The meaning of Ri Gongyi: Open source search engine Typesense Author's narrative: Write a little code every day, write Typesense

"Every day has nothing to do with nothing but donations and donations to the sea". Over time, no matter how little progress is made, there will be value. This is the meaning.

Recently, the author of the open source search engine Typesense wrote down his own mental journey during development. The following is the original translation and collation:

When I first started using Typesense six years ago, I set myself a simple rule:

Write some code before or after work every day.

A very straightforward goal, no DDL, no KPI, and no milestones.

I am about to get married and I am already working full-time. The responsibilities on my shoulders are getting heavier. In this case, I can only improve myself in this way.

As we all know, it is not easy to build a search engine from scratch, so writing some code in my free time every day can save me from having to deal with extra pressure.

Some performances take an hour to achieve, some performances take a few hours to achieve, and sometimes it even takes several weeks to complete a performance, but this is not important because there is no deadline of any kind.

In retrospect, following this rule in the past 6 years has brought me gains that I can’t believe. The process is very long, but I persisted.

In the process, my life has also brought many new changes: I got married, had a daughter, and lost my loved one in resistance, and experienced major health problems due to new coronary pneumonia...

I don’t understand at what point, Jason and I have been able to build an open source search engine from scratch. The following is a brief summary of our journey so far:

• 2015: Prototyping and benchmarking of various data structures
• 2016: The first official submission to build a core fuzzy text search engine
• 2017: HTTP API, supporting digital filtering and faceted search
• 2018: Client library, read-only copy, Hacker News release failed
• 2019: Earn the first dollar
• 2020: Cluster, Typesense Cloud, multiple demos, Hacker News successfully released. 2021: Full-time invest in Typesense and write our first blog post:)

The path we embarked on is contrary to several popular traditional views about starting a company. We did not quit our daily work and immediately started to develop TypeSense full-time, did not seek venture capital, and did not try to "monopolize" by chasing rapid growth. market.

We don't have a personal brand or extensive network to use. We didn't even make a dollar until the fifth year, but our runway was almost infinite. In this long-term process, we gradually understand: Only when we fully reduce the risk of this option, we can invest in TypeSense full-time.

If you want to achieve a certain degree of success and independence by launching your own products, and want to reduce risks, my suggestion is: get an idea in a big market that is always in demand, and develop a product that is suitable for some use cases .

Choose an idea that is not mainstream, and ensure that it is in a large and preferably a growing market, which can reduce risks with a high probability and develop slowly and steadily.

As long as you persevere, you will succeed at some point.

Take Typesense as an example:

We started with a simple idea: to build an out-of-the-box and well-used search engine to democratize search. When we first started, we knew that websites and applications always need to be searched, and this is a huge and growing market in demand.

As we slowly build Typesense, we have seen several other companies launching attacks on the same market (this is market verification). Nonetheless, we continue to move on purposefully and better understand the slow growth of market demand.

Worked at Typesense for 6 years, but our story has just begun. Now Jason and I are developing TypeSense full-time, and we look forward to developing it in our own way.

Never Give Up !