You're locked out. You've tried everything. In a moment of desperation, you type a hopeful phrase into Google: "ai tool to find passwords." You see ads and websites promising to use the power of artificial intelligence to magically discover your lost password and get you back into your account.
It sounds like a perfect solution, a futuristic key to unlock any digital door. But is it real? Can AI actually "find" a password you've forgotten? We went down the rabbit hole to find the truth.
You're not losing your mind, and you're not just "bad with passwords." What you are experiencing is a powerful biological response. Your brain, under even a small amount of stress, is actively hiding the key from you. Understanding why this happens is the first step to overcoming it.
check this creazy idea : PHARMAICY* brings you research-based drugs to unlock your AI’s creative mind. Feed it modules like Ayahuasca, Weed or Ketamine, and watch it push into new territory."
Thanks, that explanation is clear. Two things I am still trying to picture in practice.
- For ""derived values from DOM on demand"", what do you do when the derived value is expensive or used in multiple places. Do you just accept recomputing, or do you have a pattern to keep it from turning into lots of repeated DOM reads.
-And for bigger interactions like table row selection, keyboard navigation, drag and drop, does your approach still model that as DOM attributes and queries, or do you keep a small in memory store for that kind of state.
The MutationObserver tips are useful too, scoping and attributeFilter feels like the difference between this being neat and this being a footgun. I will take a look at the repo and the book. Thanks
Haha yeah you are right, that is exactly the thing I was tripping over, I had React state just babysitting the DOM for no real product win.
And your approach is a cool mental flip ""Stop syncing, just read what is already true"" that fits perfectly with the whole HTML first then enhance idea.
So, I have two quick questions though, because this is where I always get nervous:
- First, where do you draw the line between DOM state that is fine to trust and app state that should still live in data, like validation errors, derived values, server driven defaults, stuff that needs to be consistent across SSR and client.
- Second, how does the MutationObserver part behave when things get busy. If a page has lots of inputs or updates, does it stay cheap. Do you scope it per form or per component, or is it one observer with filtering.
Either way the 2KB no deps thing is a flex, dropping a link is fair, people can decide if they want that model. I am going to skim the repo.
I used to ship my app with a JavaScript first mindset. It felt fast on my own laptop, but users on weak phones, slow networks, or locked down corporate browsers kept running into sticky pages and half awake UI.
After I finally measured what I was shipping, I realized most of the client code was not adding value. It was just making hydration heavier, breaking accessibility, and hiding simple HTML solutions.
The post walks through the process I used to cut around 80 percent of the JS without going full “no JS”: listing real interactions in human language, leaning on native elements like details and dialog, using bundle analyzers, and deleting dependency creep. It also shows the small performance and accessibility checklist I use now.
PS: Ahead of China’s Yulin dog meat ‘festival’, a new survey reveals most Yulin residents don’t eat dog or cat meat and say a ban would have no impact on their lives LOL
PS: Ahead of China’s Yulin dog meat ‘festival’, a new survey reveals most Yulin residents don’t eat dog or cat meat and say a ban would have no impact on their lives
Thanks for the feedback. I understand that playlists are not the usual type of post here. The idea was more about seeing what people do with a shared notebook space rather than focusing on the music itself. It is just a small experiment to test how people use an open board when they can create their own section inside it. Glad you found the design clean. I tried to keep everything simple and fast.
I opened a shared notebook where anyone can create their own playlist of songs. You can add any genre, any year, any mood. The idea is to see all the different playlists people build in the same space. Just open it, press the add playlist button, and paste your song's link.
I want to see how many unique playlists this community creates.