com/blog/2021/12/04/</a></p>
]]></description><pubDate>Wed, 28 Jan 2026 19:30:17 +0000</pubDate><link>https://news.ycombinator.com/item?id=46800372</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=46800372</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=46800372</guid></item><item><title><![CDATA[New comment by juliusgeo in "Ask HN: I'm an MIT senior and still unemployed – and so are most of my friends"]]></title><description><![CDATA[
<p>As Percy Shelley said about Ozymandias, "Look on my Works, ye Mighty, and despair!" We're all sand on a long enough time scale, and getting a TC 800k+ will not save you from that fate, in the same way that conquering an empire will not save you either. Other people have pointed out that you likely make a decent wage, but understandably that does not address your concern. If you feel that your only value is your accomplishments, then no matter how much you achieve it will never be enough. You get to decide your own goals, and whether you're reaching them. Don't let these proxies for ability and class decide what you want or how fast you should get it.</p>
]]></description><pubDate>Wed, 09 Apr 2025 13:27:09 +0000</pubDate><link>https://news.ycombinator.com/item?id=43631817</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=43631817</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43631817</guid></item><item><title><![CDATA[New comment by juliusgeo in "Smuggling Python Code Using Shrugging Faces"]]></title><description><![CDATA[
<p>> I don't like shrug! I don't want any shrug!<p>¯\_(ツ)_/¯<p>> If people don't equally well suspect `eval` then education attempts have seriously failed.<p>Haha yes, that statement was mostly made in jest. I would hope most people would be just as suspicious of an eval one liner with a singular emoji. Does make it slightly less suspicious than an equivalent eval which doesn't have 50 invisible bytes.<p>Good shout on fixing issues with eval in OSS--I think I might do that!</p>
]]></description><pubDate>Wed, 19 Mar 2025 15:48:52 +0000</pubDate><link>https://news.ycombinator.com/item?id=43413574</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=43413574</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43413574</guid></item><item><title><![CDATA[Smuggling Python Code Using Shrugging Faces]]></title><description><![CDATA[
<p>Article URL: <a href="https://gist.github.com/juliusgeo/bb62acdcc9f88cda852178e5361437f2">https://gist.github.com/juliusgeo/bb62acdcc9f88cda852178e5361437f2</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=43407260">https://news.ycombinator.com/item?id=43407260</a></p>
<p>Points: 3</p>
<p># Comments: 2</p>
]]></description><pubDate>Wed, 19 Mar 2025 01:08:17 +0000</pubDate><link>https://gist.github.com/juliusgeo/bb62acdcc9f88cda852178e5361437f2</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=43407260</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=43407260</guid></item><item><title><![CDATA[New comment by juliusgeo in "I Fired My Product Team and Replaced Them with AI"]]></title><description><![CDATA[
<p>Funnily enough, in the bash script itself .DS_STORE is the second in a massive list of hardcoded exclusions.</p>
]]></description><pubDate>Fri, 24 Jan 2025 16:41:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=42814843</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=42814843</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42814843</guid></item><item><title><![CDATA[New comment by juliusgeo in "TikTok goes dark in the US"]]></title><description><![CDATA[
<p>I'm curious about your third assertion--I am definitely not a fan of the US government's current laissez-faire approach to regulation, but I was able to find a few examples of the US levying larger fines:<p>British Petroleum (total settlement was 20b, but the Clean Water Act penalty was only 5.5b): <a href="https://www.justice.gov/opa/pr/us-and-five-gulf-states-reach-historic-settlement-bp-resolve-civil-lawsuit-over-deepwater" rel="nofollow">https://www.justice.gov/opa/pr/us-and-five-gulf-states-reach...</a><p>Wells Fargo: has had so many scandals, some of which were over 3b: <a href="https://www.justice.gov/opa/pr/wells-fargo-agrees-pay-3-billion-resolve-criminal-and-civil-investigations-sales-practices" rel="nofollow">https://www.justice.gov/opa/pr/wells-fargo-agrees-pay-3-bill...</a><p>In the article you linked, it mentions that this fine was 2.75b, or, 4% of Alibaba's 2019 revenue. I'm not knowledgeable enough in finance to state this as fact, but it looks like BP had a total revenue of 222B in 2015 [1]. 5.5b/222b = 2.47%. The total settlement would be 20b/222b = 9.01%<p>Now, obviously there are many examples of companies being fined paltry amounts for massive violations in the US, and I'm not sure how to reconcile 5.5b for destroying an entire ocean ecosystem vs roughly the same fine for anti-trust violations. But I don't think it's true that the US <i>never</i> enforces its laws against large and valuable companies. Do you know of any good sources that compare the history of corporate fines in China vs the US in more detail?<p>[1] <a href="https://www.bp.com/content/dam/bp/business-sites/en/global/corporate/pdfs/investors/bp-annual-report-and-form-20f-2015.pdf" rel="nofollow">https://www.bp.com/content/dam/bp/business-sites/en/global/c...</a></p>
