Ask me about:

  • Science (biology, computation, statistics)
  • Gaming (rhythm, rogue-like/lite, other generic 1-player games)
  • Autism & related (I have diagnosis)
  • Bad takes on philosophy
  • Bad takes on US political systems & more US stuff

I’m not knowledgeable about most other things

  • 23 Posts
  • 31 Comments
Joined 2 months ago
cake
Cake day: September 15th, 2024

help-circle


  • Currently live in a condo, I think every unit in the building came with one

    The biggest advantage I could find is that they are insanely convenient for making French press coffee! French presses are otherwise a pain to clean (since there’s no filter to aggegate the grounds), but having an in-sink disposal means I can just flush the coffee grounds directly into the sink. Besides this though I’m pretty indifferent to them




  • So this is a bit counter to the news article’s point, and apologies for linking to Reddit… but there has been a fairly hot post on the subreddit r/USCIS. A practicing immigration attorney was sharing some thoughts on how feasible the promises are https://www.reddit.com/r/USCIS/comments/1glflxy/so_what_now_an_immigration_attorney_perspective/. Some quotes:

    IMO, no-- the economy makes way too much money from DACA folks. I do believe that they will dangle it like a carrot to appease right-wing voters. Major corporations employ DACAmented folks. The SSN from work permits have allowed more tax revenue to come in. Too much is at stake. Legally, the legal arguments at the courts surrounding DACA involve constitutional rights, which themselves aren’t going anywhere anytime soon. It’s honestly just a topic that is often talked about, but hardly understood by many.

    I want to put this into perspective. There are 11 million undocumented immigrants in the US. Currently, DHS has about 92,000 officers, and ICE has about 21,000 officers. It is asinine to try to achieve this.

    Let’s say it actually does begin and people are getting rounded up. Guess what? Not all undocumented folks are just undocumented-- many have TPS, pending asylum applications, pending T/U Visas, and work permits (see my point regarding #1). Unless a migrant has an expedited removal (not likely), DHS/ICE still needs to process each deportee, assign them A#s, and follow basic procedures. If they don’t? That’s a very easy way to reverse a deportation order. It’s the equivalent of convicting someone of murder using a confession made under a very obvious 4/5th amendment violation. Slam dunk case.

    Oh, and you know who has to handle all of these deportation cases? Federal DHS attorneys. They’re already overworked, and they tend to exercise discretion. If no discretion, the overworked ones tend to gloss over cases and provide weak arguments. Only major attention is paid to serious crimes. You’d be surprised the amount of times DHS attorneys have gotten my clients’ names wrong or made procedurally embarrassing typos.

    … assuming the administration still follows basic social contracts, that is. If the Trump administration actually uses the military to forcefully enforce mass deportations, then I feel the US is going to be fucked on so many different more levels… and there would be way more to worry than just the deportations















  • Oh my! I didn’t know what to expect, and I have to say… I was quite surprised by some of your answers. Also confirmed to me that I am definitely not normal

    Not many replies that are indicative of Aphantasia so… here goes nothing. I tried really hard at this okay

    spoiler

    I don’t “see” see anything when I close my eyes. I can create a very vague concept of a ball, a table, and… kind of a person in my head, but I don’t actually see the scene, I used to think when people say imagining things they were just making a metaphor. Things get really funk from here… But the overall schema feels more like one of those badly drawn scenes from the hit visual novel Slay the Princess. And yes I imagined it in 2D for some reason

    • Color: the ball doesn’t have a color
    • Gender: it wasn’t even a real person; it seems like a silhouette of the hand and back of a person
    • Looks: As I said, the person isn’t even facing me
    • Size: No idea; in retrospect it’s fairly large compared to the table (diameter probably 1/2-1/3 of table?), but the table is also an abstract concept so…
    • Table: no clue, it is a square table but that’s it. If anything it looks like the things served on Pizza Hut pizzas
    • Well I spoiled the question for myself so… but I didn’t have to choose, heck I couldn’t choose even if I know what the questions are






  • I don’t believe anyone mentioned this yet so… here goes nothing, there is a suspicion that this is due to A/B testing

    This is a bug report from the Invidious project; this is back in June 6 (so four months ago), but the hoster of a fairly large instance noted a very bizarre error message on the Invidious project…

    Conclusion is that Youtube is very likely rolling out A/B testing of requiring all clients to login before viewing videos

    Refreshing will probably work considering this is most likely result of an A/B test, but unfortunately I don’t see a way of this problem going away



  • So it was the physics Nobel… I see why the Nature News coverage called it “scooped” by machine learning pioneers

    Since the news tried to be sensational about it… I tried to see what Hinton meant by fearing the consequences. Believe he is genuinely trying to prevent AI development without proper regulations. This is a policy paper he was involved in (https://managing-ai-risks.com/). This one did mention some genuine concerns. Quoting them:

    “AI systems threaten to amplify social injustice, erode social stability, and weaken our shared understanding of reality that is foundational to society. They could also enable large-scale criminal or terrorist activities. Especially in the hands of a few powerful actors, AI could cement or exacerbate global inequities, or facilitate automated warfare, customized mass manipulation, and pervasive surveillance”

    like bruh people already lost jobs because of ChatGPT, which can’t even do math properly on its own…

    Also quite some irony that the preprint has the following quote: “Climate change has taken decades to be acknowledged and confronted; for AI, decades could be too long.”, considering that a serious risk of AI development is climate impacts


  • Oh my… I had a slightly similar incident. New phone number, had a bunch of random strangers texting me (some even calling!) asking for Ethan. My name is not Ethan, I didn’t know who Ethan is

    No idea what was on my mind back then, but I somehow got the contact info of this mysterious Ethan, called him (hilarity ensued since he got a call from someone on his contact list named “Me”), confirmed his up-to-date number, and promptly referred everyone looking for Ethan to the real person for over a year…

    Life is strange sometimes



    • A privacy-respecting mail service: I use mailbox.org since it follows email standards, but I think many ppl like Proton mail/Tutanota. Recommend because they are privacy-respecting, and self-hosting email is way too difficult
    • More of a yearly subscription per-se, but a personal domain from any domain registrar. Recommend because why not? There are so many cool things one can do with a domain: custom email, your own blog, professional website for job, …
    • A VPS from Linode (or any reliable provider). Recommend because some things are better done on a VPS… and I want a public-facing IP that is not directly from my bedroom
    • I used to have subscriptions to the local arcade. Recommend because I basically get cardio workout on the DDR machine (and it costs less than a gym. And easier to cancel)




  • Based on my understanding of how these things work: Yes, probably no, and probably no… I think the map is just a “catalogue” of what things are, not at the point where we can do fancy models on it

    This is their GitHub account, anyone knowledgeable enough about research software engineering is welcomed to give it a try

    There are a few neuroscientists who are trying to decipher biological neural connections using principles from deep learning (a.k.a. AI/ML), don’t think this is a popular subfield though. Andreas Tolias is the first one that comes to my mind, he and a bunch of folks from Columbia/Baylor were in a consortium when I started my PhD… not sure if that consortium is still going. His lab website (SSL cert expired bruh). They might solve the second two statements you raised… no idea when though.