toobsock's recent activity
-
Comment on What are you reading these days? #8 in ~books
-
Comment on Who do you think Dems will run in 2020? This article says it's Sanders or Warren...but is it? What does Tildes think? in ~talk
toobsock (edited )Link ParentI’d like to see him in the primary debates at least. I agree, he’s not a sharp statesman, but him being on stage at least helps rebrand the party image in a way that’s accessible to middle...I’d like to see him in the primary debates at least. I agree, he’s not a sharp statesman, but him being on stage at least helps rebrand the party image in a way that’s accessible to middle America. We need to change the perception that we’re all about socialism and identity politics.
We’re the party that actually will help blue collar Americans, and yet they still don’t like us. I grew up listening to Rush Limbaugh in a red state. That audience will always hate us as long as they can brand us the “weak, hippy, intellectual, safe space, college types without any real life experience.” I really just like the idea of The Rock, because he totally flips the script on the entire republican strategy. Hopefully in 20 years we can start voting in AI candidates.
-
Comment on Who do you think Dems will run in 2020? This article says it's Sanders or Warren...but is it? What does Tildes think? in ~talk
toobsock Link ParentHad never considered The Rock either, but it’s genius. The optics of those two on stage. He’s physically more imposing, and likely funnier. Can play the WWE part when necessary, or put on glasses...Had never considered The Rock either, but it’s genius. The optics of those two on stage. He’s physically more imposing, and likely funnier. Can play the WWE part when necessary, or put on glasses to transform into serious and intellectual Rock. These family friendly movies like Moana and Jumanji, makes him seem like a big softie who can also protect our kids with his he-man strength. This may be Trump’s only weakness...
-
Comment on California abolishes cash bail, replacing with algorithmic based risk assessment in ~news
toobsock Link ParentIf the algorithm has been trained on a huge dataset of inmates, and is incredibly accurate in predicting recidivism rates using a test set of historical cases, then absolutely. This seems easy...If the algorithm has been trained on a huge dataset of inmates, and is incredibly accurate in predicting recidivism rates using a test set of historical cases, then absolutely. This seems easy enough to answer. Just compare the results to the accuracy of the humans.
-
Comment on Would you consider it unhealthy to smoke pot twice a day recreationally? in ~health
toobsock LinkI started smoking when I was 17, and finally quit when I was 27. I’ve had many beautiful and worthwhile experiences while high, but I was never able to keep it a “once in a while” thing like some....I started smoking when I was 17, and finally quit when I was 27. I’ve had many beautiful and worthwhile experiences while high, but I was never able to keep it a “once in a while” thing like some. Neither could the majority of my friends, although we liked to assume we could quit anytime if we just had a reason to.
This may not apply to you, although it probably does, but daily marijuana usage made me dumber, slower, and less capable of planning for my future. It created a thin, lingering fog around my mind that was imperceptible at the time, but regularly led me to make decisions that were rewarding in the short term at the expense of my long term happiness and well being.
My dreams were always just a few years away. Every year was going to be the year I got serious about doing [insert dream here], although tonight I just feel like getting high and chilling out. The days blur together, and you may find yourself having little to show for it years later.
How about you keep smoking everyday, but set daily, weekly, and monthly goals for yourself. Exercise X times a week, spend X hours practicing piano a day, etc. If you’re meeting your own expectations you have nothing to worry about. Either way you’ll have your answer.
-
Comment on Y Combinator plans to start doling out $60 million next year to study universal basic income in ~finance
toobsock Link ParentThe New Yorker did a long piece on him, and he seemed very intelligent. I wouldn’t be so quick to call him clueless. And I don’t think it’s completely obvious that limiting hate speech will make...The New Yorker did a long piece on him, and he seemed very intelligent. I wouldn’t be so quick to call him clueless. And I don’t think it’s completely obvious that limiting hate speech will make for a better society 20 years from now. We might be breeding more alt-right youth by dividing ourselves into ideologically homogenous groups. I was bullied in school and romantically undesirable. A place like 4chan might have appealed to me because at the time I had some unhealthy views towards women. Now after being in the 4chan echo chamber, I might develop unhealthy views towards minorities and start using the word “cuck”, and I’ve got a deeper hole to dig out of.
But I also like being on tildes and not having to read racist shit.
-
Comment on What computer/programming/etc. project are you working on this weekend? in ~comp
toobsock Link ParentI’ve only ever worked with images really, so this is quite a new space for me as well. I’ll look into NLTK. I’ve only done one NLP project in the past which was a classifier for suicidal comments...I’ve only ever worked with images really, so this is quite a new space for me as well. I’ll look into NLTK. I’ve only done one NLP project in the past which was a classifier for suicidal comments on reddit. It was modeled after some existing research on Twitter comments. It used a bag of words method and completely disregarded grammar.