]]></description><pubDate>Sun, 19 Jan 2025 16:53:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=42758719</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=42758719</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42758719</guid></item><item><title><![CDATA[New comment by juliusgeo in "Ask HN: How do you prevent the impact of social media on your children?"]]></title><description><![CDATA[
<p>I can understand your POV perhaps surprisingly well, as my father was secular growing up and then chose to join Protestantism in college (against his parents wishes). I wonder if at the end of the day it's just teenagers wanting to rebel. My dad's parents were secular, so he became Christian, and I became secular again. I can definitely relate to not enjoying support groups where the suffering is "valorized" to a certain extent. I think I was mostly reacting to the sentiment of superiority in general, but that is also an interesting case because it is pretty clear that families like that tend to have better outcomes overall (at least in monetary terms). My POV is that WASP culture in general breeds these perspectives, and also reinforces them because of the monetary and social inertia.</p>
]]></description><pubDate>Wed, 15 Jan 2025 01:09:24 +0000</pubDate><link>https://news.ycombinator.com/item?id=42706145</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=42706145</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42706145</guid></item><item><title><![CDATA[New comment by juliusgeo in "Ask HN: How do you prevent the impact of social media on your children?"]]></title><description><![CDATA[
<p>I am someone who was raised with a very similar set of values. I was homeschooled, and often believed that "public-schooled" kids had a worse, more limited set of values. I was not allowed to use computers till I was in 11th grade, and dove into reading as an alternative. Very little screen time, but I ended up with a lot of issues that did not even begin manifesting until I was an adult. I would urge you to re-examine your beliefs around this topic. It is too easy to elide the issues by reframing them as "a bit of arrogance". Based on my own experience, listening to the people around me, they are not experiencing it as "a bit" of arrogance. It is too easy, almost intoxicatingly so, to believe that you are better than those struggling. As long as you frame your own struggles as unique, you will deprive yourself of both 1) commiseration and 2) knowledge on how to progress past. Rather than say "everybody who sought solutions to their issues had issues", ask the question "how many people that did have issues did not seek solutions".</p>
]]></description><pubDate>Wed, 15 Jan 2025 00:45:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=42705993</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=42705993</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42705993</guid></item><item><title><![CDATA[New comment by juliusgeo in "Before Squid Game, there was Battle Royale"]]></title><description><![CDATA[
<p>The outrage trilogy is amazing, I would also recommend Sonatine and Hana-bi.</p>
]]></description><pubDate>Sat, 11 Jan 2025 16:53:24 +0000</pubDate><link>https://news.ycombinator.com/item?id=42667109</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=42667109</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42667109</guid></item><item><title><![CDATA[New comment by juliusgeo in "Self driving 1993 Volvo with open pilot"]]></title><description><![CDATA[
<p>There are many young people (myself included, though not nearly as skilled as most) that are modifying their cars in pretty substantial ways. Just in my own friend group there are multiple under 30 with their primary car being a heavily modified sports car. I myself own 2 relatively old sports cars (90s Acura, and 00's Audi), both obviously in manual, both of which have had significant work done in my driveway. The way that I would frame it instead is that the concentration of interest has increased. Nowadays with social media, etc, there are infinite ways to learn about, do, and compare various modifications to cars. People who do not care about cars are now in a position where essentially zero knowledge is required to use them as a method to get from A-B. However, those who are interested in it for process of building itself still exist, and the resources are better than ever. For the time that I've been around, I've been seeing increasing, not decreasing, interest in older car platforms primarily for the reason that they are easier to work on compared to new cars.</p>
