Filipe Marchesini

[Question] What is the ap­pro­pri­ate way to com­mu­ni­cate that we are ex­pe­rienc­ing a pan­demic?

• Filipe Marchesini 4 Mar 2020 1:16 UTC

  1 point

  on: Coron­avirus: Jus­tified Prac­ti­cal Ad­vice Thread

  Should I expect a faster infection rate on my country (Brazil) because most people here use paper money to make trades? Should I recommend people to stop using paper money and instead opt for a contactless card? Most people don’t have access to banking services; so is there any option for them?

• Filipe Marchesini 11 Mar 2020 9:27 UTC

  17 points

  on: Wel­come to Less Wrong!

  WHO I AM: I have 24 years of existence. I give math, chemistry and physics lessons to high school students since 17. I am pretty good at it and I never announced anywhere on planet that I give lessons—all new students appear from recommendations from older students. On the end of 2016 I already had 38 months going to the university, trying to get mechanical engineering credentials. I wasn’t interested on the course—I really liked the math and the subjects, but the teachers sucked and the experience was, in general, terrible. I hated my life and was doing it just to look good for my parents—always loved arts and I study classical music since 14. I heard about “artificial intelligence” just once, and I decided all my actions in life should be towards automate the process of learning. I started a MIT Python course and then dropped out university. I am completely passionate about learning.

  WHAT I’M DOING: (short-term) I am currently learning and doing beautiful animations with the python library called MANIM (Mathematical ANIMations). I am searching for people to unite forces to transform tens of posts in The Sequences into video content with this library. I hope to gather money from it and spread rationality in general, turning it more popular. My family and my best friends have shit quality life and I would like to get enough money to change that. Most of my reasoning is “explore the solution space and find the best ways to help most people, if you really help and it is something scalable, you will get money on the way”. If I get the money (instrumental goal), I will help those who need (relatives and close friends). As soon as I achieve this goal, I will jump to my long-term goals on helping ending disease, extending life (first) then killing unwanted death, refining art, playing games.

  HOW I FOUND YOU: I googled “pascal wager artificial intelligence” after seeing a Robert Miles (AI researcher) video. Then I found Roko’s basilisk, read a lot about it and had a bad week. Then I found Eliezer Yudkowsky talking about it, and my fear went away. Then I discovered this forum and it turned out to be my main source of truth-seeking. I don’t know personally anyone who has access to this kind of source of information (Lesswrong), collective truth-seeking with strong grounds.

  I VALUE: I value people who works towards making the life of others better. I value people who really seeks truth. I value people who takes ethical problems seriously. I value people who spends more than 5 minutes by the clock looking for better solutions for our everyday problems.

  TO ACHIEVE WHAT I VALUE: To make the life of others better, I am daily trying to discover how can I use the programming knowledge I am acquiring daily about blockchain, artificial intelligence and mobile/​web/​software development to create real solutions for real problems—solutions for drinking water, food (automate prodution/​distribution), education (for all ages), energy, housing, income, health and environment. I didn’t develop a scalable solution for any of these problems, because I am still learning and it is very hard, I admit, but I will help, no matter what, and if you wanna lose, just bet against me. I just need a little more time.

Hy­drox­y­chloro­quine: the day sci­ence stopped

• Filipe Marchesini 5 Apr 2020 18:35 UTC

  4 points

  in reply to: ioannes’s comment on: How strong is the ev­i­dence for hy­drox­y­chloro­quine?

  Gautret et al. 2020° shouldn’t be considered a study of hydroxychloroquine efficacy. It should be considered a failed attempt at studying hydroxychloroquine efficacy. Here is why I believe that.

• Filipe Marchesini 5 Apr 2020 19:58 UTC

  1 point

  in reply to: renato’s comment on: Hy­drox­y­chloro­quine: the day sci­ence stopped

  Thanks, corrected it now.

• Filipe Marchesini 5 Apr 2020 22:25 UTC

  3 points

  in reply to: ChristianKl’s comment on: Seek­ing opinions on the pros and cons of var­i­ous telep­res­ence tools

  Are you aware or worried about reported Zoom privacy and security issues?

Look­ing for a Team to Par­ti­ci­pate on a Competition

• Filipe Marchesini 29 Apr 2020 8:56 UTC

  10 points

  on: Prospect­ing for Con­cep­tual Holes

  Learning a new language (English) was decisive in my life to fill a lot of type 1, type 2 and 3 holes as you presented on this post.

