Earlier quoted context omitted.
> Release early Though never release half-baked cr@p. You have exactly one chance to make the initial splash, don't waste it.
"Though never release half-baked cr@p. You have exactly one chance to make the initial splash, don't waste it." This is absolutely not true! Release ANYTHING. A half baked product will provide you with required flexibility youll need when the market tells you what you really should be making.
That's what the market research phase of the project planning is for.
If you need to fine-tweak your feature set, do a closed alpha with a focus group, that's fine. But if your project plan relies on a public beta feedback for flushing out core feature set, that's just wrong. Or more specifically - it is very risky, because you are essentially implementing a wild guess product for a wild guess market and then hoping it would converge to something of a value.