Today, I put myself in the shoes of someone "writing contracts on Dusk": the most annoying part is not the concept, but whether the toolchain can keep up. @dusk_foundation The point that matters to me now is that it wants to make privacy capabilities a chain-level primitive, rather than having applications piece together blocks themselves. For developers, this makes a big difference: you don't have to design the proof logic yourself every time or take responsibility for the boundaries; instead, you can directly call on the underlying capabilities and focus your energy on business contracts and risk control rules.
But I also have to admit that this path can easily go wrong: as long as the documentation, SDK, and debugging experience are a mess, no matter how good the underlying technology is, no one will want to migrate. The long-term value of $DUSK will ultimately be determined by "whether anyone is truly running products on it," rather than who shouts the loudest. Right now, I'm focused on one thing: whether developers will continue to stay.