Well said. I'm gonna steal this explanation.
Also reminds me of the famous Carbonara quote: "if my grandmother had wheels, then she would be a bike" [1]
5,638 karma · joined February 24, 2020
Well said. I'm gonna steal this explanation.
Also reminds me of the famous Carbonara quote: "if my grandmother had wheels, then she would be a bike" [1]
Let's imaging I have an API. This API tells me how much money I have in my bank account. One day, someone hacks the API to always return "One Gajillion Dollars." Does that mean that "One Gajillion Dollars" spontaneously emerged from my bank account?
ToM tests are meant to measure a hidden state that is mediated by (and only accessible through) language. Merely repeating the appropriate words is insufficient to conclude ToM exists. In fact, we know ToM doesn't exist because there's no hidden state.
The authors know this, and write "theory of mind-like ability" in the abstract, rather than just "theory of mind."
This is a cool new task it ChatGPT learned to complete! I love that they did this! But this is more "we beat the current record BLEU record" and less "this chatbot is kinda sentient"
> Is this really necessary?
> Let’s just be safe
"Look! I built a whole App that does the thing! Isn't it cool? Oh, yea, you need to use this gnarly url and - crap - yea, use chrome. Refresh it... Hold on, let me restart the Raspberry Pi... Got it! Look at that! So cool! Wait - did someone HACK this?"
The grown up version of this might be just building for one platform, and one happy path. Who cares if lots of people can't use it? If it's popular, you can turn it into a job then. But until that happens - just make the computer do something cool!
Sounds like a great sibling gift tbh
Is this something in the same problem space as ChatGPT? I'm not sure an LLM would be well suited for that, but could be wrong.
https://www.reddit.com/r/Fitness/comments/37ylk5/a_linear_pr...
It’s hard to give advice without knowing why you feel overwhelmed. Is the job crazy hours? Too much new stuff to learn? Different than you expected?
How are your colleagues - supportive, engaged, helpful? Or aloof, focused on their own work, perhaps critical at times?
What’s your feedback at work like - positive + encouraging, or negative and highlighting weak spots?
The solution could be anything from “see a therapist” to “be more proactive” to “this is just what imposter syndrome feels like.”
- They have the worlds largest messaging platform (WhatsApp) that is basically unmonetized rn.
- They are one of 3-5 major players in AI. Amazon was one of the 3-5 largest internet companies... and then created AWS. (e.g., they released-then-unreleased a GPT3 clone before ChatGPT, because they got slammed in the press for ethics. doesn't mean they can't build a B2B business with it.)
- The regulatory landscape is shifting towards tech nationalism. Yea, there'll be regulation, but nationalism will limit the scope + create a moat against upstarts.
Recessions are great times to build new businesses - if you have cash on hand. They have cash.
I find this helps a heck of a lot with maintainability + debugging as well
The first censuses in the US were basically "a dude rides into town, asks how many people live there, then leaves" and got surprisingly close to recent estimates.
I just wonder if this is a matter of "there's error everywhere, it's basically stochastic, the # is close enough"
I totally agree that "people living together in a small area" can pretty much do what they want. The only issue is when that starts to screw over other folks, and when all the other small areas get together to do the same.
Then it's pretty reasonable for the state to go "you can't just never build houses. That's a decision that screws the rest of the state" and enforce that decision
An open-minded willingness to change one’s mind is admirable. But if you change your mind too often, you end up thinking less yourself - you’re outsourcing learning. It should get incrementally more difficult to change your mind the more you learn. Otherwise, you’re just collecting facts, not actually understanding anything.
Besides, you don’t really hope you’re wrong, and it’s a bit arrogant to say so. Have you ascended above these petty emotions of the flesh? Being wrong is A Bad Thing, and “I believe I’m wrong” is an oxymoron.
I think a better approach is “I’m always willing to change my mind, and hope I never have to.” I’m often wrong, about things of which I’m unaware, and I always pursue the opportunity to be less wrong.
But it’s a worse title.
If we purposefully impoverish black people (e.g., via Jim Crow), the destitution of their kids isn't _just_ an economic problem.
And note that this is the literal case: I know a bunch of poor older folks whose parents struggled to get good jobs because of Jim Crow policies.
Like, are you telling me that NYC landlords in the 90s weren’t trying to scam their tenants? That Philly police in the 80s were clean, efficient, trustworthy public servants? That there was less fraud in the .com boom, or before the Great Recession, then now with AI-SaaSchain nonsense?
I used to go to a burger place where the cashier would just tell me to fuck off if I ordered too slow. I can’t imagine trying to tell that dude he forgot _napkins_ in my bag.
I totally get the urge to rant too. People suck. But don’t lose hope! Because people always sucked, and it’s nothing new.
This reads like a board scared of a hostile takeover, which seems reasonable given geopolitics rn
https://www.investopedia.com/terms/p/poisonpill.asp#toc-what...
Every bodybuilder knows that you have to push to failure and rest a lot between sets. The term for low-intensity or unrested lifting is “junk volume.”
Every runner knows that 80% of your volume should be easy. It’s called “80/20 training.”
Every CrossFit bro knows you need both intensity and volume - and maximizing that over time is what’s important.
They all work, but in different ways + for different reasons.
There’s lots of studies that try to break the mold. Usually, they’re limited by the impracticality of a large N, or perfect adherence, or exogenous factors, or avoiding hype. It’s a limitation of the field. It just means things will go slower. Don’t worry too much about any one study.
We’re building more cheap crap, sure, but also a few rare, well-built buildings.
The best advice I got was to solve one problem at a time. I went all in on getting a job. Then getting an apt. Then getting in shape. Then reconnecting with friends. Then dating again. Then quitting smoking again.
I definitely had to lean on family/friends for help. Especially my parents, who had their adult son back in their basement for a few months, and a few friends/former colleagues, who worked their networks on my behalf.
> I have expertise in Ruined Life Reclamation
In hindsight, the person who gave me the best advice definitely did have some expertise themselves.
But still, I was surprised how much of “how to make people like me” was pretty explicitly learnable.
But your point does still stand. If you’re happy where you are - you have “other, better options” - def doesn’t make sense to invest in something where you’re not gonna get the return you want.
Did you notice that they’re talking to a new grad?
“I can’t believe a young person would suggest youth oriented thing to another young person!”
(0) Reach out to everyone. Everyone. That old college acquaintance you just noticed is in your city? Get coffee/happy hour. You never know who's looking for more friends.
(1) Encourage people to invite/bring their friends when you organize something. Hit up one friend for happy hour/food. "You've mentioned [coworker] a few times before - would he/she want to join us?"
(2) Host (implied) large group gatherings. Easy to say "come over, and bring whoever!" I like sports, so I host for the super bowl, World cup, March Madness, etc. - but could be anything. Throw parties for other peoples' birthday/going away/coming-to-the-city. Advance level: some bars are often open to "hosting" an event - they'll just rope off an area if you promise X people will show up.
(3) Host recurring events. Board Game nights, poker nights, dinner parties, wine club, etc. Great way to deepen relationships.
(4) Start group chats. Much easier to maintain a relationship with 4 group chats than 20 individuals. If one can't get critical mass, post the plan in another.
(5) Be actively welcoming. Make a point to intro yourself to new people/social stragglers on the edge, and greet people by name when they show up in a new-ish space (e.g., my friend brings his new gf to a board game night; she doesn't know many people here; stand up and say "Hello Janice! Great to finally meet you. How has your week been?"). Forge connection, don't just co-exist in the same space.
I'm sure there's more...