On June 5, 2021, the 2021 China Developer Ecosystem Summit hosted by SegmentFault came to a successful conclusion. Wen Ming, Apache Apisix PMC Chair and CEO of Branch Technology, delivered a speech titled “Explorations of Apache Apisix’s Community Operation and Commercialization”.

Guest: Ming Wen, CEO of Branch Technology, Apache Apisix PMC Chair

Sort and publish shorthand: SegmentFault Editorial Department

My name is Wen Ming, from Apache Apisix community. The content of Pengfei and I are actually somewhat similar. We are both an open source project based on the same global, and we will also make some attempts to commercialize it. We both went from 0 to 1, but the difference between us is that he had Qingyun and the resources behind him. We had nothing at all. When we started, there were only two people with no money.

How did we get this project from zero to one? Do Apache, make it a commercial company, and then make it a commercial company with dozens of people, and have commercial customers. So today I’m going to share some of our own insights.

First of all, let me introduce myself. I am a member of Apache Foundation. What’s special about the Apache Foundation is that unlike CNCF, it does more code from Contributor, and then you have write rights and become an Apache Commiter. After commiter, you will continue to contribute and participate in the community, and then you will become a member of PMC. PMC is the project management committee, where you can vote, elect, become a PMC member and then move up to PMC Chair. Apache Foundation has more than 300 Top Level Projects, so there are more than 300 PMC VPs, namely PMC Chair. There are more than 700 Apache members in the world, and more than 30 Apache members in China. After becoming a member, you can choose the board members.

As Wu Sheng shared today, Wu Sheng is one of the nine board members of the Apache Foundation. I myself am PMC Chair of Apisix, Member of Apache, Commitor of Skywalking, and Member of K8S. These titles are earned by your code and your contributions to PR in the open source community. Many of the people who share today are community operators, marketing people, or community people, but I am actually a developer.

About tributary technology

We formed a commercial company in ’19 called Tributary Technologies, and we were going to commercialize Apisix.

Branch Technology is a commercial company we established in 2019, and we will commercialize it based on some Apisix. Our current product lines, including gateway, K8S ingrass, and our service grid, are actually some products related to users’ four-layer and seven-layer traffic under the cloud source. The three product lines are linked together.

Last year or so, we got several rounds of financing, and now we have more than 20 employees, all Remote, in 12 cities in three countries. We connect people through things like Slack, Tencent Meetings, JIRA, GitHub, etc.

So this is actually a technology company in China that has a little bit of Silicon Valley style, and all of us now, except for the administrative students, all of us have a Developer background. You can see our relationship with various foundations. We are the only project in China that has been donated to the Apache Foundation by a start-up company, and we are the only Chinese project in the API Gateway in the CNCF Panorama, and we are also the founding member of the Linux Microservices TART Foundation. The TARS Foundation is a micro-service foundation initiated by Tencent, so we have some connections with the three major foundations in the world.

This is our team. Besides me, there is a small partner named Yuansheng Wang. The two of us have been writing codes in the dark room for more than a year. Then Yeliang Wang, as we know from Chinese technology circles, is in charge of our BD and products.

Apache Apisix community operations

Just to give you a general idea of what Apisix is, because a lot of people don’t know what Apisix is, and a lot of people ask us, The reason why it’s called that is because it’s not really a word, it’s a word we made up Apisix, we do API, API is a very important thing, all of the services that we use on the Internet, all of the services in the cloud are exposed through API. So API is a very important thing, and we want users to manage API efficiently and safely, so we choose SIX, and then our enterprise version is called SEVEN, because SIX is a lucky number in China, and SEVEN is a lucky number in foreign countries.

We entered Apache Incubator in October 2019 and graduated in July 2010, which became TLP (Top Level Project) in a very short period of time. During this process, we released one release per month, and when we first entered Apache, we only had 20 Comtributors. Now we have 250 Comtributors. In just over a year, the number of Comtributor has grown tenfold.

As Pengfei said just now, we have made a small open source tool. The purpose of this open source tool is to see the growth curve of the contributors to open source projects, because now there are so many people doing open source projects, so how to evaluate an open source project? Some people look at Star, some look at Comtributor, some look at Issue, some look at PR, and some look at some features or performance compared to similar projects.

But for us, whether it’s an open source project or a company, the core of it is people, so I think it’s OK to just look at people. For example, how many of the best people in the world are working on your open source project and contributing to you. How many companies with very big logos are using your products? In fact, it just depends on this.