  In 2013 “I felt” I should learn English. I didn’t have a clear goal on doing it, but “it seemed important on the long-term”, but I couldn’t explain why it would be important to care about really learn another language.

  In 2016 I heard from a YouTuber from my country talking about AI. My goals changed completely, and I could only find real quality discussions about AI or state-of-the-art information in English.

  Then it came programming, blockchain, decision theory, etc, etc. Subjects that I only started caring because I learned English and found really good English content about these subjects.

  This post makes me wonder what would happen if suddenly I started to learn Chinese.

• Filipe Marchesini 29 Apr 2020 13:38 UTC

  1 point

  on: Filipe March­esini’s Shortform

  Suppose a dynamic in which a creator announces that wants to complete a task. E.g.:

  Creator: “I would like to give a lesson teaching how to remove the background of a photo”.

  The creator optionally puts a threshold in which, once the users already demonstrated enough interest, the creator starts streaming his job. Let’s suppose, e.g., that the last creator have put the threshold: “$50.00”.

  Users decide to “put money on the table” on the platform, signaling interest in seeing the creator accomplishing the task.

  Users can also signal interest without putting money on the table. The conventional “Like system”. The creator can also put a threshold on the number of likes he would be willing to create the content.

  Suppose during an hour these users demonstrated interest:

  Users_and_values = {User1: $0.50; User2: $2.50; User3: $1.50; User4: $0.25; User5: $0.00004522; … ; UserN: $0.10}

  Once the sum(Users_and_values.values()) == 50 (it hits the threshold), the platform signal to the creator that there are enough users willing to pay for his content.

  The creator starts streaming what he promised he would do (all registered users on the creator channel receive notification messages). During the stream, the users that had showed interest (and those that hadn’t too) in paying can pay any time, any amount of what has been promised. For example, once the streamer already have accomplished 50% of what he promised (and signal that), users could decide to pay 50% of what they promised. “Paying as he goes” would be very cool and it would have a bar being automatically updated showing the amount already collected by the creator during the stream. If the job is not good or it is not like what has been promised by the creator, users can simply not pay. Users could simply pay 100% of what they said they were willing to pay once the streaming ends. I think you get the idea. If not, let me know if this is not clear.

  Rules:

  Users can not show willingness to pay more than what they have on their wallet;

  If a user says he is willing to pay $1.00 for seeing a task accomplished by a creator and the user didn’t pay anything after the creator accomplished his job, the user should justify why. If he doesn’t justify, his karma will be lowered and the next time he promises to pay $1.00, the amount would be multiplied by the probability of he paying.

  There should be a threshold on maximum willingness to pay. Otherwise, an user with $1,000,000 could just show fake willingness to pay for all creators.

  The payments would occur using the digital currency NANO, that allows free and instant transactions. I would like to see your input on the general idea, there are several details not discussed yet.

  For instance, this could be used to reward LessWrong creators for creating their posts. Or even noobs like me would be willing to try more crazy ideas just because there are enough money on the game. Normally, I wouldn’t randomly “create a sale system for bakeries”. But if people suddenly said they would pay $50 bucks for seeing me doing that, that’s a completely new story.

  Users could also create their posts requesting what they want:

  Eliezer: “Suppose you underwent a low-risk Corona exposure situation, but you very much want to avoid infecting a pool of friends or housemates. Is there a ⁵⁰⁄₂₀ (if maybe not ⁸⁰⁄₂₀) on quarantining yourself away during the time when you are most likely to be (quite?) contagious without having developed symptoms? [...] $500 for a definitive response of this kind with citations. ”

  So Eliezer creates the post, and puts the money. Users start submitting their answers to the post, and whenever Eliezer sees a satisfying solution, he simply clicks on “reward user”.

  If Eliezer never pays anyone on a sucession of posts and without justification, he could have his karma lowered. Users could report other people that are creating baits

• Filipe Marchesini 29 Apr 2020 16:01 UTC

  2 points

  in reply to: lsusr’s comment on: Prospect­ing for Con­cep­tual Holes

  These points really make me update my beliefs on the importance of learning Chinese. Thanks for that.

  What am I missing by not speaking Portuguese?

