On Friday, our company organized a face-to-face event with CEOs, and it was naturally difficult to ask some particularly pointed questions in this real-name format. One of the questions was very interesting: "Why do cloud computing?" In Mr. Huang's opinion, business under the sky can be divided into two kinds, one is good business, such as high-speed rail, which shortens the distance between people in time and space and improves the efficiency of the whole society; the other is bad business, such as opening casinos and selling drugs. He said that QingCloud wants to be a 100% righteous company, 99.9% can't, not even a little bit. After all, the last company that said Don't be evil has become the object of ridicule, today we want to talk about good business and bad business.
Obviously, there is no clear line between good business and bad business, and this is not a black and white topic. A business can have both good and bad parts, and we need to look at this topic in a dialectical way. For example, is TikTok a bad business? From a positive point of view, TikTok does make people feel relaxed and happy; from a negative point of view, TikTok is rarely creating value, but more consuming value. But people are social animals, emotional animals, it is unrealistic to ask a person to think only about creating value for society every day and only think about improving their productivity. So in my opinion, the business is not that bad, I just don't really like it. What does the business I like look like?
In 2021-10 I mentioned my dream of starting my own business.
- Open source: the source code is completely open, open source operation (e.g. publishing its own profit statement every month)
- Useful: able to increase, not decrease, the productivity of users, thus improving the effectiveness of society as a whole
- Profitable: able to have positive cash flow without relying on external investment
The project used as an example at the time was Plausible, a privacy-friendly open source web analytics service, and the founders laid out their idea in full in About Plausible Analytics. It seemed like 100% good business to me, and I backed it up with my own actions:
I recently came across a project like this: SealYun:
My initial reaction was to feel ridiculous: is it not an IQ tax to make money this way? After a few days, when I think about the project, I feel that the initial idea was too condescending, because this company is perfectly in line with my dream of starting a business.
- Open source: SealYun's project is fully open source on Github and licensed under the Apache 2.0 protocol
- Useful: 2000+ users have purchased SealYun's services and successfully deployed K8s
- Profitable: Based on the annual fee of 99 * 2000+ users, this company can reach 20W in annual revenue, so it must be in a profitable state
Why did I think this project was a bad business in the beginning? It may be an unconscious chain of contempt: doing infrastructure is more high-end than doing business, doing back-end is more high-end than writing tools, and building wheels from scratch is more advanced than calling other people's APIs. The problem is that users don't think that way. They think that a graphical interface is better than a command line, and that using tools is more convenient than tuning APIs. Elegant APIs need to be designed and implemented, but grounded tools are no less valuable than APIs.
Thinking this way gives me a little more confidence in my current path.
Starting from a unified abstraction, it docks storage services downward and wraps user applications upward to solve common user problems in migration, management, backup and other common scenarios.
Compared with making a distributed file system system, the technical content may not be that high, but the value for the user, I think, is not much smaller.
For self-motivation.
See you next week!
Translated via DeepL with a bit modification.