In business, which companies with big logos use your products? Who are the top people using it? Just know how many people from different parts of the world contribute to your open source project, the STAR number doesn’t mean anything.

This is a growth curve for Apisix.

Another question is, should we look at monthly activity? There are a lot of open source projects. It attracts people, but it doesn’t keep people. This is a big problem.

This is the monthly contributor curve for Apisix, which is how many people contribute to Apisix each month to be incorporated into the backbone, so we’re probably around 25 people.

Compared with some other projects, for example, a similar company of ours in the United States, Kong, has more than 400 employees, while his Comtributor has 15 employees per month, while our company only has more than 20 employees. Therefore, most of the contributors of monthly activities still come from the community.

Are contributors paying attention to your project? Are they contributing to you? This is a point that we are very concerned about, the core of the whole open source project is Developer, and then whether it is our publicity, our documents, our videos and so on, it is all means. The core is whether I can attract the attention of the world’s top developers, because now is the era of many open source projects, an open source project has many competing products, a company, a developer has a lot of choices, why does he choose you? Why not choose another open source project? You have something to attract him.

After you’ve hooked him, he’ll use it and then find your bugs and contribute code to you. This actually forms something similar to the whole life cycle, so what do I do to keep my contributors active all the time, this is a key point in our whole community.

Again, we never look at the star count.

A lot of open source projects, it will add a probe into its project, is that I install it and I will return some data back, the role of the probe is to know which company is using me, which city is he from? How much data is there? But Apisix being an Apache project, we can’t have any privacy violations, so we don’t have probes. We can only rely on open source users to report online, so we can only know part of the users. However, these users are mainly in Internet companies. In fact, online education is also related to the recent epidemic. We started writing the first line of code in April 2019, and two months later, in June, we had an open source demo. And our early users of Apisix were we were taking people to dinner and saying do you want to try it? Do you think I am a reliable person? If you think I am reliable, can you use my open source project and go to the production test environment for a run?

When you sell your product, you’re selling your people. So how does Apisix maintain an iteration of one release per month? How do we get our program to be used by more and more companies?

We use stupid methods, we don’t use dark technology, we use very stupid methods. We put the user problem is in the first place, I don’t know what the author of an open source project, like us, every day is busy and user communication, every day we look at the user’s feedback, there is a problem can direct feedback to development, I don’t know how many people can see user feedback every day, but we want to see every day.

Another way is to preach. The sermon stuff is actually very slow. Apisix has an offline meetingup once a month, not during the epidemic. Now it’s an offline meetingup every month. Last time it was in Shenzhen, this month it will be in Chengdu, and next month it will be in Hangzhou. We will brush up on Beijing, Shanghai, Guangzhou and Shenzhen, Hangzhou and Chengdu, which is equivalent to saying that we will visit one city every month, and when we visit cities, we will be the CTO, and the CEO will visit customers personally. Because only by getting closer and closer to your customers and listening to their feedback, can you know what your product will do?

In addition, obsession is also important. The core reason why many open source projects don’t stick is that the founders may lose interest. This is an important thing, no matter how big your project is, like Redis, when the author of Redis announced his retirement, Redis actually had a big impact on open source projects, and on commercial companies. Just like the K8s Ingress Controller, which is the second highest open source project in K8s, when its author announced his retirement, the project stopped updating for more than 20 days. No matter how popular an open source project is, there are only two or three core people behind it. If one day these two or three people feel that they don’t want to work on it, are not interested in it, or do something else, the open source project will surely die. Therefore, obsession is very important for us. The project of Apisix must be a lifelong project of Wang and me. When we entered Apache Comtributor, we went to our tutor in China, Mr. Jiang Ning, who told us that there was a price for you to enter the incubator. The label on this thing will be with you for the rest of your life.

We’ve actually learned a lot from Apache. It took us more than half a year to graduate. However, there were many cultural conflicts and we learned a lot during the period. We found that what we learned was not only applicable to open source projects, but also helpful to commercial companies. Apache’s belief is that community is more important than code, and we believe that community is more important than code. So when we communicate with our customers, we think, for example, even if he can’t become my paying user, can we cultivate a Comtributor? Let him contribute my code? So this company he’s not spending money, but he’s spending money on my open source project, and we’re going to build the whole community. One of the things that’s really important about Apache for the public good is that I do so much code, I’m a charity, I have so much code, I have Hadoop, I have Spark, I have Kafka, it’s free, I do so much good to the world, I save so much money, It has brought benefits to mankind. And what about a commercial company? A commercial company can’t just do it for money, it must have big goals, it must do something for the public interest, so that it can run in the long run.

