Prophet

The blockchain is essentially a ledger that tracks transactions (including the creation of currency). One thing that is not always clear is how important it is for a blockchain to be decentralized. When I say “decentralized,” I mean that many different people are operating a server that performs transactions on a larger network. These people are rewarded in currency for their efforts, and are sometimes referred to as “miners,” though this term is changing somewhat.

There are thousands of these servers in a network that are operating on and tracking the ledger for blockchains like Bitcoin or Ethereum. Any updates to the ledger are verified by all of these nodes. As long as 51% of nodes can verify a transaction, it will be added to the ledger. This means that as long as someone doesn’t own 51% of the network, they can’t just inject whatever transactions they want (i.e., fraudulent activity). In practice, this makes these networks very resilient to fraud.

I think this paves the way for a lot of the practical examples you’re looking for. For example, there’s no way for the network to decide to just give tons of money to a single entity for some “economic policy” like Too Big to Fail (i.e., corporate bailouts). This means you don’t have to wake up one morning worrying about whether or not your currency will rapidly inflate because of things like corruption. Another example is the true ownership of digital assets. NFTs have (rightly) gotten a lot of flack for being overpriced JPEGs, but there are real use cases here. A random middleman can’t just decide to price gouge because they own all the tickets first (Ticketmaster). Instead, artists can mint tickets on the blockchain (very important: this ensures authenticity) and then fans can buy them on the blockchain - no middle man required. You still show a QR code at the door for verification like you would now.

It is entirely job dependent. I have been in jobs where it was just a grind and going the extra mile simply put a smile on my boss’s face. In jobs like these the best thing you can do is carve out as many hours as possible during the work week to build new skills or apply to other jobs. I’ve also been in jobs where going the extra mile directly contributed meaningful skills to my resume/portfolio and helped me get a new job with way better pay.

I would add Pinta as another drop-in replacement for MS Paint

Lexmark seems like such a niche thing to put here but I know about it because it’s based in my hometown. Also it deserves to be placed next to the soyboy.

The guy who leads this group is extremely vocal (almost weirdly so) about white privilege and systemic racism. He is also white. It’s true that many AI models have white-bias. The reasons for this are multi-faceted. Our datasets are grossly imbalanced against racial minorities. I also think I understand that for some darker-skinned races, it is more difficult for the model to extract relevant features from the shitty Flickr photos they scrape for these models.

That said, injecting words into the users prompt to force the model to generate minorities more often is an extremely naive approach. Kind of like if Google added “reddit” to all searches just because it worked for some specific test cases, but ignoring that you now no longer get any site except reddit. Probably the solution here looks like paying a lot of money for high quality datasets as well as investing in user education and more AI explainability of these tools.

So why is palworld seemingly so much more successful? Niche games like these mean nothing if they don’t appeal to a broad enough audience to be successful. I would have initially said the same thing about palworld but now it’s the #1 game on steam of all time and in hindsight it is so easy to see why.

How does one even fix this? Is it just dragged out?

I agree, in the context of the tweet, that purchase history is enough to build a working product that roughly meets user requirements (at least in terms of predicting consumed items). This assumes you can find enough purchase history for a given user. Even then, I have doubts about how robust such a strategy is. The sparsity in your dataset for certain items means you will either a.) be forced to remove those items from your prediction service or b.) frustrate your users with heavy prediction bias. Some items also simply won’t work in this system - maybe the user only eats hotdogs in the summer. Maybe they only buy eggs with brownie mix. There will be many dependencies you are required to model to get a system like this working, and I don’t believe there is any single model powerful enough to do this by itself. Directly quantifying the user’s pantry via vision seems easy in comparison.

Also quite difficult from a vision perspective. Tons of potential object classes, objects with no class (e.g., leftovers, homemade things), potential obfuscation if you are monitoring the refrigerator/cabinets. If the object is in a container, how do you measure the volume remaining of that substance? This is just scratching the surface I imagine. These problems individually are maybe not crazy challenging but they are quite hard all together.