It definitely gives me pause. ESR clearly doesn't know when he's out of his depth (classic Dunning-Krueger).

I'm not enough of a programmer to judge his programming texts, though I am enough of a sysamdin to find his Unixy sysadminish stuff generally valid.

I've found CatB itself aging poorly and question a number of the assumptions behind it, particularly as concerns anthropology. It seems shaky. Though I think the general principles behind Free Software and the open source model have their merits. Just, possibly, not quite those ESR describes.

ESR expressed views on all sorts of items which are unknowable and much debated; programmers cling to this idea that there's a right and a wrong (protobuffers not JSON! One True Way vs TMTOWTDI! Emacs vs Vi! JavaScript is a reasonable choice etc etc). Generally there's not, there are just ideas and opinions without hard data. Systemd contradicts pieces of the original article.

So you tell me: how would we know if his technical work stood up to scrutiny? If I have experience that agrees, does that mean it does? What if my experience contradicts IT?

Indeed, this is the critical question. If L. Ron Hubbard secretly but accurately predicted the lottery numbers for last week, it doesn't mean we have to go back and change them. Things can seem wrong/impossible/against your worldview, but that sense doesn't help quite so much as _just looking_.

Fundamentally, calling things into question has little value until we generate an answer to the question it was called into. Considering it's relatively easy to judge him on the technical work, why not?