Sounds like the system needs tweaking to be less disruptive. Maybe with some actual rules on usage and reach? Especially restrictions on alerts during sleeping hours

My cat was a nut. We had this game where he would immediately jump into the most interesting-smelling grocery bag when we got home, and then expect to be carried around like a furry watermelon lol

Wow, yeah, 380V / 1.2A can absolutely be lethal if electrocuted… even in healthy adults. The most dangerous path for the electricity is from the hand or arm, as it can pass through the heart; which of course is exactly where you would grab a person with the glove.

The cognitive friction of a slow PC is like sandpaper to my brain

So much room for activities silliness!

Yeah I used it to manage my finances. In my banking app it just shows up as:
Amazon $24
Amazon $12
Amazon $16

Before, I checked the emails to label and tag the transactions according to what they actually were: electronics, furniture, etc. Now I have to log in and track down that one hidden page that splits your purchases into actual transactions instead of one infinite scrolling list of items, because Amazon will group the purchases seemingly arbitrarily based on it’s own shipping algorithm.

They made the entire UX hostile just to spite A.I. users. Not the first time they’ve done something similar; I’m getting pretty tired of their shit. In fact, I think I’m going to try some alternatives in my country and see if they’re better to use.

*Gives A.I. agent access to credit card*

I need a tungsten cube

Did you actually look at how it’s implemented in the linked GitHub thread?

How it works is it takes the child view, and stuffs it into the password field. iOS automatically hides the password content (the dots) in screenshots, so by stuffing the view into there it gets hidden in screenshots. I then added the logo, and absolutely positioned the follow button on top of it. when the button is hidden, the watermark is revealed!

Absolute positioning layering hacks based on a quirk of how iOS currently handles child nodes of password fields when the dots would push content out of the parent node in screenshots. The filename of this feature is literally called “growth-hack”. I dunno what else to tell you man, but if you like it, I’m happy for you. But it’s clearly bad code.

As for the second part of your comment; normal users don’t look at source code. That argument means nothing in this context.

The consequences could be anything, that’s the problem with unintended usage of features. Could be an ugly black box drawn on screen. Could be nothing. Could something subtle like the follow button doesn’t work anymore because the invisible logo now blocks tap events.

This is exploiting an edge case behaviour of a security feature. At a programming level, this is bad code, and will break at some point, possibly in an ugly way. I also don’t like the idea of apps being sneaky and hiding behaviour from users.

I use Voyager, and there’s a button “share as screenshot / image”, which allows you to include a watermark, and shows you exactly what the end result will look like. Just do something like that instead?

Yeah I feel like these inventions such as pervert glasses, smart doorbells, and now this, is to get as many cameras running 24/7 around the planet as possible. The best part is that consumers will subsidise them

What’s the bet the logo is A.I. generated as well?

I wonder if at some point we will just get a completely A.I. interface? Like no buttons, no text fields, nothing for people to worry their pretty little heads about and get confused with. You have to describe what you want done to the A.I. who will do it for you. “You missed a comma just after the word ‘backwards’ but before the ‘the’. No not that one… yes you got it. Can you undo your previous change that you did by mistake? God fucking dammit Gemini. Ok I’m sorry, can we keep working?”

Also Paint it Black

This is specifically for neighbourhood-level news. Think those Facebook groups people report lost dogs and cats and plan events.

Chicago has a relatively rich nonprofit news scene that includes robust neighborhood-level news coverage.

They hope to build a participatory community resource that informs everything from where to volunteer, to plans for the weekend, to how neighbors can meet one another.

I’m not sure Lemmy has all the features they want out of their platform. They could host it, but it sounds like they want to verify membership, charge fees, have calendar integration, etc. Lemmy is decentralised, it sounds like they want to build something centralised.

It has many of the hallmarks you picture when you think of a social platform: a feed, posts, reaction buttons. The feed is ordered reverse chronologically, not algorithmically, and can be filtered by topics like arts & culture or pets. There’s a separate, second feed that features posts the community steward has flagged to bring to members’ attention. Posts might be introductions, or events, or requests or offers of help, or steward-posted discussion prompts or guides. There’s also a calendar feature, where users can submit or look for events.

No idea how they plan to monetise this. Also of course they have an A.I. clanker moderating all the posts, but at least they have an escape hatch.

Posts are automatically moderated by an AI-powered tool that checks them against community guidelines, though posters can also flag posts for review by the human steward.

If the system thinks a user’s message may not meet community standards, like it’s inflammatory or obscene, they can’t post it right away. However, users always have the ability to go around the automation and have a local human steward review their message. This is important because the AI might not be correct, and it gives appropriate agency to users and stewards.

People love llamas

Andon Labs, which tests the limits of AI agents, gave Luna a $100,000 budget, internet access, and a corporate credit card, along with instructions to open a store and turn a profit.

Luna, built using Anthropic’s Claude models, selected merchandise, hired contractors, posted jobs on Indeed, interviewed applicants, and hired employees for Andon Market, which sells books, candles, prints, games, and branded merchandise. Andon Labs provided support for more difficult tasks, such as permitting, but said the lab tried to be as hands-off as possible.

The store has generated sales but is not profitable.

That’s pretty impressive. It constructed an entire business using only some string, a squirrel, and a business.

Anecdotally, all our young family members don’t trust A.I. at all.

Once, I asked it a question as a joke, (I think it was about "Labubu"s or something I’m out of touch about); they told me it sounds like fake news 😂. I asked the tool if we could trust it, and it said something like “no, don’t trust me”, which was hilarious.

I think kids these days have a good bullshit detector, by necessity.

I’m hoping to get a VR set at some point, but I’ve heard it’s difficult to get working on Linux so I’ve been putting it off. Maybe once the steam frame exists. I looked at the Valve index but it currently costs $4,000 in my country which is insane lol

Euro truck simulator is the closest game I’ve played in that genre. Sometimes I just want to do something with my hands and turn my brain off for a while, and playing a “boring” game is pretty good for that. Think of it like watching re-runs of familiar shows. Sometimes I play House Flipper as well, literally just painting walls and mowing the lawn lol