HNHacker News
TopNewBestAskShowJobs

juliusgeo

338 karma · joined October 1, 2018

github.com/juliusgeo
submissionscomments
juliusgeo··on US hits Chinese labs over fentanyl
Yeah, consumption went down, but it caused numerous other issues that proved to be far greater. All you’ve shown is that reduced consumption of alcohol reduces the metrics of social problems associated with alcohol. That says nothing to my point which is that the other problems end up being far greater when you go in with such a profoundly callous mindset. Prohibition caused a 24% increase in crime overall with murder in particular going up 13%. And at the end of Prohibition people still drank and we still have all the same issues with alcohol and drugs now, just worse. And then Reagan tried the same approach again, and it failed again. How many dead bodies do you wish to see on top of the 100k overdoses to somehow solve this issue? Would a stack of dead bodies twice as tall satiate? At some point we as a society gotta realize that murdering people en-masse as a response to social issues is just not effective.
juliusgeo··on US hits Chinese labs over fentanyl
We tried that. It was called The Prohibition. It was a massive failure for many reasons but to your point: it created an environment where only the most violent, most criminal gangs end up controlling the entire market. The justifications given were also almost interchangeable with what you just said, mostly relating to scare tactics being perceived as a panacea to “society’s ills”. Most of the time if your solution is “scare people into doing what’s right” it’s been tried before by someone else and failed because it is a fundamentally broken method of solving societal issues in a meaningful way.
juliusgeo··on Show HN: Open-source resume builder and parser
Yeah, I see what you mean. In my case it is because I use LaTeX to create my resume, and my main concern is just that in some ATS systems it doesn't parse correctly and it would be nice to know when I'm making the resume, rather than when I apply for the job. For people without an existing opinionated workflow (likely the bulk), the resume scanner would be a good way to help them understand the value proposition of a unified builder that works with most ATS systems by design, rather than as an afterthought (which seems to be the case with most resume builders online besides this one).
juliusgeo··on Show HN: Open-source resume builder and parser
Just checked out the resume scanner and it’s very nice! I appreciate how it only took a few clicks to get from the home page to a real example with my own resume. It also works much faster than the resume parsers I’ve used before (most of which take on the order of 10 seconds just to get from the upload screen to a parsed output). One minor nit is that I think the resume scanner deserves its own button on the front page—the current placement makes it seem like an addendum to the “no sign up requirement” imo.
juliusgeo··on Show HN: My recommendation engine for Hacker News
It might be useful to pose this problem in terms of a precision vs. recall curve.
juliusgeo··on Reddit CEO Spez AMA Overview
My point was less so about whether or not the price was fair, but that it illustrates issues with Reddit's leadership. Whether or not the prices quoted during negotiations between Reddit and Apollo are fair is somewhat orthogonal to the issue of weaponizing those negotiations to score PR points.
juliusgeo··on Ask HN: What would an IDE built for the Apple Vision Pro look like?
Thanks for the reply! The main reason I ask that is I primarily use Python and Julia, which have pretty decent tracing abilities but there are still a lot of rough edges when it comes to visualizing those traces. Particularly with Python, the default tracing packages such as cProfile are quite lacking in terms of support for visualizing and exploring the output of the profiling (most likely by design, but still frustrating nonetheless). It's necessary to bring in Snakeviz for visualizing, and it is quite good, but the HTML UI that it is run has very limited options when it comes to distinct ways of visualizing the data.
juliusgeo··on Modern Life Is Rubbish: Tao Lin’s Taipei (2013)
The scene you’re referring to at the end I believe is from American Psycho, with Willem Dafoe playing the detective doing the interrogation.
juliusgeo··on Ask HN: What would an IDE built for the Apple Vision Pro look like?
Wow, this looks very impressive. Going to check it out. Is there any plan to support languages other than Swift? I’m assuming that has quite a few technical hurdles associated.
juliusgeo··on Bad waitress: Dying on your feet
I completely agree with this, but fundamentally I think that there is more value from identifying with the similarities between workers of all wage range. I was mainly just arguing against the point that the top reply in this sub thread was implying that high wage workers are somehow more exploited.
juliusgeo··on Bad waitress: Dying on your feet
The top comment is this sub thread very clearly included the words “even” to imply that while most would not assume software engineers are not exploited because they’re relatively wealthy, they are still being exploited. This very clearly implies that SEs are less exploited—there was absolutely nothing in the original comment that supports your mischaracterization of their point.
juliusgeo··on Reddit CEO Spez AMA Overview
Apollo’s dev would likely have never talked about the “buy me out thing” unless Reddits CEO had publicly implied he was trying to extort Reddit for money.

Also, I would love to see an in-depth explanation of exactly what “math tricks” he used. Not being sarcastic—I read his whole post and it seemed like he was very clear that his calculations were back of a napkin type.

juliusgeo··on Bad waitress: Dying on your feet
I worked as a busser and at multiple fast food places before getting a job as a software engineer. I have also been reprimanded for going to urinate at unscheduled times. I don’t think that gatekeeping the experience of being exploited for your labor is useful. Creating divisions between people who are only exploited, vs those who are exploited even more doesn’t help anyone except for those who are exploiting.
juliusgeo··on International Obfuscated Python Code Competition: Submissions Open
Weirdly enough, I can only replicate that behavior when I go to scale=80% and below, but it also varies based on the exact zoom. Perhaps if you zoom in one level below or above 125% it will resolve? I am admittedly not really good at CSS or HTML.
juliusgeo··on International Obfuscated Python Code Competition: Submissions Open
Also, one question I forgot to ask about the widescreen support. What browser are you using?
juliusgeo··on International Obfuscated Python Code Competition: Submissions Open
Thank you for the feedback!

