Perfect example of a base rate fallacy - https://en.wikipedia.org/wiki/Base_rate_fallacy

What percentage of GitHub activity goes to GitHub repos with less than 2 stars? I would guess it's close to the same number.

> As far as end-to-end encryption, on SM sites (social media or SadoMasochism, however you want to read it) I don't really see the need.

You don't see any benefit to allowing people to encrypt their private communications in a way that can't be accessed by the company?

It's weird to see tech news commenters swing from being pro-privacy to anti-privacy when the topic of social media sites come up.

Many will cheer for any case that hurts Meta without reading the details, but we should be aware that these cases are one of the key reasons why companies are backtracking from features like end-to-end encryption:

> The New Mexico case also raised concerns that allowing teens to use end-to-end encryption on Instagram chats — a privacy measure that blocks anyone other than sender and receiver from viewing a conversation — could make it harder for law enforcement to catch predators. Midway through trial, Meta said it would stop supporting end-to-end-encrypted messaging on Instagram later this year.

The New York case has explicitly gone after their support of end-to-end encryption as a target: https://www.reuters.com/legal/government/meta-executive-warn...

According to a writeup at https://www.chainalysis.com/blog/lessons-from-the-resolv-hac... this started with a plain old hack that compromised their signing key.

They also had a smart contract which didn't do some proper checks, but the hack was only possible with the stolen private key. Whoever held the private key was able to mint a lot of money, unchecked.

So there was a traditional hack at the core of this heist, not just a smart contract exploit.

I thought you were being sarcastic until I watched the video and saw those words slowly appear.

Emphasis on slowly.

> Though iPhone Pro has very limited RAM (12GB total) which you still need for the active part of the model.

This is why mixture of experts (MoE) models are favored for these demos: Only a portion of the weights are active for each token.

> Before I felt mostly invisible but since then got approached in bars all the time, which rarely happened before.

Physical attractiveness is extremely relevant in the context of cold approaches in a dating environment. I won’t disagree with you there.

However getting approached at bars is very different than working with someone in an office setting or having your papers graded in a university setting.

I don’t think it’s that simple. My assistant professor friends build relationships with students and help them work on topics and weaknesses. They get to know students and how to help them on problem areas.

For deeper courses they may help them pick topics to write about and sources to read.

Having that context for the ongoing feedback from grading and mentoring is valuable. Depending on the work it simply might not be possible to do anything blind.

Even without names, handwriting and writing styles are obvious. Even in an office setting I can always tell who wrote something as small as a sign or a note by handwriting or word choice alone.