Huh, this sort of issue is what made me leave KDE in the first place. Haven’t had such problems on gnome.

“what are you saying? That I can quit vim?”

“no Neo, what I’m saying is - when you’re ready, you won’t have to.”

Audio communication is too slow. If Humans can’t evolve to communicate telepathically, language itself should evolve to account for this. Here’s what I propose:

• we develop a mathematical formula to generate a fingerprint/signature for sentences we intent to say;
• before saying anything, we calculate this fingerprint in our head, then say this fingerprint result, followed by the actual thing we want to say
• the listener then gets this fingerprint result and keeps in mind; whenever it tries to predict what the full sentence from the speaker is going to be, it calculates the fingerprint for the predicted sentence and compares it to the fingerprint received at the start.
• if the fingerprints match, then the listener reports: “I got it” and the speaker can then skip saying the rest of their sentence.

Surely this is bound to improve communication for everyone and would have no downsides whatsoever.

(Sorry, the amphetamines must be kicking in right about now).

I somehow managed to get pretty good at getting a good feeling of “did this person understood what I want them to understand?” and I adapt my level of overexplaining based on that feeling.

It’s something I wish other people did to me as well, as I hate it when people keep talking more to make the same point I already got. Tbh sometimes I even wish people would stop mid-sentence if I already autocompleted their sentence in my head.

I remember at one point the front-end guys I knew were laughing that it didn’t even support iframes. But I imagine it eventually got decent enough.

Meanwhile in my company the leadership just thinks that we have a messaging problem after the new AI stuff we implemented made absolutely no difference in the sales numbers.

Windows Vista is Microsoft’s greatest success, because it’s main purpose was to make people forget the promises made for Longhorn.

AI is not the new NFT but also not the new Internet. It’s the new touchscreen. Amazing in some contexts, but forced down on every other.

I think both of them have Japanese (I remember seeing Rosetta Stone being praised for its Japanese content 20 years ago and I hope it would only have improved since), but I haven’t gone very far in the language in either app.

As a complete beginner, Drops is pretty good for learning random words and increasing vocabulary. As you advance through it you start seeing sentences too, but it doesn’t teach you how to make your own sentences, only to memorize the ones they pre-created.

Rosetta Stone doesn’t translate anything. All of the content is in the language you want to learn and it tries to introduce you to things in a natural way. For example it shows a picture of someone biting an apple and says “the man eats an apple”, then later shows other pictures related to one or multiple men, fruits and verbs, so you can get used to the differences between things just by observing those.

Sometimes the icons annoy me too and I wish the app had an option to always show the icon’s label, but at least you can tap on the icon to see the label.

Check out “Language Drops” and “Rosetta Stone” if you’re looking for replacements. They both have very different approaches to language learning (both from each other and from Duolingo), but their content is at the very least much better curated than Duolingo’s.

I haven’t gone out of my way to check but AFAIK neither of them is jumping on the AI-before-anything-else train.

While I knew since last year the the story was fake (and I had random people mention it to me on online games when they learned I was Brazilian), I’m also skeptical that this story alone was all it needed to kill that guy’s company.

Monster.

There’s some difference in the fences on the left side at the exact time the car passed by on the other lane. My guess is that the timing of the other car made the software interpret those changes in the input as something moving instead of simply something being different.

There’s a typo there so I think it’s fake.

So I guess it’ll probably block screenshots too?

There is no customer involved.

1. They place an order and pay for it
2. They force the order to be assigned to one of the drivers
3. Driver claims the order is delivered
4. Doordash pays the driver
5. They overwrite the data so that it goes back to step by 2.
6. They “deliver” and get paid for the same order hundreds of times.

Books were probably red, but it takes a while to identify something like this.

If you actually think about things and form your own opinions you’ll usually be treated as “the other side” by everyone who signs and follows any pre-made set of opinions.

If you hate AI but thinks there is some specific situation in which it doesn’t 100% suck, you’ll be treated as a troll in anti-AI communities. If you’re MAGA but disagrees with anything Trump says, you’ll be called a leftist in conservative circles. If you’re a fierce active defender of LGBTQ+ rights but thinks it’s OK for a white American to dress up as a Mexican character for Halloween, you’ll be ostracized in many left wing groups.

Disagreeing is often treated as trolling by those you disagree with, depending on the subject. Mostly because those disagreements are often bad faith talking points from some groups of people.