Join xLog#
In March 2023, the new wave of AI sparked by OpenAI's release of ChatGPT last December has caused an information overload for almost everyone interested in new technologies. Every day, there are messages on the timeline about companies releasing new AI tools and joining waitlists, or new tools developed using OpenAI's API. This really made me feel that this is a special moment and I want to record something. I thought about using Apple's Notes to record, but I heard about a user losing their iCloud data and unable to recover it, so I wanted to find somewhere else. It was also on the Twitter timeline that I saw @DIYgod and @songkeys mention xLog, saying that there are bonuses for writing blogs (now that I've looked at their homepage, it seems that they are both stakeholders in this platform, haha). But I didn't join immediately. However, yesterday on the timeline, I saw @pseudo_yu and clicked on his blog to find out that he was an English major student who later became a blockchain developer, which interested me. So I decided to join xLog (although it seems that his work is also related to this platform), which means that all the users of xLog that I know of are internal to this platform. Some people say that this AI wave has overshadowed the news of the metaverse and blockchain, but I still have a curious attitude towards these things.
WhoIam#
I am a graduate student in the field of optics from an ordinary university in China. I am currently working as a front-end developer in a company in Hangzhou. I chose optics because when I was in high school, I saw a post on a certain university's BBS about a PhD student in optics who was a celebrity in that BBS, which left a deep impression on me, so I chose this major. I consider myself talented, but I didn't study hard enough. Since junior high school, mathematics has been my weak point (I think it is related to the changes in my family at that time), but relying on my strengths in other subjects, I was outstanding in high school. In the end, I completely failed the math exam in the college entrance examination and only barely got into an ordinary undergraduate school. During my undergraduate years, I felt that other people's minds were not as good as mine, and it was difficult to communicate. Only one or two people could meet my standards. Now that I am over twenty-eight, I realize that my perspective was so low at that time. There are many young and outstanding people, and I am far from being one of them. The professors at my undergraduate school just followed the textbook in class, and I liked to pretend to be a good student in class, but in fact, I didn't study. I wasted a lot of time and failed many courses. At that time, I felt that my family couldn't support me to study abroad, and I didn't care about my GPA. I liked to browse Zhihu at that time, and there was a Zhihu user @bobozeng who was actually in the same major as me, but of course, he was much better than me. After reading his experience and many other contents related to my major on Zhihu, I started to consider my future. (I still remember the year when Overwatch was just released. I had a great time that summer and also played Hearthstone.) I decided to take the postgraduate entrance examination. The first step, of course, was to graduate successfully. That year, I passed twelve exams in one semester, including retakes. I also wanted to learn Python and read all the content on Liao Xuefeng's website (I have a habit of reading the documentation completely first). At that time, I had a roommate who was self-learning JavaScript. Looking back now, if I had chosen JavaScript like him and then went to work, I could have caught up with the previous wave of the Internet boom. For the postgraduate entrance examination, I chose to take the exam in my major, and I had to take the math exam, so I did math problems for several months, but I still couldn't grasp it. I seemed to understand but not really. I was confident that I didn't need to study other subjects. After the postgraduate entrance examination, I felt that math was beyond redemption. I had a splitting headache that night and went to the cinema to watch "Legend of the Demon Cat". During the vacation that year, I had nowhere to go, so I went to a relative's house in another city. Unexpectedly, she was involved in a pyramid scheme, and I had nothing to do there, so I watched American TV series every day. When the results of the postgraduate entrance examination came out, I just passed the national cutoff for math with a score of around forty (maybe it was a fateful choice, math was very difficult that year). I felt that I couldn't pass the interview of the schools I applied for, so I started to look for information on transferring to other schools. I didn't prepare for the interview. Who knew that later they actually asked me to go for an interview, but I had already contacted the school for transfer, and I was not prepared, so I gave up the interview. After coming to the school for transfer, I still had the same feeling that everyone around me was mediocre. During my graduate studies, I needed to use lasers for experiments, and I worked very hard, but I still didn't understand the experiments. There was also a task in the lab that required me to repair the experimental equipment myself. I felt that this was what I learned the most during my graduate studies. During the repair process, I experienced the feeling of an engineer solving problems for the first time under the guidance of my advisor. During my graduate studies, I had a few insights: 1. The illustrations in papers should be exquisite. My advisor even bought us a membership to a design teaching website (I wasn't very interested). 