Photo taken in 2019 in Tibet

With “double 🥚” in mind, I’d like to take a look at some of the things I did in 2020 as a front end. There are a lot of discussions about technology direction, including Serverless, WSAM, Flutter, everything. I thought about it, and with my business background, I would like to talk about what MY team and I have done in the front direction this year and the thinking behind it. Before we start, let’s first say that my team is the first “proprietary Dingding” front end team ** under Dingding Business Division of Ali Cloud Intelligence. The young team, little more than a year old, is working with large clients based on the firm’s existing standard capabilities. In particular, large-scale organisations, such as governments and state-owned enterprises, cater to their fancy business needs.

It is also due to the keynote of serving large customers and Ali’s “cloud nail integrated” strategy. Our entire team this year built our technology layout around the words “proprietary,” “modular,” and “customized.” Based on the existing infrastructure of Ali Group, the team has grown rapidly and vigorously. With the proposition of broadening the frontier of the front-end 2B2G field, we have been walking for one year. Therefore, here we would like to talk about our team’s exploration and thinking on these three keywords, which is also a good introduction to the jade.

proprietary

Mature 2B business, little not the shadow of proprietary change. Imagine if your entire suite of products and services (front and back) were to be deployed to a proprietary environment without a public network. At this point the front-end application will:

  • CDN cannot be used to accelerate resource acquisition
  • Limited Internet access
  • A large number of mature public network-based services and solutions are no longer available

There is a sense of isolation at first. Fortunately, the team did a lot of engineering work this year to solve the proprietary problem:

  • Static resource build time analysis, local aggregation, output build package
  • Docker image of front-end resources and one-click release with operation and maintenance platform
  • Front-end page script localization client integration (CEF container offline)
  • .

Solved the “food and clothing problem” of proprietary environment delivery 🍚, making project delivery a little more flexible. However, this piece of technology still has a lot of space to be explored, especially the deployment mode of Ali small program system under the hybrid cloud architecture, which is an important topic next.

modular

Modularity has always been an important proposition in programming. This year, the “modularization” faced by the team is to turn the “Dingding” App into a skeleton under the background of the integration of cloud Dingding, and the flesh, tissues and organs carried on it can be customized and replaced. This puts higher abstraction standards and requirements on existing business modules:

  • Modules need to be “highly cohesive” and have convergent external capabilities
  • Modules need to be “structured” to meet the demands of pluggable and assemblable
  • Modules need to be “standardized”, extensible, customizable, and replaceable

In this context, thanks to this period of time of the server and the native client students create and crazy input, in DDD (domain driven design) and so on thought under the influence of, let’s try in inventory module, to explore these modules “DNA” 🧬, out of their fixed points and core design as a “core module”, Further precipitation into so-called “business assets”.

This is essentially a high level of business abstraction, a process of “continuous refinement” that each core business module (chat IM, contacts, etc.) continues to explore for the next 3-5 years. In this way, is it possible to set a “small goal” for myself, to become the most NX front-end team in China collaborative application module design and implementation 😀, it is also quite interesting.

customized


You can never predict what fancy requests your customers will make.

It’s also hard to say no to the needs of your large (BA) customers.

Cloud nail nails under one nail is “thousands of faces”. In the face of a large number of business customization requirements, the team is not willing to unlimited pile of people to meet the customization demands. So what we want to build is “stage”, let our partners become “performers”. Looking forward to a set of models to enable partners to do in-depth customization and expansion on the basis of the capabilities of the dingding platform, extending new possibilities!

In addition to “proprietary” and “modular” above, we need to find ways to “empower” our partners to extend and customize existing “platform capabilities,” and we need to open up everything possible. Thanks to classic programming ideas such as AOP, SPI and IoC, we give them new weapons on the front end to extend and customize design.

This piece is currently in the high speed start stage, the team is desperately trying to find a feasible solution, can be handed over to three customized road. Along the way, IT became clear to me that the front-end could also benefit from these classic programming ideas, such as front-end frameworks like Inversify that focus on IoC and DI. This also makes me understand that as an FE, I should never set limits on myself, so that I can go further

The last

Here’s one of my most important takeaways from this year:

  • The business tone determines the technical tone.
  • The scale of business determines the scale of technology.
  • Business diversification, decision technology diversification.
  • .

Booming business, will bring boiling technology! I also worked with my team during the year:

  • Studied the cross end material Rax of the group and implemented it in the business
  • Developed full TS, React Hooks, Redux stack
  • Toss over the fancy gameplay of Rx. Js in message (IM) module
  • Try technologies like Flutter and WASM to find a niche in the business
  • Have you ever thought about how to make the micro front end (MS) become the landing scheme of the “eternal youth” vitality of the middle and background applications
  • .

Also experienced the expansion of the team from a few people to a dozen people. This process made me understand that each team has different rhythms at different stages, and certain rules should be followed to do things. Especially when it comes to technology, one should neither be too slow nor too hasty. In the end see or step by step 👣, steadfast do technical talent is king.

Well, ha, is also to 2020 handed over a relatively satisfactory answer sheet 😉 believe that 2021 is bound to be the organization of digital intensified year ~ road long its repair xi, I will be up and down and search!

Finally, I would like to give a small advertisement to this young front end team — the proprietary Nail front end team. We welcome people with ambitions to define the future of 2B2G front end research and development system together. I will shamelessly add your wechat 😀

Denver annual essay | 2020 technical way with me The campaign is under way…