> There's a survivorship bias in play here as we don't have a good other sample or more to compare to.

It is not survivorship bias to point out that the survivor survived.

> Even Google's Knol project, which was intended to be a Wikipedia competitor, faltered after a few years.

Not “faltering after a few years” is part of “succesfully navigating that minefield”. If you fall out of the “race” no matter how good your policies would be otherwise you won’t be a reliable source of information. Because your can’t be if you no longer exists.

This is not a statement about what could have worked, this is a statement about what did work. And there survival is a necessary ingredient of success.

It is indeed a survivorship bias since we have no good other sample in the form of competitor to compare to, like how Pepsi is to Coca-Cola. Which part of my statement you find difficult to understand?