There is also an open channel in Apache where we discuss things on mailing lists, which is a very old way of doing it. In fact, many people don’t read mailing lists, or in China many people don’t use email, they use chat software, but why does Apache always use mailing lists? In fact, we also went to ask, said that now everyone is so developed, right? Why can’t we talk about it on GitHub? Since all of our projects are on GitHub, why do we discuss them on the mailing list? Then the Apache guy replied, because Apache has some project Comtributor in Africa, or in some very poor countries, even within the region where GitHub is blocked, how can he contribute code to these people? How does he participate in the discussion? We need to involve everyone in the discussion. Email is the most reliable, because you can’t send an email, the email is plain text, there are no pictures or videos, so he can participate in the discussion even on the poor network. So Apache wants to get everyone on the planet involved. It wants everyone to be able to participate in the discussion. And there’s no way to change anything on the mailing list, but GitHub can delete or change your posts.

At Apache, we talk about putting something on for 72 hours, which is three days, because people don’t see it over the weekend, so it’s three days before people can push it forward. It’s the same in our company. Because our company is also fully remote, people also connect their content through Slack, through GitHub, through Google documents, through email. If you can’t write your proposal clearly, don’t drag others to the meeting. You’re wasting their time. In Apache is also the same, when you tell someone something, you must write it very clearly, otherwise no one will pay attention to you. There is also an elite democracy in Apache, as Tujunping said this morning, in Apache, everyone is democratic, who can vote, but your vote and my vote weight is not the same, for example, I am a PMC, my vote is valid, you are not a PMC, Your ticket is invalid, but the friendship ticket does not count. The more you contribute, the greater your voice will be, so Apache is an elite democracy, that is, everyone has the right, but the more you contribute, what you say will count. In the company is also same, when employees do you give the greater the value of your company, he also will say, but we hope that everyone involved, so we are similar to push OKR, everyone can see everyone OKR, all things are in the open, but you can do it, It’s another story.

Apache Apisix commercialization exploration

Then let’s take a look at open source commercialization. Open source commercialization is a hot topic, especially in China in recent year. It seems that something can be commercialized after open source, but in fact it is different. For example, you can see tens of millions of projects on GitHub, but few of them can be truly commercialized and have commercial value. Many projects, such as front-end projects and tool projects, are difficult to be commercialized. In fact, there is no inevitable relationship between open source and commercialization. In fact, open source is just a programmer who thinks this tool is fun. For example, Linus, when he first developed Linux operating system, was not for commercialization, he thought he wanted to make something more fun. Many open source projects are not intended for commercialization, but commercialization is a great success. There are also good open source projects that can’t be commercialized, such as some front-end projects, which are difficult to be commercialized, but can earn tens of thousands of dollars per month depending on sponsorship. He acted as a consultant and then got sponsorship, which is also very good. These projects are difficult to be commercialized, but it is a very good and successful open source project. Many of the software companies that are well commercialized are not open source, such as Windows and Office, which are all not open source, but they are very successful in commercialization and even have no substitute for it. But do you think open source can kill them? Very difficult, so a thing business is not commercial, can be successful, and open open source is not necessarily related.

Open source projects and has a large gap between commercial product, actually I have been reminded of the company’s friend don’t think you did a very cow force, good performance, high star with a lot of people, because companies want to is a commercial product, but you do is an open source project, commercial product to what is, is safe, stable, Then out of the question, can you help him think of a variety of similar to high availability, rollback, audit, even can reduce cost, improve efficiency, to solve these problems in our business, but open source projects can not directly solve this stuff, do a good job in an open source project you just commercial a foundation, also has a gap to jump, there is a book is “crossing the chasm”. When you have a project that goes from Startup to commercial scale, there are a lot of gaps, and the role of a contributor to an open source project is different from that of a commercial company manager. This is a big challenge for many open source business companies, because both of us are developers and we have been writing code for more than ten years. We even managed a small team of a few people without managing anyone in our former company. What should we do then? Now I want to be a business company, I have to talk about money with investors, I have to recruit people, I have to talk about cooperation with external partners, I have to write articles, and then I have to become a web celebrity… You have to do a lot of things. You have to hand out money and negotiate options with employees. Therefore, I think working in Developer is the happiest, because you don’t have to take care of all these chores, and how happy you are to write code with peace of mind. Why do you want to start a business? There is a big difference between his roles, sometimes feel that programmers choose to start a business is not necessarily a good thing, then why there are so many people out? I think the biggest attraction is that the challenge is so big, that you feel like I used to be just a developer, but now you can directly decide something, decide the direction of the company, the direction of the open source project, the strategy behind the company. The challenges are going to be very big, and the challenges are going to be very high, so if you’re going from being a core contributor to an open source project to being a manager of a commercial company, it’s going to be very important to know whether or not you’re going to be able to experience the highs and the challenges. You know, if you feel good about it, then you can do it, but some people think it’s too hard and it’s not right for them, and then you can’t do it, and it’s very challenging to change roles.

