I spent 2 hours at work yesterday reorganizing the walk in fridge to work with our new menu effectively and cause we were short someone the day before and it becomes a mess whenever im off. I fucking love organizing stuff. Spending 2 hours downsizing shit, getting rid of shit we couldn't use anymore and placing it all in logical places is what my brain loves to do, it feels satisfying to finish and see how much better things look and everyone else at my work is so fucking bad at it that it makes me seem amazing

Musicbee

the one thing i miss from windows

I put videos into music, music into pictures and pictures in video folders

ah yeah, going into my old Downloads folder, going through history like a digital archeologist:

"Hmm.. a 2020 download of that old HOI4 mod..? How'd it get here..?"

Kids for the past decade plus go through school using google drive. Highly searchable. Teachers believe that kids are "tech savvy" because they know how to use tablets. So the entire process of teaching kids how to save, organize, find, and open files is gone. There are zero consequences for not knowing it either. But collages have noticed this trend for some time. Professors asking for project folders and discovering their students have no idea what they mean. It's not universal, but when the windows monopoly ended for schools, it meant kids were learning a whole different way to manage files.

Assumptions that they only know how to use a phone/tablet rather than a desktop/laptop computer, I guess. I feel like I hear most of it from their former teachers/university professors.

You can turn unstructured data like strings and images into vectors, which are 1D arrays: [1, 4, 8, 2, ... 8, 3.5, 9, 1] is a vector. Each number in a vector is basically a score of how close that original data is to a semantic concept. A picture of an apple and the string "apple" both score 1 on Apple and highly on Fruit and a zero or something on Mineral. Vectors can be any length really, so any one vector can define the proximity of original data to tens of thousands of concepts. An orange scores high on Fruit, but not Apple, so in a vector search, if you search fruit you'll get both apples and oranges. If you search Florida fruit, you'll get oranges but not apples. Search pie recipe, apples show up but oranges don't. And so on. Vector searches will retrieve things that are semantically related.

You fill vectors by training computers to sort the data themselves. You train the computers by exploiting tens of thousands of third world workers to manually categorize information and double check the computers until the success rates are high enough from the automated categorizers to fire the workers.

A very simple vector example:

Assume we have an array of 26 binary variables, one for each letter of the alphabet. We flip to 1 if the letter is present in the word. Dog -> [0,0,0,1,0…] Cat -> [1,0,1,0,0…]

Then we can do a search by taking a target word and doing a cosine similarly search, roughly we find the binary array most similar to a given array to compute a score.

Modern models compute a much more complicated vector by using the context of the words around them. Multimodal image models use a combination of images and text to train a model so later you can use pass it some text to get a vector.

There are some things out there about doing math on vectors because the form a latent space, for example: ‘king - man + woman = queen’ but empirical tests show that this doesn’t quite hold up on modern models. 3blue1brown has a video on vector math which is worth a watch.

while this is a funny comment, my cat died and so i've been rewatching avatar and my god the earth kingdom stuff is straight up sinophobia, jesus christ, cant westerners ever be normal?