2. Academic circles have their own cliques, and ivory towers are not pure. Everyone will look at your academic lineage. My advisor was a young professor who had just started taking students, so he couldn't provide me with much help. He saw that I was interested in programming and couldn't find a way in experiments. It happened to be the wave of deep learning at that time, and all disciplines were riding the wave and publishing papers, so he let me do experiments in this direction. He didn't know much about deep learning, so I read some papers on my own, learned a bit of TensorFlow and PyTorch, and ran some data on the old computers in the lab. At that time, I often watched videos on Bilibili. Thanks to the Bilibili content creator @霹雳吧啦 Wz, I truly began to understand some of the basics after watching his video on ResNet. At that time, there were not many people who truly understood and were willing to explain like him. Later, more people discovered the value of his videos, and now there are many people calling him a mentor in the comments. Of course, now professionals like Li Mu and others also share hardcore knowledge, and I asked ChatGPT the questions that had troubled me for a long time and got answers. Beginners nowadays are much happier than I was back then. After entering the field of deep learning, when I looked at the articles in optics that were riding the wave, they were using very simple methods, and they didn't publish the code, models, weights, or datasets traditionally, making it difficult to reproduce. After exchanging ideas with visiting professors, I also lost interest in this aspect. At that time, I was about to graduate, and I felt that with my level, it would be difficult to make deep learning my career, and I didn't have time to learn anymore. To be honest, I didn't learn anything related to optics during my graduate studies (the school did not have a program specifically for traditional optics, and the only professor in this field had retired). After doing some research, I decided to self-study front-end development. I learned HTML, CSS, JavaScript, and Vue. During the learning process, I read the entire website of Ruanyifeng's JavaScript (https://es6.ruanyifeng.com/) and the Modern JavaScript Tutorial (https://zh.javascript.info/) several times (thanks to these excellent documentation). After learning enough, I remember it was November of that year, I started to send out resumes. Because my undergraduate and graduate studies were far from my hometown, I only applied to companies in Hangzhou that were close to home. At that time, I looked at interview experiences on Nowcoder. The interviews were all conducted over the phone. I didn't hear back from the first few companies, and I wasn't well-prepared either. During the interviews, I learned about the Alibaba Front-end Training Camp. I joined, but because I was busy with graduation, I didn't have time to study much. Students who want to learn front-end development can join and it should be very helpful. ByteDance also has a similar training camp. I interviewed with another small company. The conditions were very good, and I really wanted to go, but I didn't pass the coding test. Finally, I applied to another company. The interview was very simple, but they offered me a salary that exceeded my expectations when I was desperate. I am very grateful. After joining the company, the office conditions were very poor and the work was busy, but I also learned a lot. Recently, I feel that I can't learn anything new at work, and my colleagues don't seem to have a passion for technology. The technical level of the new graduates who joined is very poor, and they can't compare to me when I first joined. The working hours are particularly long, the company's computer network is blocked, and it is difficult to find time to study privately. I want to quit my job, but I often hear news that the job market is not good, and I think my current job is okay. I want to go abroad to study again, but I am burdened by my low GPA at that time. Now I try to find some time to learn technology and English. This is my current situation. The recent AI technology wave has given me hope for the new era. At this point in time, health has become even more important. Maybe I will be able to see completely different new technologies in a few more years.
Let's talk about something else below:
Hong Kong and US stocks#
I watched a video introducing China's previous stock system, which was really immature and childish. Now, A-shares have been hovering around 3000 points. The companies I know are listed on the Hong Kong and US stock markets, so I want to trade Hong Kong and US stocks. I registered an account on the Long Bridge trading platform, but I still need a Hong Kong bank card that can receive deposits. I learned that Minsheng Bank can handle it through mainland China's witness, but it requires a passport, long-term visa, and a threshold of 300,000 yuan. When I learned about it some time ago, I could still spend 100 yuan on Taobao to buy a Nepalese visa, but now it's no longer possible. I can only spend 400 yuan to apply for a Turkish visa at (https://www.evisa.gov.tr/zh). I also consulted bank staff and learned that I need to purchase a life insurance product with a return of 60,000 yuan over five years in order to apply. The conditions are so strict that it is difficult to accept.
Blockchain Wallet#
A long time ago, I opened a blockchain wallet and bought some Bitcoin (I learned about Bitcoin in high school, but my family was poor at that time). When I bought it, the price was already high. The mnemonic phrase of my wallet at that time was still 12 Chinese characters. Just now, when I tried to log in to xLog using MetaMask, it doesn't support Chinese mnemonic phrases. After creating a new wallet, I found that MetaMask doesn't support Bitcoin, so I had to choose another wallet, Coinbase. However, the mnemonic phrase in MetaMask can still be used in Coinbase. Has it already been added to the blockchain? To be understood. When logging in to xLog using Coinbase's wallet app and scanning the QR code, it first prompts that it is connected to Crossbell, but after waiting for a long time, there are no other prompts or next steps. After a few minutes, when I used the wallet app to scan the QR code again, it prompted to switch to the Crossbell network, and there was a warning: "Be careful when switching to an unknown network. Please verify the details before switching." What risks are there? To be understood
Topics I want to write about next#
- Experience with ChatGPT, OpenAI, and other related projects from Microsoft
- Using AI to help learn English
- Foreign phone cards suitable for use by domestic users for registering accounts that cannot be registered with +86
- Virtual bank cards suitable for use by domestic users for paying for services that cannot be used with domestic bank cards
- Experience with Midjourney
- Reading the new React documentation
Experience with xLog#
I recommend using the MetaMask browser extension to log in. I initially chose Coinbase Wallet, which requires interaction with the app on my phone, but messages often fail to be received on both sides, resulting in a poor experience. However, MetaMask wallet does not support Bitcoin.