@coffeeaddict @rudytheelder

I've thought about that like every time I hesitate to boost something lacking an alt text. But:

- I bet the bot wouldn't be able to make head not tail out of an image like the stack of spheres,and would certainly miss the significance of there (probably, I didn't confirm) 2024 spheres in the stack

- Half the images I see on Mastodon are visual puns otherwise meaning-doubled

So it would be challenging. Plus the models would be expensive to run on a Mastodon server.