What’s your project idea?
-
Comment on What're you drinking this week? in ~hobbies
toobsock Link ParentWater is a classic. Sodas are good too. Carrot juice if you’re feeling adventurous. But seriously, aside from Welch's sparkling grape juice, I’ve not come across a good non-alcoholic anything.Water is a classic. Sodas are good too. Carrot juice if you’re feeling adventurous. But seriously, aside from Welch's sparkling grape juice, I’ve not come across a good non-alcoholic anything.
-
Comment on What computer/programming/etc. project are you working on this weekend? in ~comp
toobsock Link ParentAre you using conv nets or something else? I’m playing around with using NLP to build a social platform inspired by some things I’ve seen on tildes and reddit. It would auto apply tags to comments...Are you using conv nets or something else? I’m playing around with using NLP to build a social platform inspired by some things I’ve seen on tildes and reddit. It would auto apply tags to comments and posts in an attempt to group discussions more naturally. I’m envisioning moving through a 3D landscape of past and present discussions, so when the next gender pay gap or taxation debate comes up, they would be identified by an algorithm which links them to previous debates in order to prevent the same discussion from taking place again, and hopefully a new one can be built on the existing pile. Still working through the details with a friend.
What’s your personal assistant tool like? Is it like a simpler version of Siri?
-
Comment on The impossible job: Inside Facebook’s struggle to moderate two billion people in ~tech
toobsock Link ParentThe salaries for 100k employees making minimum wage would be 1.5 billion a year. That's not including health insurance or the office space for this army of workers. Not saying you need 100k...The salaries for 100k employees making minimum wage would be 1.5 billion a year. That's not including health insurance or the office space for this army of workers. Not saying you need 100k workers to do this job effectively, but it doesn't seem as easy as you make it out to be.
-
Comment on Facebook is rating the trustworthiness of its users on a scale from zero to 1 in ~tech
-
Comment on Facebook is rating the trustworthiness of its users on a scale from zero to 1 in ~tech
toobsock Link ParentI don't think hiring humans to do this task is even an option. If you're trying to assign a trustworthiness score to 2 billion people, it would take an enormous amount of people years to sort...I don't think hiring humans to do this task is even an option. If you're trying to assign a trustworthiness score to 2 billion people, it would take an enormous amount of people years to sort through, and the judgement of each employee is going to vary to the point that a score will likely be worthless. Humans are good at determining if something is child porn or not, and those cases are rare enough that perhaps humans can manage those reports.
This case seems much more suited to an algorithm, and it's not untestable. Create a data set by manually looking at a lot of profiles, and labelling them with a trustworthiness score. Divide the data set into a training set and a test set. Train the algorithm on the training set, then see how it does on the test set. If it did great on the training set, but terrible on the test set, then your algorithm needs tweaking. If it's able to predict the scores on both sets, then it learned what the humans were looking for when determining the scores themselves, and your model is ready for production.
-
Comment on Aquarium/fish enthusiasts, what are you up to? in ~hobbies
toobsock LinkAnyone running a heavily planted tank or aquascape? I’d love to see some pics of your tank if you feel like sharing.Anyone running a heavily planted tank or aquascape? I’d love to see some pics of your tank if you feel like sharing.
-
Comment on Deepfakes are coming. Is Big Tech ready? in ~tech
toobsock (edited )LinkAnytime a Reddit thread shifts my opinion on something, I have to wonder if it was engineered to do so. And if it wasn’t, then I’m staring at the blueprints. Use two accounts to fake a discussion...Anytime a Reddit thread shifts my opinion on something, I have to wonder if it was engineered to do so. And if it wasn’t, then I’m staring at the blueprints.
Use two accounts to fake a discussion about illegal immigration between a compassionate republican and a snarling liberal with unpopular views.
Claim to be a black woman, and describe the abusive treatment you’ve received from all your previous black boyfriends, and that you now only date white men. Reply with other fake accounts that, as a black woman, this has also been your overwhelming experience, and that you’re glad it’s being brought up.
Claim you’re a cardiologist that’s helped low income seniors for the last 2 decades, but will likely have to stop due to Obamacare. Add in jargon to taste.
Most readers will consume it quickly and move on, and their model of the world will have been updated.
This end-of-the-world feeling may not be unique to modern society. I recently finished The Forge of Christendom by Tom Holland, which is about European Christians around the year 1000 expecting the antichrist to appear and the world to end at any moment. Maybe we’ve always been secretly hoping it would end.