Readit NewsIn my experience, next edit is a significant net positive.
It fixes my typos and predicts next things I want to do in other lines of the same file.
For example, if I fix a variable scope from a loop, it automatically scans for similar mistakes nearby and suggests. Editing multiple array values is also intuitive. It will also learn and suggest formatting prefences and other things such as API changes.
Sure, sometimes it suggests things I don't want but on average it is productive to me.
There are only five apps on my phone, out of over a hundred, that use critical alerts.
PulsePoint, if someone near me is having a heart attack
Messages, if one of my kids is in trouble
Health, if I am having a heart attack
Home, if my smoke alarm is going off
ActiveAlert, my fire department’s dispatch notification app, which will tell me where to drive the ambulance if someone is having a heart attack
If I’m in a darkened theater and someone nearby needs cpr, my house is on fire, or one of my kids is in trouble I want the phone to make a sound.
I want someone else’s phone to make a sound if they get those notifications, too.
If it’s time to take their atorvastatin I don’t give a shit their phone better stay shut the hell up.
If someone’s calendar app slipped through the cracks and got permission to issue critical alerts, THAT is the problem, not the fact that a pill reminder app can’t.
edit: though if I remember or see the initial reminder and log it, it obviously won't go off with sound. If it pings, I've basically always already forgotten.
If this is a genuine belief you have, I think you need to break out of your echo chambers a lot more.
I’m exploring building a game around it (not in it) with Kikai, an unholy marriage between zachtronics games and starcraft, where units (and buildings) are uxn machines with devices to interact with the outside world. This allows you to fully customize the whole army and strategy to be whatever you like.
Buildings for example, take a bit of the role of linkers (or even compilers), they are just another Uxn machine with a Factory device. The most basic form of a building and the one that you will have by default is just a memcpy of bytes from a source ROM to bytes in the RAM of the newly created unit. But by investing a bit of time you could for example have pre-made behaviours that you can link on runtime based on that particular match needs. Units also have a radio and can send messages between them. WIth some more code you could have a handler that rewrites parts of the code broadcasted from a building allowing you to deliver OTA updates!
My objective for the game is that it works out of the box as a normal RTS but that once you get the gist of it you can start automating here and there so there is no high cost of entry but there is infinite extendibility.
There is another very interesting project in the same direction that Devine (the creator of Uxn, Varavara, Orca and Nebu, as well as many others) shared with me recently when I explained my project to him: Doldrusidus which is incredibly fascinating. It goes in the same vein, small ships in a multiplayer universe each of them running Uxn machines inside.
Kikai and devlogs: https://marcecoll.itch.io/kikai Doldrusidus: https://desertslug.itch.io/doldrusidus
`git commit --fixup=X foo; git stash; git rebase -i X^; git stash pop`
`jj squash --into X foo`
> can't find better things to do, such that it makes them poorer, or anti-social its a loss
I feel like this misses the point a bit - lost income/sustainability for artists is obviously a big issue we'll be facing, but looking for a performance indicator in an artistic endeavour doesn't really get you anywhere. There's more ways to value a painting than "what the market would pay" and "potential heat output as firewood", right?
> Run asynchronously in the cloud
> cloud
Reality check:
https://huggingface.co/Menlo/Jan-nano-128k-gguf
That model will run, with decent conversation quality, at roughly the same memory footprint as a few Chrome tabs. It's only a matter of time until we get coding models that can do that, and then only a further matter of time until we see agentic capabilities at that memory footprint. I mean, I can already get agentic coding with one of the new Qwen3 models - super slowly, but it works in the first place. And the quality matches or even beats some of the cloud models and vibe coding apps.
And that model is just one example. Researchers all over the world are making new models almost daily that can run on an off-the-shelf gaming computer. If you have a modern Nvidia graphics card, you can run AI on your own computer totally offline. That's the reality.