SsgMshdPotatoes

Browser tool to get instant database schema images from DBML schemas. There's an SVG version too: change /dbml-to-png to /dbml-to-svg.

Mostly relies on the placement logic, which I'm aiming to keep improving (so if you get a bad placement, I'd be eager to know!).

It has a little DBML reference in the code panel and also a '+' button for quickly inserting tables, relations and enums.

Feedback is welcome!

Browser tool to get instant database schema images from DBML schemas.

There's an SVG version too, just change the /dbml-to-png to /dbml-to-svg.

Mostly relies on the placement logic I've been working on. Should be pretty decent at this point, but if you get ugly placements on your schema, I'd love to know (contact info can be found on the bottom of the page).

A paper that is somewhat related to the question of how well this works is "Do Large Language Models Know What They Don’t Know? Evaluating Epistemic Calibration via Prediction Markets"

There a model gets a prediction market scenario (that in reality has already closed, but not from the model's POV), and it is tasked to predict the outcome AND give its confidence in the prediction.

Conclusion turns out to be "systematic overconfidence across all models". Probably worth keeping an eye on such research, might enable you to make the product better over time as new research comes out etc.

"The entire appeal is that it isn't "human"" - yes, part of the appeal is perhaps that the medium isn't human, but also that the message IS still human, right? And that is the combination I'm talking about.

I guess the question body wasn't clear enough, because that's not what I meant. My bad if so - I just rephrased the part in parentheses to make it clearer.

Maybe I could've worded it a bit better, the question is not whether forums such as HN will continue to exist, but whether/how it is possible to keep them "human", and also what appropriate definitions of that could be.

I thought about this too recently. I guess documenting every consideration along the way would take way too much time (would be longer than the documentation of actual implementations), but one of these days this seems likely to change?

Lol it only took 20 years