So, I'd say my Google usage is split maybe 50-50 between looking for a thing and trying to answer a question. When I am eagerly anticipating an answer to a question, I try to take the time to make a guess first.

It (ideally) keeps me calibrated, helps me learn how to make better guesses.

I think this is a very good policy, but that's because I value the type of understanding that lets me make better guesses. But I think there's something under-examined about that. Do you need good guesses for questions you can just Google?

So, obviously, the answer is no. Surprise! That's right, you might have guessed, in keeping with the theme, that I'd talk about how good predictions help you understand and critically listen to new things, and help filter out nonsense as well as bring a more coherent understanding to things you're still learning.

I'm sure that's all true.

But you don't need it. Obviously, you can just Google it. Get a data plan for your phone. And today, for reasons unknown to man or beast, I've decided to stick to the strict literal meaning of the terms, despite this clearly being coded as advice and not, ya know, an actual description of systemic necessity.

Because my job as a writer is to provide information, and new information, to an active guesser, should feel like surprise or a new tool for their tool belt. This was way too simple to really take effort to explain, so I thought I'd surprise you a very small amount. Or at least do something that was clearly meant to be a surprise – then, you, having read without making any guesses, suddenly learned you did it wrong by not considering the text more carefully.

I think some writers optimize for surprise, and that seems like it stinks, but I feel I should write at least 300 words so the thought doesn't seem totally insubstantial – despite not being persuasion so much as admonishment for carelessness.