Givewell, for instance, publishes their findings as spreadsheets and lets you set weights on different aspects of human experience you consider to be good or bad.

This is still a fixation on an individual subjective human perspective. Which is a bit confusing, given that the EA manifesto explicitly leans on Bayesian statistical analysis. The end result is a round peg (perceptions and emotional priors) being shoved into a square hole (hard numerical figures). It also isn't effective as a policy guide, because the layman fiddling with weights on a spreadsheet still doesn't have any actual control over the scale of political economy that a government or a mega-millionaire commands.

This is also just objectively not true and suggests to me you’ve never even talked to an effective altruist.

We're running into a Jordan Peterson line of argument, wherein "you just don't understand my line of thinking" is used to dismiss critiques you're not equipped to rebut.

Can I counter with "You've never even talked to a non-effective altruist?" and conclude you've been too cloistered to explore ideas outside the EA space? Or would you consider that a personal attack rather than a statistically informed observation?

horrible suffering, such as the kinds of suffering caused by easily-preventable illnesses, is much worse than any of the subtle and varied experiences of a good life

This isn't either/or. You can go back to the old Bill Gates plan to mitigate overpopulation in the third world. He initially tried to push out contraception to the local populations of communities he'd hoped lower birthrates would help. Instead, what he discovered was routine vaccination and standard modernized health care drastically reduced infant mortality and resulted in parents choosing to have fewer kids as a result.

In hindsight, we discovered similar patterns of behavior across the US and Europe, Latin America, India, and China. But as a knock-on effect, we've seen the US/EU focus so exclusively on disease mitigation as a strategy for improving relations in countries they wish to ally with that they neglect their domestic populations (who are comparatively much wealthier, but see the foreign aid as coming at their expense). The iterative result has been a series of claw-backs of positive disease mitigation policy fueled by a popular media that's vilified the very act of disease mitigation and denigrated the people who received it as subhuman. And the true irony of the affair is in how many of these popular media institutions are owned and operated by self-proclaimed EAs.

The EA strategy of trying to decouple and distill policies into their individual components, then min-max solutions at a spreadsheet level, have produced a backlash their narrow focus failed to anticipate.

I think this should be taken evidence against charity being rooted in the need for a popular buy-in.

Until the AI wonderkin can fully divorce themselves from the public at-large, they're going to need to rely on human labor and ingenuity to accomplish large, complex projects. The strongest card that EAs have to play is typically their ability to quickly roll up a highly educated, multi-talented workforce underneath them. Even then, they're notoriously inefficient in their application of these skilled technicians.

But we're already seeing the results of the Bullshit Jobs and Bullshit Bosses, as the bigger Tech companies stumble through the 2020s. Without people who want to work beside you on a project they are deeply invested in, the work slows down and the work product becomes flimsier and more ineffectual. In the end, you're left with Bloomberg 2020 tier work, where you've got tens of thousands of people collecting a paycheck to do nothing.

I’m advocating for socialism, not charity.

Socialism requires a popular consensus to function. You can't impose a collective project by executive fiat.