2020 Work Log

Today’s task was to fix HDMAPMDK-1199, which I had forgotten about earlier.

Finally, I got it done at 8 o’clock.

Reviewed HDMAPMDK-892 - provided experience trajectory

Found that MDK has not yet made the corresponding adaptations, so I spent this afternoon making those adaptations.

Finished the adaptations. So tired.

Continued tinkering with my Arch, trying to record some steps of setting up my system as part of automation.

Today I plan to do the following:

1. HDMAPMDK-1199 RoadMark becomes a subclass, providing richer information
2. HDMAPMDK-892 Provide experience trajectories

If I have extra energy, I’ll work on the release documentation.

In the end, I completed 892.

Found some rhythm, got Arch installed – feels even smoother than macOS?

Played a bunch of Battlegrounds with Yuzhang. - -

I moved to Beijing for work on Saturday,

Returning to work in Beijing feels a bit unsettling—everything has changed. First of all, I no longer have my own desk (Xiu Yun hired an intern, and for convenience, she let the intern use my desk). This is the most uncomfortable part of all the disappointments. It’s like when I transferred to Class 1 in my sophomore year of high school: sitting in a new seat, not knowing anyone, and not wanting to talk to them. Everyone here seems familiar yet distant. I feel like I have very few friends here, though rationally I know this is just an illusion. But the desire to avoid talking to people and find a corner to stay in is real—and I did find one, next to Yu Zhang. Why don’t I want to communicate with others? I still can’t figure it out. Maybe it’s because I lack a clear reason for being here? When asked why I came to Beijing and how long I’ll stay, I mumble, “A while… maybe a month or two.” Or maybe it’s because, deep down, I’m still an introvert who dislikes complexity and cares too much about whether I appear foolish to others. Yet the more I write, the more I feel that the second reason carries more weight and is even more unnecessary. First, appearing foolish is better than actually being foolish. Ask others for advice earnestly, even if it’s clumsy. Don’t be afraid. Don’t be afraid when you’re young. Don’t be afraid at any time. All is well.

Now, about work. I’ve been busy with delivery recently, mainly the refactoring of TLM (typed lane model). This time, I need to set clear goals, control myself, avoid impulsive refactoring, and be strategic about it—keeping my progress in check.

Also, I got my desktop computer working again. It turns out the reason I couldn’t connect to the internet earlier was that I changed my password, and after a power outage, the network connection was completely lost. So I had to set up the machine again. The priorities this time are:

1. Productivity
2. Coolness
3. Good looks

I spent a long time reading Wang Yin’s blog today—still in the process of detoxification. I used to be a frenzied religious zealot, never seeing the essence of things. I always thought something was the ultimate solution to all problems. But in reality, “ultimate solutions” are rare; most of the time, it’s just religious fanaticism, like:

• Vim vs Emacs
• OOP vs FP
• Go vs Rust

…and so on. When I first got into OOP, I thought it was invincible and could solve everything. But in the end, all I wrote was a big, all-encompassing class—one that didn’t really matter. What really mattered were the functions under that class. Later (in the past two years), when I got into FP, I thought it was a silver bullet for all problems. But after writing some programs in Scala, I found that my productivity didn’t improve much. Instead, obsessing over syntax and immutability made me less efficient.

An intense and thrilling day. Completed HDMAPMDK-1347 and HDMAPMDK-1352, and also made a major refactor.

Learned a valuable lesson: Don’t make huge changes close to delivery, because without sufficient QA testing, even the most brilliant refactor can bring risks. Also, without enough time for design review (I mean by myself, because sometimes while writing you might realize the design was wrong—though discovering design issues during implementation is actually another problem: starting too early).

Monday supplement: During this intense refactoring day, I barely had a moment to rest, even not enough time to think. I could only rely on familiarity and flow to write code with extremely high efficiency. Actually, if we ignore the principle of not doing a big refactor before delivery, I quite enjoyed this state. But everything must have a goal and meaning; you can’t just satisfy yourself, otherwise this pleasure becomes low-grade fun.

This morning I need to do three things:

1. Clean the data and upload the final 0605 data.
2. Investigate the width anomaly issue - it’s because there is no lane at that position.
3. Check the issue where difference points were assigned to the wrong lane.
4. Re-check.

First thing in the morning, I went to help Yiming resolve the issue discovered the previous night – the problem of not being able to fetch the traffic lights.

Came in the afternoon. To get Yiming (镱铭) on board, I fixed some bugs, and managed to catch Yiming (逸铭)’s evening onboarding.

This morning, as soon as I arrived, I went to analyze the issue with Yiming. We spent the entire morning on it. The result was another problem caused by a previously known issue (point loss due to differences in the old compiler): when capturing the road, it happened to be near the missing segment, so the lane offset turned out to be a negative value.

I ended up working until around 4 AM, after a period of inefficient and painful late-night work (it seems that under the current state, whenever I stay up late, it almost always leads to inefficiency).