I think one thing missing is that these hype cycles can be stacked on top of one other. The Internet isn't simply one cycle; there's two other cycles including cloud computing and mobile.
Blockchain goes through a few. First with Bitcoin as a currency, then things like contracts, then NFT. It likely fails at some - we're still not seeing Bitcoin as a currency. But the overlapping cycles tend to mess with it; people who get excited over NFT also get excited over Bitcoin.
IMO Peaks can only occur when a tech has gone past the Early Majority market into Late Majority. Something can still peak while it's in Late Majority, but never earlier.
How do we measure this? I normally assume HN is in the front of Early Majority. I'd say universities and Fortune 500 are at the end of Early Majority. People who hang around TikTok and Discord are at the start of Late Majority. FB users are at the end of Late Majority. That relative who doesn't use Internet and gets all her news from TV is in the Laggard phase, though arguably not in the market at all. I could be wrong though, these are just random samples from the market I'm in.
Ride hailing, banking apps, SMS, and GPS maps are in the Laggard phase. In my country, QR payments have just entered Late Majority.
ChatGPT and Midjourney is probably in the Late Majority phase - that rando comic artist who still draws for newspapers is already playing with Midjourney.
Most people don't know about GPT-4, though. That's another hype cycle. Asking GPT via photos is still in the Early Majority phase, so it'll likely go up a lot higher. GPT-4 can answer exams and take home interview questions via your phone, but most exams aren't built against this.
