Genuinely curious if you've used "plan mode" (with perhaps a plan feedback tool) to get clarity from your coding agent before unleashing it on a feature like "add a pods view with live updates"?

Getting a plan isn't a panacea but is a better way to limit downstream slop than just vibing without one.

The Doge had no power.

Welcome to my rathole: Venetian history. To expand on your thesis, the Council of Ten was the executive power in Venice, but the Doge and minor council attended the Council of Ten meetings and the Doge often chaired those meetings. And even if he was powerless, the Council was made up of other old men. I don't want to push this whole argument, I'm not recommending we adopt the Venetian constitution. But I don't think you can blame current chaos on age.

There is another variable, the rate of change. When a Venetian Doge was 80 years old, the world was.. more or less the same as when he was 20.

Contrast that to now, where an 80 year old in his twenties likely did not even have colour TV.

Can you explain how is the example of Venetian republic relevant for something like a massive multi-cultural, multi-racial federation in 21st century? One that's led by 100s of old people not just ten rich family heads?

I clicked through the $100k credits link and didn’t see Cloudflare listed as an Atlas partner? (Maybe not updated?)

This looks interesting nonetheless.

Came here to say something similar. Adults are just as hooked to their addiction feeds as kids.

These are uber-personalized feeds optimized to keep you scrolling to the next item (story / video / post) so companies can show more ads.

"Social media" is a textbook example of a euphemism. We should be calling this what it is: "addiction feeds".

Seconded. Interesting read.

Haha. Underrated comment!

Some folks have been using the term “promptbaiting” to describe these obvious engagement tactics.

Thanks for this coinage. FYI your comment got killed.

Working on that too. Lmk if you’re up for a chat?

yea I'm down - feel free to send me an email ben@benr.build

I salute this artistry

Why are completion tokens more with image prompts yet the text output was about the same?

Some multimodal models may have a hidden captioning step that may take completion tokens, others work on a fully native representation, and some do both I think.

"Thinking" Mode

it doesn't say that anywhere.