It's an unknown. While we have manned rockets, we have never entirely isolated a group of humans from the biosphere for even a year, never mind indefinitely. The times it was tried, it failed. As the article points out, it is not an experiment that can be done piecemeal. All elements must be in place and interacting before the effects are known: lack of gravity, radiation shielding, recyclers, scrubbers, years of time, etc
Gravity and radiation are well understood and we can definitely build machinery that can tolerate them well for many years.
> we have never entirely isolated a group of humans from the biosphere for even a year, never mind indefinitely. The times it was tried, it failed.
Why doesn't the ISS count? It gets resupplied every few months but a Mars base would be resupplied every two years so systems only have to scale to last 10x as long.
Lunar Palace 1 made it to 370 days: https://en.wikipedia.org/wiki/Yuegong-1
This idea that "life support" is the most difficult aspect is unfounded. The most difficult problem is transportation and reducing the cost of transportation means that you can attempt to solve every other problem by simply throwing more mass at it.
NASA does much of its work through contractors anyway and there are several companies working on independent space stations. I'm confident that the "life support is too hard" argument will be proven false in a few years.
A similar argument is made that "cryogenic fuel transfer has never been demonstrated at scale therefore Starship can't work". But nobody has even seriously attempted this! And the reason is that without reusable rockets every refueling mission is unbelievably expensive and the business case never closed.