[interlagos]

It is also an instructive example of novelty. His initial experiment was reported very widely, and tens or hundreds of thousands lined up to "prove it could work", putting their dollars to demonstrate it.

Once the novelty wore off, the viability of the approach disappeared. The opt-in payment ratios started generously, and rapidly approached zero.

Little can be learned from CK's experiment because it was novel -- it got covered far and wide, encouraging a lot of people to "show" that the model works.

[dodedo]

That's an interesting theory but I'm not convinced it has a factual basis. Maybe readers decided that King's book just sucked and were waiting for it to improve? Neither you nor I can draw a meaningful conclusion on that point.

What can be demonstrated, empirically, is that money can be made without imposing draconian controls.

[interlagos]

Empirically what has been shown is that if you do an IAMA on Reddit leading up to the release of your "experiment", a lot of people will play along. It does not carry over to any other release, and says absolutely nothing about DRM or atypical release plans.

Lots of people sell videos online without the middle man. Most see no sales.

[dodedo]

"It does not carry over to any other release, and says absolutely nothing about DRM or atypical release plans"

This is demonstrably false. In fact, every artist who's ever done an online release of their content without DRM has seen sales in proportion to their general popularity. We've seen this with Radiohead, with NIN, and with many other lesser artists.

In every case, the profits were roughly in line with the general popularity of the artist's content.

"Lots of people sell videos online without the middle man. Most see no sales."

And most content distributed through the various middle-man networks likewise is a failure. Most OSS software is a flop. Most commercial software is also a flop. This statement is neither surprising nor relevant.