If a project like the LHC were really producing useful results, the free market would jump to fund it.
Actually, businesses rarely looks farther than 5 years in a business plan.
If a research project can't make a profit in that time, they don't pursue it.
The LHC took 10 years to build, from 1998 to 2008. Therefore nearly all of the physics research that has been performed and its resulting discoveries and breakthroughs would never have happened if it was left to the "free market".
Science and understanding can not progress through simple theory. The ideas must be tested and validated. That's the reason for facilities like this.
A huge effort was put into it to understand what its intended use was and how to make something people wanted.
They should not have replaced Gnome 2 with Gnome 3, it should have been a whole new project. Gnome 3 has nothing to do with the original objective that was Gnome 2.
Remember that many double blind tests have errors in the setup that can remove the possibility of a positive result. For instance, if the switching system degrades the signal significantly, the possible further degradation of the ADA can be masked.
Having said that, I have done my own tests with SACDs.
When I compare the sound of the red book (CD) layer to the SACD layer, I rarely hear a difference. But when I do, It may be that the red book layer is not really a direct down sample of the DSD encoding, so inconclusive, but it showed to me that red book is better than I thought.