I think this might be a case of doing something rather than hyperanalyzing every possible situation and doing nothing.

Might not be ideal, but at least he's doing something.

Absolutely. He's doing way better than "nothing".

But you present a false dilemma; it is possible to do some analysis without entering a hyperanalysis spiral of inaction.

Exactly. But by continuing to debate this, we're perpetuating the doing nothing aspect. If you were really concerned about efficiency, you wouldn't be debating this on hacker news.

Compassion is not an algorithm to be optimized.

"Compassion is not an algorithm to be optimized."

What, if anything, do you think you mean by this sentence?

Compassion isn't always about efficiency. Doing something to make a difference in someone's life isn't always about allocating resources most efficiently.

If every single person who wanted to help out just kept saying "But these people need it more!" Then these kids wouldn't get the morale boosts they so sorely need.

People need to do these things. You can't just say "He isn't being efficient enough! Malaria patients need it more!" Because that's just unfair on these kids.

No.

If every single person who wanted to help out looked for the most efficient thing to do on the margin, then there would hardly be any malaria patients. The most efficient opportunities for philanthropy would be exhausted. Then the next most efficient, then the next, and so forth. Maybe eventually we'd get around to giving these kids morale boosts, but if we didn't, it'd be because there were more important things to do.

It's a cliche, but life isn't fair.

I could turn this around and say that what he's doing is "unfair" to malaria patients, who need his help more but aren't getting it.

If every single person who wanted to help out would find a cause, then find a better cause, the world would be a better place.