Blog styling is a bit weird and for the actual copy I kind of don't get its direction

I feel like anything above 4k is impractical. thats 8,294,400 million pixels and 8k is 33 million. sure thats a near 4x increase, but visually its the exact same. all 8k movies do is allow a higher charge because they can label the movies to be better with their "higher and better" number

It’s how the screen looks paused with your nose up in it.

You could send people 720i over Netflix as long as pausing it would download an 8K still frame; they’d howl how good it was.

I mean, that is in your comment. According to the usability research, more people prefer the other approach. You SHOULD adapt your default style to something more universally applicable.

And, no, I do not want the text I am reading to be full-width on my 32-inch monitor. I'm jumping between a lot of other sites that need to take up the entire space.

I mean, newspapers have a smaller columnar style than any website and have been around for generations.

love how frictionless the site is, really good ux other devs should take note

This is a cool idea but surely you can use the (I think its) octo API to send the issues automatically instead of making me open github?

Thanks! Using the Octokit API would require users to authenticate, which adds extra friction. Redirecting them to a pre-filled GitHub issue feels more transparent and secure. They can see exactly what they’re about to submit, and they don’t have to sign in to any third-party app or website. Just one simple click to say hello.

I'd recommend (if you can afford it of course) to take some time off and see if you can work on something for yourself without a deadline or someone in your ear, just peace and building. Like a structured hobby.

Interesting read. I’ve tried Alloy and Dafny for verification before. Seeing how this integrates with real code would be useful. Does it handle concurrency or just sequential logic?

Thanks a lot. It does handle concurrency.

https://fizzbee.io/testing/tutorials/quick-start/#parallel-t...

Sequential logic is generally easier to test (also concurrency testing of linearizable systems). FizzBee specification language is created primarily to express concurrent behavior of non-linearizable systems - like eventual consistency, etc.

Nice work. I’ve been frustrated with how closed off location history tools have become lately. This looks like a solid step toward giving people real ownership of their data again. Definitely checking this out.

Thank you. Yes, I feel the same!

what’s interesting is how every few years we circle back to the same conversation with new names for old problems. the industry keeps rebranding the same complexity as innovation, but it’s mostly the same tension between abstraction and control. every new framework promises to simplify things, but each simplification hides an entire new layer of assumptions that developers eventually have to learn anyway. maybe that’s just the cost of building software at scale now: we’re layering human preferences and historical context into the codebase as much as we are logic. in that sense, modern stacks aren’t complicated because of bad design choices, they’re complicated because they’re living artifacts of collective compromise.

I think in some places AI's arent strict enough, its all an imbalance

This is also true.