Ask HN: Did APIs Kill the Semantic Web Dream?

No, in fact, APIs are one of the few ways the Semantic Web lives on, even if most HTTP APIs serve content that's not formally typed with intrinsically meaningful mediatypes [1], and that's not linked to other resources with registered rel types [2].

Nonetheless, the data is still pre-structured, and is many steps more reasonable than the alternative, where the consumer scrapes HTML from the webpage and hopes for the best.

Limited time and lack of business incentives contributed to this end. Design and onthologies are hard, and when the only reason for doing so is to interoperate with your competitors, and if and only if they do the same, then no one wants to spend the time. It's also clear that elements of the Semantic Web live on in areas of life that aren't so subject to these pressures: government datasets, academia, library science, and hobbyist efforts.

[1] https://www.iana.org/assignments/media-types/media-types.xht... [2] https://www.iana.org/assignments/link-relations/link-relatio...

I don't understand the question. How does one get "a web of data that can be processed directly and indirectly by machines" without some sort of API?

Wasn't it issues like https://en.wikipedia.org/wiki/Semantic_Web#Challenges which killed the Semantic Web dream?

API didn't kill the semantic web, but the possibility to use API that does not implement standard that push semantic forward.