]]></description><pubDate>Sat, 04 Jan 2025 22:14:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=42598070</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=42598070</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42598070</guid></item><item><title><![CDATA[New comment by juliusgeo in "My Colleague Julius"]]></title><description><![CDATA[
<p>I would also.</p>
]]></description><pubDate>Tue, 24 Dec 2024 10:15:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=42500936</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=42500936</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42500936</guid></item><item><title><![CDATA[New comment by juliusgeo in "Engineering Sleep"]]></title><description><![CDATA[
<p>Xyrem is just the brand name for GHB. The company that currently owns the rights (Jazz Pharmaceuticals) pled guilty to felony misbranding in 2011, and in 2013 raised the price by 841%. More recently, in 2017, they sued multiple other companies attempting to produce generics, before settling on an exclusive licensing agreement with one of them. Ironically the headline on their website is "Improving Patients' Lives", but I imagine they aren't reducing the price because Xyrem makes up 74% of total sales [1]. The entire thing reeks of PE--tons of acquisitions.<p>[1] <a href="https://en.wikipedia.org/wiki/Jazz_Pharmaceuticals" rel="nofollow">https://en.wikipedia.org/wiki/Jazz_Pharmaceuticals</a></p>
]]></description><pubDate>Sat, 30 Nov 2024 19:46:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=42283570</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=42283570</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42283570</guid></item><item><title><![CDATA[New comment by juliusgeo in "Car tires shed a quarter of all microplastics in the environment"]]></title><description><![CDATA[
<p>Not really relevant to your overall point, but I found it interesting that apparently F1 already tried that:<p><i>In 2005, tyre changes were disallowed in Formula One, therefore the compounds were harder as the tyres had to last the full race distance of around 300 km (200 miles). Tyre changes were re-instated in 2006, following the dramatic and highly political 2005 United States Grand Prix, which saw Michelin tyres fail on two separate cars at the same turn, resulting in all Michelin runners pulling out of the Grand Prix, leaving just the three teams using Bridgestone tyres to race.</i><p>[1] <a href="https://en.wikipedia.org/wiki/Formula_One_tyres#History" rel="nofollow">https://en.wikipedia.org/wiki/Formula_One_tyres#History</a><p>[2] <a href="https://en.wikipedia.org/wiki/2005_United_States_Grand_Prix" rel="nofollow">https://en.wikipedia.org/wiki/2005_United_States_Grand_Prix</a></p>
]]></description><pubDate>Fri, 29 Nov 2024 04:36:04 +0000</pubDate><link>https://news.ycombinator.com/item?id=42270838</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=42270838</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=42270838</guid></item><item><title><![CDATA[New comment by juliusgeo in "Supreme Court blocks controversial Purdue Pharma opioid settlement"]]></title><description><![CDATA[
<p>I can’t respond to any of your comments about policy, because you’ve never elaborated your ideas, and you seem to be assuming a great deal about mine and the other commenters.<p>1) Real heroin does <i>not exist</i> anymore in real life. All of it is varying cuts of fentanyl and research chemicals.<p>2) Your original comment asked the question “how many have died from oxy vs hepatitis”, which, along with your comments about heroin, made it clear that your knowledge isn’t up to date. It strains credulity to say that more people died of hepatitis than they did of the opioid they’re injecting into themselves regularly. Do you really think hepatitis is the root cause here? Or is it the proliferation of opioids <i>much stronger</i> than heroin. I’m sure you’ve heard of fentanyl and carfentanil, but there are entire other classes (such as Benzimidazoles) that have been rising to prominence in the last few years that are more dangerous.<p>3) I’m very anti DEA as well, and agree that prohibition kills. However, this is <i>not</i> simply a case of holding a manufacturer for a substance responsible. It’s what the Sacklers did <i>after</i> learning of the addictive power of these medicines that gives them culpability. It’s like a local bar serving someone even when they’re visibly drunk, which confers criminal responsibility if that drunk person later does something while under the influence. Making opiates == legal, being negligent == not legal. I wish we didn’t have a prohibition, and drugs could be acquired safely, but that doesn’t mean that the ends justify the means, and any drug manufacturer is automatically the good guy.