I am now sitting in the room of the quarantine hotel in Qingdao and not surprisingly, love the food they serve. I should have posted earlier yet as always my laziness delayed the post.

I have some thoughts to share. How to understand “less is more”? Just like Jobs has advocated the line of code that never breaks is the line you never had to write. The more one adds to his life the more he needs to maintain and the easier to “break”. And that’s one of the reason we need PCA in our lives. I put 5 items in my shopping cart in JD several days ago, planning to set up my new working desk for new life. But I deleted them from the cart 2 days ago and feel so great now. So when we want to solve a problem or achieve something, the first thing we need to do is to check if we can use the tools at hand rather than rush to buy or employ new tools. New tools bring new problem and sometimes it is not because we need the new gadget but because we need the new gadget to use a new tool, thus putting ourselves in a vicious circle of adding entropy yet without adding value. We can add entropy of course. As a matter of fact, we are adding entropy to the universe every second when we are alive. But only those who add value to their life and the world can make their life better and the world better. So the equation would be Value - Entropy > 0.

So does the scientific research. I am planning my project these days and wondering why the old prediction method is not generalized enough or become our routine practice except for clinical research. I guess it is because the value added to calculate before surgery does not offset the entropy created by doing so. People automatically choose the easiest route in their practice. It is only for a standardized measurement so they use STS in the paper. Then it would be nonsense to only improve the algorithm in its accuracy while the main reason people do not use it is not because of the accuracy, not to mention its accuracy has been good enough that new algorithms do not have apparent advantage over its original design. We may need to focus on other aspects of the potential of improvement that would add value while keep the entropy unchanged such that the new prediction models may take one step forward to a more generalized clinical usage in real world.

Add value. Maintain entropy.