That's a really bad way to go about usability. It's pretty frustrating you're looking for a feature to have to hunt for it like you're playing a point & click game. At that point I wouldn't even call it a feature, it's an Easter egg.
> If it's not obvious, it's probably not a critical feature, just a nicety for those who discover it.
There is a huge difference between discoverability and hierarchy. If the user can't distinguish between an interactive element and a static label, the UI has failed. But there is definitely nuance in the degrees of hierarchy and importance. If every button looks the same with every affordance necessary for a user to recognise it as such, it might be great in terms of discoverability, but in most cases it would still be bad design because as a user I can't easily distinguish between primary, secondary or tertiary actions.
This is of course also highly dependent on the business goals and overall context of the design. If you have a marketing page you might want to clearly distinguish actions that lead to the sign up or download of your product (this – in most cases – is what you would like your users to do and also what your users eventually want to do themselves). If you have enterprise software that is highly customizable and completely different for each and every individual, you might want to shy away from opinionated hierarchy of importance. In this case, it would be very much dependent on the user and therefor their choice.
Typically the context is – at scale – more on the former side than the latter. If I have, for example, a notes app I would say the hierarchy favors the creation of new notes rather than going into settings to change the appearance of my notes. That doesn't mean it shouldn't be discoverable, only that the design should clearly dictate what's more important (especially on a temporal dimension: I'll add notes more often than I would edit the style) and therefor influence the appearance of both actions.
Let me know what you think.
The problem is that design is always a compromise between discoverability and post-discovery efficiency. Consider the swipe-left-to-delete pattern. It's not discoverable, but once you've learned it, you wouldn't want it any other way. Designers sometimes compensate for the lack of discoverability by adding a slower but more discoverable way of doing the same thing, like an "Edit" button that reveals a delete button on every item. Thus making swipe-left-to-delete a non-critical feature, and a nicety for those who discover it.
Another example is clickable usernames. We have an unspoken expectation that clicking a username in any app takes you to the user's profile. This expectation is so ubiquitous that apps (like HN) don't provide any additional indication that usernames are clickable. Imagine if every username on every app was a full blown bordered, textured button. Even someone with zero design sense would feel something's off.
It simply isn't possible to maximize both discoverability and efficiency at the same time. Every feature needs to decide who it wants to cater to and who it's comfortable leaving behind.