I tend to disagree: from what I have seen having a lot of "pluses" is more a distraction from actual thinking that a plus in terms of productivity. Same goes with a lot of things in life, the amount you invest (in time/ money) for the extra luxury generally outweighs the benefits of a simple and effective solution.
Also I can't really see the correlation between building a good product and a desk or its organization. What build a good product is a good team, a good organization, good and proven tools.
It reminds of a story with a friend, this guy was quite rich and was trying to learn guitar, so when he started he bought a $3000+ guitar (Martin D20 or something approaching), but given his skills a 1000 or even $500 dollars guitar could have suited his needs for the next 5 years. He didn't play better than somebody else with better tools, maybe worse because of the complexity of repairing the guitar. What drives the product is the skill of the programmer, the tools are secondary.
The difference between your friends guitar story and most of HNs computer usage is that your friend was just trying to learn guitar while most HNers use a computer as a job. I think you would agree that someone whose job it is to play guitar should probably spend (money doesn't equal quality, but just using it as a point) whatever they think is necessary to have the best guitar for them.
When it comes to using a computer all day, every day, having great (not just okay) input devices, and great (not just okay) monitors not only can make you more productive, but also save your eyes and prevent RSI. Desks also fall into this category of saving your actual health.
Also I can't really see the correlation between building a good product and a desk or its organization. What build a good product is a good team, a good organization, good and proven tools. It reminds of a story with a friend, this guy was quite rich and was trying to learn guitar, so when he started he bought a $3000+ guitar (Martin D20 or something approaching), but given his skills a 1000 or even $500 dollars guitar could have suited his needs for the next 5 years. He didn't play better than somebody else with better tools, maybe worse because of the complexity of repairing the guitar. What drives the product is the skill of the programmer, the tools are secondary.
But I might be wrong!