I am happy with today’s progress. The 4th lap is so interesting. It was not that there-is-no-tomorrow kind of sprint. I was just chilling and relaxing. Why were they always faster. Also this week’s activities are recorded by a forerunner 965. It is a nice watch!

The only way to get my pace under 1min/100yd is to use butterfly. It is my first fly today after ~3 years, and I was pretty anxious before that start, but it felt excellent! Also added 10x50 kicking drill for my freestyle everyday. I will see how it goes.

Tiny improvement recently. Stroke count seems much better. The two gaps in the chart are kicking drills. There is one issue - Garmin’s Auto Rest occasionally has +- 3 sec when I finish an interval. I start to consider Apple Watches.

I feel I will be stuck at 1:30+ One Hundred forever. There is still some room for longer distances. I am around 1:40+ zoo. Probably that is my next thing to improve. Also swedix goggles arrived today!

Finally finished active October (20 swims)

I usually use bilateral breathing but still prefer left side. This is common for many people. However in single arm drill I am always better on the right side. I don’t quite understand it.

🌈

Suppose I have a collection of identical, non-overlapping, unit-height rectangular blocks in 2D, each given by a real coordinate x and integer coordinate y. What is the simplest algorithm to check if these blocks are in static equilibrium?

Pace is approx. the same but stroke rate is reduced. That is good. Best pace should be around 1’30. That 1’13 is probably a mistake of my watch.

Also tried slight long distance (1.59mile) the other day. Considering to participate this year’s Alcatraz event. The pace looks … fine.

Received an interesting book today.

Got two chess books for my daughter :) then realized the openings one seems beyond her age

One year of swimming. Start to wear HRM recently. It feels a bit dork but captures my aerobic/anaerobic load accurately.

Switched to Apple Watch last month. Missing distance alert a little bit. But it can automatically track kickboard drill! It also feels much thinner when running. The lte edition allows me put my phone at home.

Sadly I have to decline it because it was automatically timed by auto-rest rather than manual starting/stopping button.

Bumpy objective, and high discrepancy in a single gradient vector happen at the same time. I can see gradients can be quite different, but why discrepancy in a single vector? Why local min likes noisy grad? Problem of my annealing rate?

Probability. Which side are you on, Kolmogorov or Cox?