Then why do people choose open source now? In fact, I think whether you choose open source or not has no inevitable relationship with the success of your project.

Open source is a means, for example, some people open source for marketing, hope that more people know, hope to be on the hot search, this is one aspect; On the other hand, I hope to get access to the world’s top developers and more scenes with big logos, which is one of the purposes of open source. Others open source, just for financing purposes. So open source has many purposes, many ways, but if you really want to do it for a long time, the most important thing is to depend on your core developers or your founders, in the end what is your original intention? And then when I went to talk to my partner, I was like, why did you choose us? Why did you choose to open source this thing? Two people are very helpless and painful, no money, writing more than a year of code, 19 years of no one to look at open source, investment is the coldest time, why we choose to do this thing? And then my partner’s thought was, I’m not after money, I just want to put my name in history. He talked to people like this every time, and he did, and he had a miserable life. Actually I am also similar ideas, my first job was working in Jin Shangan for five years, and then the slogan of jinshan is to make our code, our software runs on every computer, and my partner’s dream is to a name written in the book of the annals, we all hope to be able to make some contributions, we in the application layer for more than ten years, like a lot of developers, I’ve been developing my business for more than a decade, but after more than a decade of development, where is my code? Where is the code now? It’s all gone, there’s nothing left, but you do a top project at Apache and you do a good commercial company based on top project, it’s going to be left, and it’s something that’s very pleasurable to us.

conclusion

To sum up, open source is really just a means. It depends on what your founders want.

On the other hand, open source projects need to find users who can run together. At present, people like Tencent, Weibo, Shell, and Zhongyunhui can accompany you in a long run. They can put forward a lot of bugs of your products, and then they can become your contributors and PMC. Then gradually you can play from one scene to another scene, and then play to another scene, do not build castles in the air, do not build cars behind closed doors, let the user’s needs take you to run, you can ensure that you will always be the most advanced.

We think we have a clear business path because a lot of open source projects can’t be commercialized because users don’t want to pay for them.

So what kind of open source project users are willing to pay for?

If your open source project fails and his company fails, he is willing to pay for it. For example, a database, in which TP is not the same as AP: when TP’s database goes down, the user hangs up. It doesn’t matter what AP is, I can also change it. Apisix is the traffic portal, and the traffic portal cannot be suspended. If it is suspended, all the connections between APP and users will be completely cut off. This is also a very critical place, only the critical place can receive the user’s money.

In addition to Apisix, we now have the Ingress Controller pushed to Apache and the Service Mesh just opened source. We went from one open source project to three open source projects, where 90% of the functionality is the same, but the scenarios are different.

Finally, open source is only suitable for some companies, some projects, it is not a silver bullet, not to say that I this thing is not good, I open source once I fire. Open source commercial have a bright future, because it can really help users reduce costs, to help it reduce the cost of operations, the cost of the engineer, help him to improve efficiency, help him to do the digital transformation, to help him to go to cloud and open source projects can help him to do, but is very clear, from the project to the product actually need to do a lot of things, And this is something that the founders have to be very clear about.

In addition, now is the best an era is an engineer, if we think back to more than ten years ago, or two years ago, you don’t have the chance, to try to do business, such as many open source projects, now I take a sum of money to try to open source, even if failure is also a good thing, because you won’t have the chance. Now Chinese engineers have accumulated some, Chinese capital has also accumulated some, he can do such a trial and error, there are always some projects and some commercial companies can come out. And I think this is also a very good time for the operation of open source projects or marketing students. If you are interested in what I just talked about, you can add my WeChat. Thank you.