For most of history, the constraint on building things was the building. Making was hard, slow, and expensive, so the people who could make at all had an advantage just by being able to. That world is ending. Tools keep getting more powerful, the cost of producing keeps falling toward zero, and when everyone can produce, production stops being the advantage. Something else has to be, and it's worth thinking carefully about what.
What's left, when making is cheap, is judgment — knowing what to make. Call it taste: the ability to tell the good version from the merely possible one, to feel that this is right and that is almost-right, to know what to leave out. As the cost of building falls, taste rises in value, because it's the thing that can't be automated away, and the gap between people who have it and people who don't gets wider, not narrower.
The reason taste resists automation is that it doesn't live in any single decision. It lives in thousands of small ones, most of them invisible, each one a tiny judgment about what serves the whole and what's just noise. You can copy a person's output, but you can't copy the sequence of refusals that produced it — the hundred things they considered and rejected, none of which appear in the final result but all of which shaped it.
This is why taste can't be learned from a tutorial, and why it makes such a good moat. A tutorial teaches you what to do. Taste is mostly about what not to do, and the negative space is much harder to transmit. You can hand someone every technique and they'll still produce something subtly worse, because the techniques were never the constraint. The constraint was knowing which technique the moment called for, and that knowledge has no shortcut.
You can feel the same shift inside a single product. The hard part of building, increasingly, isn't adding things — anyone can add things, the tools make it trivial. The hard part is knowing what not to add, which feature to refuse, which clever option would quietly make the whole worse. Subtraction is where taste shows up most clearly, and it's precisely the part that cheap, powerful tools make harder, because they remove every reason to say no.
So how do you get taste, if not from a tutorial? The boring answer is the true one: exposure and reps. You study a lot of great work until you can feel why it's great, not just see that it is. And you ship a lot of your own work until you can feel the difference between your good and your almost-good. There's no compressing this. It's slow, and the slowness is the point, because if it were fast everyone would have it.
That slowness is exactly what makes it defensible. In a world where the technical barriers keep falling, the things that take years to acquire become the durable advantages, precisely because they can't be bought or downloaded. Taste is one of those. It's impossible to fake for long, impossible to acquire quickly, and impossible to copy from someone else's output, which is the whole definition of a moat that lasts.
The strange implication is that as tools get more powerful, the human part of building matters more, not less. When making was the bottleneck, makers won. Now that making is cheap, the people who know what's worth making win, and that knowing is irreducibly human, irreducibly slow, and irreducibly personal. The future belongs less to the people who can build anything and more to the people who know which things are worth building.