1. Yes! Underhanded entries are welcome. The obfuscation doesn't only apply to the code, but also the purpose of that code.

2. No tricks are in poor taste! The only limit is that we asked for submissions that don't re-use the common types listed by the IOCCC: https://www.ioccc.org/faq.html

3. No max length limits, and yes if you are able to create a very short program that will help. If you submit a longer program it would help the judges to provide a de-obfuscated version as well.

4. Yes, as long as it doesn't require any human intervention to produce output. There can be an alternate mode that allows human intervention, but the main goal is just to reduce confusion from potentially incorrect inputs.

5. Good point, will fix.

juliusgeo··on International Obfuscated Python Code Competition: Submissions Open
That is a good point. The plan right now is to re-use the mailing list to send out submission information for each contest, so subscribing to that mailing list should give you some of the functionality. I will have to investigate setting up a real RSS feed.
juliusgeo··on International Obfuscated Python Code Competition: Submissions Open
Thanks! I'm very excited to see it. You can submit any Python program as long as it produces an output of some kind without human intervention. More details are here: https://pyobfusc.com/#rules
juliusgeo··on International Obfuscated Python Code Competition: Submissions Open
Yeah, I tried a few times to fix the syntax errors but couldn't. I love that method of obfuscation, though.
juliusgeo··on International Obfuscated Python Code Competition: Submissions Open
Directly inspired by the classic C coding competition, the IOCCC, the IOPCC is now accepting submissions. This is my first time running a competition, so there will likely be some issues in this first year. Tentative plan for submissions end-date is outlined on https://pyobfusc.com/#submit.
juliusgeo··on International Obfuscated Python Code Competition
Directly inspired by the classic C coding competition the IOCCC, the IOPCC is now accepting submissions. This is my first time running a competition, so there will likely be some issues in this first year. Tentative plan for submissions ending is outlined on https://pyobfusc.com/#submit
juliusgeo··on In vitro biological system of cultured brain cells has learned to play Pong
I would dare to say that's not even an outrageous proposal, more of a modest.
juliusgeo··on In vitro biological system of cultured brain cells has learned to play Pong
Maybe they could rebrand to "Brainlent Green".
juliusgeo··on Radiation-hardened Quine (2014)
mame is also very well known for their IOCCC submissions. My favorite is https://www.ioccc.org/2012/endoh1/hint.html.
juliusgeo··on Two men got jobs at Amazon just to steal copies of Zelda
The US federal government has a similarly high conviction rate, especially considering the number of plea deals (90% of defendants plead guilty). https://www.pewresearch.org/short-reads/2019/06/11/only-2-of.... I went to Wikipedia to find more information and it seems that your characterization of it as very unjust is a common misconception:

> According to Bruce Aronson of New York University School of Law, Japan's conviction rate is misleading because it is the rate at which defendants admit guilt in the cases they are charged with. According to him, if the method of calculating the conviction rate in Japan is applied to the United States, the conviction rate of federal defendants in the United States in 2018 was also over 99%.

https://en.m.wikipedia.org/wiki/Criminal_justice_system_of_J...

juliusgeo··on Children's Games (Bruegel)
I first saw this painting in the intro to Melancholia, by Lars von Trier. I’ve had it as my desktop background for years. It reminds me a lot of the feeling of watching all your neighbors shovel after a snowstorm. Funny that the movie communicates an unsettling and existential story that is the opposite of cosy.
juliusgeo··on Young Americans Struggle to Hit ‘Adult’ Milestones
I can count on a single hand my peers that have fulfilled any of those three milestones. I just turned 25.
juliusgeo··on Micromouse: The fastest maze-solving competition [video]
There was a very brief mention in the video, but this link goes into more detail: https://micromouseonline.com/micromouse-book/rules/. The score of each run factors in 1/30th of the time spent to search. There is a separate period where the mouse can search, and then it goes through the maze at high speed using the route it planned prior. Obviously if you spend a massive amount of time searching, that will impact your score but it seems to prioritize the later, faster runs in terms of scoring.
juliusgeo··on What number comes next? The On-line Encyclopedia of Integer Sequences knows
Another comment in this thread mentioned that Neil Sloane now has his own channel, and is making some videos [0]. Very fun from that 20 minutes I've watched so far.

[0] https://www.youtube.com/watch?v=9ogbsh8KuEM

juliusgeo··on What number comes next? The On-line Encyclopedia of Integer Sequences knows
Paywalled article, but I’m assuming it’s talking about the OEIS located here [0]. Such a cool resource, especially if you are trying to calculate S-boxes[1], or a lot of digits of an irrational number, or Bernoulli numbers or something. The links section for each sequence is super useful for initial research.

[0] https://oeis.org [1] https://oeis.org/A354500

← PreviousPage 2 of 5Next →