  Brazilian speaks with the body, so you are not missing that much. Haha just kidding (although it’s true that body language plays a big role here). But there are 250 million total portuguese speakers, and by just knowing Portuguese you will understand Spanish easily, because there is 89% of lexical similarity. I can understand Spanish with no problems and I never studied it. If you know how to search, there is a vast and extensive Brazilian music that you wouldn’t like to miss. The literature written in Portuguese remains untranslated, except for very few authors, so you would be able to read very good books (if you like certain genres). Also, Brazil is about half of South America, and is a fantastically interesting place to travel. When new people come here, the reception is warm. Music, cooking, sports and memes here are a real thing. Knowing Portuguese would make you able to connect with our culture and you would be able to access very good content on a lot of subjects. But I am fully biased and mostly looking into the bubble (mathematics, physics and computation) and you don’t lose much by just using English sources for these subjects.

  For example, this channel is specialized in non verbal behavior analysis and I never found anything better than that on any language. This channel is the best on musical content (how to play drums, guitar, acoustic guitar, etc, how to sing, compose, etc). This on comedy, this on history and science. I mean, if you keep looking, you will always find a lot of good content on very different subjects. You will miss all of this by not knowing Portuguese, as with any language.

  If you look on the science side, you can check Brazil here (tables from page 4~9).

  There is no good translator from Portuguese to any language yet, and your best tool to date would be DeepL, in case you need to translate Portuguese info to English.

• Filipe Marchesini 29 Apr 2020 16:52 UTC

  1 point

  in reply to: ESRogs’s comment on: ESRogs’s Shortform

  35% crypto—XTZ, BTC, and ETH (and small amounts of LTC, XRP, and BCH)

  Why do you bet on cryptos with weak fundamentals like LTC, XRP, and BCH, and not on cryptos with stronger fundamentals like Monero and Nano?

• Filipe Marchesini 29 Apr 2020 19:02 UTC

  5 points

  in reply to: ESRogs’s comment on: ESRogs’s Shortform

  Got it. I wonder why there aren’t discussions about cryptocurrencies here on LW. Like, extensively researching the existing protocols and their characteristics and then naming which one is the best and why, and betting on it. Or even using it on the forum. I would suggest you to research these two cryptos I have cited if you are interested on updating your bets.

• Filipe Marchesini 29 Apr 2020 20:00 UTC

  6 points

  on: Peter’s COVID Con­soli­dated Brief − 29 Apr

  Thanks for the great effort for bringing a summary that really makes me feel I don’t have to continue digging for new information. If this work is going to be continued, I will just wait for your new updates, and then finish looking for local news, as things on my country are largely different from yours. Keep it up!

• Filipe Marchesini 4 May 2020 8:33 UTC

  2 points

  in reply to: Raemon’s comment on: FB/​Dis­cord Style Reacts

  “I was reading your post, but suddenly I had to leave in the middle of it. I would like to say that I agree/​liked your intuition (to the point that I read it), and I would like to signal this immediately, because I’m afraid I won’t remember to finish the reading in the future, although I would like myself to finish it in the future.”

  “Although your post seemed to be going in a direction I like, I had to abandon the reading. Forgive me for leaving it and not finishing absorbing a seemingly important message. I hope this reaction helps you, I’m out of time to explain why I would like to help you, bye”
  
  ″That sounds like something someone from my tribe would say. Then you seem to be an ally.”

  “There’s only one part of your message that seems to be important.”
  
  ″I’ll hit like, because you quoted ‘the super important thing’. Although I’ve seen a lot of irrelevant information in your post, I’m short of time to comment and say which part is important and which part is not. This reaction represents that ‘there is an important thing in your post’ and ‘this thing’ is the ‘super important thing’.”

• Filipe Marchesini 4 May 2020 18:44 UTC

  2 points

  on: On Nega­tive Feed­back and Simulacra

  Hot sauce:

  “[...] insults her cooking and insinuates that she doesn’t know how to cook”

  If my girlfriend had said something like that, I think an appropriate solution would be:

  [curious] what you mean… [\curious]. [surprised] AHHHH, the hot sauce?? [\surprised]. [empathy] Ohhh, my love. Really sorry for that [\empathy]. [confession ~ lower speaking] I have to tell you something. I am so addicted to hot sauce that I even didn’t remember that I shouldn’t use it in (clearly exaggerating) every meal I eat. Your meal is [emphasis] really amazing [\emphasis] and I didn’t want to make you feel bad. This action of putting hot sauce was completely automatic (this conveys that I wasn’t carefully choosing my actions, I was just acting by default), because I am very used to put this thing on quite every meal, I swear this isn’t a special occasion. Sorry me for that, next time I won’t use it and I will really try to appreciate the meal without it. It isn’t even healthy to put this thing on every meal. Thanks for remembering me that.

