A similar point was raised in another article not that long ago.
The only problem with the idea of "merging books and the Internet" is that it doesn't account for those who are paid "by the book," so to speak. The original intended users of the Internet, those who would be putting all those documents up for everyone to access, were scientists and engineers who were earning a salary for their work, not for their web-intended publications. It's easy to put your text online, if you're already offering it for free to collaborators.
But for those who are paid for the book itself, a system of monitoring who accesses those books (and how) is needed, in order to compensate the authors for their work. Then, of course, you need a method of compensation. And oh, yes, it will have to be a universal system to work for all countries on the web that may share documents.
So the issue is not nearly as cut-and-dry as this blog or the other article would suggest; which means the phrase "about to merge" is seriously understating the timeframe needed to accomplish this. The process--if everyone agreed to it in the first place--could take decades.