<p>I just don’t think you’re approaching this subject with the care it deserves. It’s easy to make flippant comments blaming “Drug warriors” (genuinely confused what that even means—does that mean someone who is pro-prohibition?), it’s much harder to interact with the literal decades of research about this topic, and magnitudes harder to actually experience these things yourself. If you’d like a starting point, I found this report to be pretty approachable: <a href="https://www.rand.org/pubs/research_reports/RR3117.html" rel="nofollow">https://www.rand.org/pubs/research_reports/RR3117.html</a>. However, I wouldn’t be surprised if posing questions the way you have elicits emotional reactions from those who know people who have died. I lived through the first wave of fentanyl in the US, but I know quite a few who didn’t. Based off what i’ve been able to pick up out of all the vitriol, I think we would likely agree about the policy decisions, but your approach of making simplistic comments and then calling people evil based off of their response seems like an ineffectual strategy. And maybe it isn’t, but then you’d have solved the drug crisis already, right?</p>
]]></description><pubDate>Tue, 02 Jul 2024 04:34:07 +0000</pubDate><link>https://news.ycombinator.com/item?id=40853592</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=40853592</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40853592</guid></item><item><title><![CDATA[New comment by juliusgeo in "Supreme Court blocks controversial Purdue Pharma opioid settlement"]]></title><description><![CDATA[
<p>That’s a good question. You should do some research. Maybe you can ask the families of the victims?</p>
]]></description><pubDate>Sat, 29 Jun 2024 05:05:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=40828003</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=40828003</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40828003</guid></item><item><title><![CDATA[New comment by juliusgeo in "The Nissan GT-R Is Dead"]]></title><description><![CDATA[
<p>it does not take years to learn how to do most things on cars, especially if you choose an older car. garage and truck are also not a requirement unless your race car is no longer street legal. I’m a novice and have managed to replace all the coolant lines and do some not half bad paint work on my car (90s Acura). Drifting does burn tires, but also doesn’t require your tires be in good shape (they’re gonna get shredded, and the goal isn’t to set the fastest time). Time attack and drag racing are where it gets really expensive, but there are cheap options like auto-cross: <a href="https://www.scca.com/pages/what-is-autocross" rel="nofollow">https://www.scca.com/pages/what-is-autocross</a>.</p>
]]></description><pubDate>Sun, 16 Jun 2024 05:26:00 +0000</pubDate><link>https://news.ycombinator.com/item?id=40694880</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=40694880</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40694880</guid></item><item><title><![CDATA[New comment by juliusgeo in "Qlock – A JavaScript Quine Clock"]]></title><description><![CDATA[
<p>best place is the code.golf wiki: <a href="https://code.golf/wiki/langs/javascript" rel="nofollow">https://code.golf/wiki/langs/javascript</a>. They also have tips for other languages.</p>
]]></description><pubDate>Wed, 29 May 2024 15:23:39 +0000</pubDate><link>https://news.ycombinator.com/item?id=40512951</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=40512951</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40512951</guid></item><item><title><![CDATA[New comment by juliusgeo in "JetBrains releases RustRover IDE for Rust development"]]></title><description><![CDATA[
<p>I’ve been using the pre-release version for a bit because I am a big fan of PyCharm. I found it very good besides the profiling interface. The “jump to source” option never works for some reason. Overall, though, provides a great experience.</p>
]]></description><pubDate>Sat, 25 May 2024 02:49:55 +0000</pubDate><link>https://news.ycombinator.com/item?id=40472288</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=40472288</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40472288</guid></item><item><title><![CDATA[International Obfuscated Python Code Competition 2024: Submissions Open]]></title><description><![CDATA[
<p>Article URL: <a href="https://pyobfusc.com/#submit2024">https://pyobfusc.com/#submit2024</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=40408413">https://news.ycombinator.com/item?id=40408413</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Sun, 19 May 2024 17:36:56 +0000</pubDate><link>https://pyobfusc.com/#submit2024</link><dc:creator>juliusgeo</dc:creator><comments>https://news.ycombinator.com/item?id=40408413</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=40408413</guid></item></channel></rss>