While I still see a steady uptick in the number of hypermedia APIs out there in the wild, as well as conversations around the different media types that are available, I think that we severely underestimated the time it would take for the average API developer to absorb and accept the concept. When you are immersed in any of the leading formats, from HAL to Siren, and you have the aha moment about why hypermedia makes sense, it can be easy to think that everyone will see the future like we do. In reality, I just don't think people are always seeking wisdom in the same way. They are often just looking to get the job done.
It takes a lot of work to become hypermedia-literate. It's an investment that not everyone can afford. While I am seeing more APIs employ hypermedia, I have not seen an increase in the tooling and definitions we need to help developers speed up their understanding, providing examples that they can reverse engineer. Siren is my hypermedia format of choice and I found that the TV and streaming video API platform Wurl gave me a strong example to reverse engineer and learn from in my own journey.
Maybe not everyone learns like I do, but I can't help but feel like people need more common examples to learn from. I will spend some time going through the hypermedia APIs I've included in my research and generalizing some of the design patterns that are present and publish them as simple examples on GitHub. I need to refresh my own hypermedia skills, as I dive back into my subway map API design, which uses Siren as an enabler of the journey. It will also give me some good stories here on the blog (hypermedia is always an evergreen driver of users, as they Google for things in their hypermedia journey).