  Thinking again about it, if my girlfriend had said that, she wouldn’t be trying to signal that I was insulting her cooking, but she would be trying to signal that I should praise her food. I think the guy on the post wasn’t praising from the beginning. For example, once I get the first mouthful, I should already say “woooww, this is delicious”. Then if I proceeded to put hot sauce, she wouldn’t simply say “you are insulting me”, but maybe she would assert “are you really liking the food? I see you are putting extra condiments”, then I would reply, “no, this is really delicious. I am putting the hot sauce because I am addicted to it, but if you find this bad, I won’t put it anymore. I really would like to leave this addiction.”

  Anyway I really like your posts Zvi and I would like to see more thoughts of yours here on Lesswrong. I see you carefully making your steps and I admire and learn from that.

  Sorry for not exploring more on the other points of the post. Time is out, bye and thanks

• Filipe Marchesini 3 Jun 2020 10:58 UTC

  36 points

  on: Open & Wel­come Thread—June 2020

  LessWrong warned me two months before it occurred here. The suggested preparedness was clear and concise, and I felt the power on my hands. I had valuable info no one on my tribe had. I alarmed my mom and she listened me, stayed home and safe, when everyone was partying out (carnival). I long-talked with friends, and explained to them what I believed it was happening and why I believed that. I showed the numbers, the math, the predictions to the next week, next week came, and reality presented its metallic taste. Week after week, the light was getting brighter and brigher until it turned really hard to refuse to see it, or to believe on the belief that everything was just fine.

  One thing I learned is that it doesn’t matter if you just know something really valuable, but can’t convince those that do matter for you. I tried to explain my 50 years experienced physician father that he should listen to me. He blamed my low status. But even after weeks, police at the streets forcing citizens to stay at home, he could not believe. He was in denial and my incompetence to change his mind made him to the hospital, 16 days and he isn’t still back. Don’t worry, he is getting better and I am just babbling around. Father of a cousin died. Brother tested positive. Stepmother obviously got it too, but tested negative. You know, I really tried, not just tried to try, not even planned to try, I really did it right way. Successive failures. It wasn’t enough. I don’t have anything valuable to share more than ‘you have to learn ways of convincing your most loved ones’ urgently, if you don’t have this tool, but I don’t know how to do it yet, I am struggling to find a way, and I would ask you to share when you get one. Things as simple as “there is food over there” or “there is a lion coming to you”, on level 1 talk. Maybe the dark arts could have helped me when level 1 failed, but not sure. But I feel very happy for a lot of peers I helped along the way, and all is due to LessWrong, I am thankful for this community, and I changed the behavior of some people I know on the right moment of the outbreak. This simple text forum saves many lives and I am on the path to contribute too on larger scale.

  I know I am not that new on the forum, I don’t remember exactly when I started here, but I believe it was on the last months of the last year, but I still think I am noob, but learning. Don’t upvote this comment, but do comment if you wanna say something.

• Filipe Marchesini 9 Jun 2020 21:32 UTC

  7 points

  in reply to: Sherrinford’s comment on: Open & Wel­come Thread—June 2020

  I think you can refer the person to orthogonality thesis

• Filipe Marchesini 12 Jun 2020 13:07 UTC

  1 point

  in reply to: lsusr’s comment on: 2020 LessWrong De­mo­graph­ics Survey

  Exactly, I am from Brazil and I can’t see any option that fit my race.

  According to Wikipedia, “According to the American census, the Hispanic or Latin category would not include Brazilians or Americans with origins in Brazil, [3] [4] [8] as it is specific to people of “Spanish culture or origin”. [3] [4]. Technically speaking, people from Portugal or of Portuguese origin are called Lusitanians. In Portugal, the term “hispanic” refers to something related to ancient Hispania, Spain or the Spanish language and culture [9]. The common modern term for identifying both Portuguese and Spanish cultures under a single nomenclature is “Iberian”, and the term referring to cultures derived from both countries in the Americas is “Ibero-American”. These designations can be mutually recognized by people in Portugal and Brazil, in contrast to “Hispanic”, which is totally devoid of any self-identification in these countries, and quite the contrary, is used to mark a clear distinction in relation to the culture of neighboring Spanish-speaking countries (Hispanics) in relation to Portuguese-speaking countries (Portuguese-speaking).”

• Filipe Marchesini 13 Jun 2020 13:46 UTC

  1 point

  in reply to: Lucent’s comment on: Open & Wel­come Thread—June 2020

  I think you are referring to Goodheart’s law, because all the measures your examples used as a proxy to achieve some goal were gamified in a way that the proxy stopped working reliably.