Well, all of the questions are binary so they either happen or don't happen. They may not be sampled from some "objective" distribution, but you can still assign subjective probabilities to them. Just write how likely you think they are to happen.

Swedish perspective: It's pretty funny seeing these biscuits praised as some kind of perfect recipe/secret treasure when it's one of the most common biscuits here. And when I say common I don't mean that they can be found in stores or cafés (though they sometimes can) but that it's something many people make at home. I think the reason that they aren't more common in stores is that they taste much better when they are warm; from the oven. The stores can't compete with that.

The recipe I'm used to is pretty similar to yours (from the famous Our Cookbook) but without the ginger or salt.

So instead of a disclaimer saying that a tweet is false, we'll now have a market saying that it probably will be declared false in the future. Then later the tweet will be declared 100% false and the market would close. But I don't see why you would trust the final result any more than the disclaimer. If you don't trust the social media companies then the prediction market just becomes "what people think social media companies will think" which doesn't solve the problem.

Edit: I missed that future users would vote to decide what the true outcome was but my point still stands. The prediction market would become "what people think people on social media will think". I know there has been work on solving this (Augur?) but I haven't read any of it.