I'm not sure I agree with Yvain's post.

One issue, with the abortion example:

Moldbug later uses the example of pro-lifers protesting abortion as an example of an unsympathetic and genuinely powerless cause. Yet as far as I can tell abortion protesters and Exxon Mobile protesters are treated more or less the same.

Well, there are laws limiting the ability of pro-life activists to protest outside abortion clinics. There are no analogous laws for Exxon Mobile.

His claim about how social power can't overcome structural power is dubious. Tell that to Mozilla co-founder Brendan Eich or GitHub co-founder Tom Preston-Werner. To be fair to Yvain both these incidents happened after the article was written and it appears he has at least moved in the direction of updating on them.

Also Yvain says:

Social power is much easier to notice than structural power, especially if you're not the one on the wrong end of the structural power.

This is pure BS. Structural power is very easy to notice, look at the org-chart. It is social power, as Yvain defines it, that is much harder to notice.

Which one do you mean, social power or structural power?

I mean power. The ability to significantly influence decision relevant outcomes without excessive cost to self. The statement doesn't care where the power is derived and it would sacrifice meaning to make either substitution.