Why Zulip will stand the test of time

Tim Abbott 4 min read

Quill, a team chat company with $16 million in venture capital funding, was acquired by Twitter and shut down their product last week. Users were given a mere four days to export their data prior to deletion; direct messages and private channels could not be exported at all. Our sympathies go out to Quill users, whose needs were given so little consideration in this deal.

The Zulip story

The Quill story is in stark contrast to the history of Zulip, a modern chat app that is designed for both live and asynchronous conversations. The original 2012–2014 Zulip startup was acquired by Dropbox while still in private beta with a few dozen customers. We cared about continuing to support our user base, and made an informal agreement with Dropbox to keep the service running.

When it became clear that Dropbox would not be using the Zulip codebase, but our beta users still loved the product, we successfully advocated with company leadership to release Zulip as open-source software. In 2015, members of the original Zulip team were joined by Zulip users in a one-week hackathon to make this a reality.

Shortly thereafter, I founded Kandra Labs to steward and financially sustain Zulip’s development. We migrated Zulip’s original customers from Dropbox’s servers to the new Zulip Cloud offering, fully preserving their chat history. Despite the acquisition by Dropbox, Zulip customers have thus enjoyed uninterrupted service for over eight years (from 2013 to the present).

What software can I count on?

When choosing a piece of software that will be core to how one’s business operates, there is an important question: “Will this product still exist and be responsibly maintained in a few years?”

This question is harder to answer than it might seem. A publicly-traded company may be unlikely to go out of business, but large companies (e.g., Google, Microsoft, Atlassian) routinely abandon products. A startup that has raised a flashy round of VC funding may soon (like Quill) be acquired with the intent to shut down its product, or go out of business altogether. The safest products on this front are likely large open-source projects with a broad community, like Python, Linux, or PostgreSQL.

Unfortunately, not every organizational need has been tackled by a large open-source project. What about open-source projects driven by a single company? If an open-source vendor goes out of business, it’s still possible for the software to be maintained and used, which is a major advantage. Nevertheless, it can be incredibly difficult for project leadership to continue supporting a complex open-source product (e.g., a database or encrypted chat application) if funding runs out or business priorities change.

Building software that will stand the test of time

With that context in mind, we have designed our company, community, and technology with the explicit goal of Zulip being actively developed for many years to come. Here’s how:

  • We are growing our business sustainably, without venture capital funding. VCs are incentivized to push companies to gamble for explosive growth. Often, the result is that a company with a useful product burns rapidly through its resources and goes out of business. We have built Zulip as a sustainable business (also supported by SBIR grants from the US National Science Foundation), and are being thoughtful about our pace of spending.
  • Zulip is 100% open-source, and easy to self-host. Our high quality export and import tools let you move seamlessly between Zulip Cloud and your own self-hosted Zulip installation. When you self-host Zulip, you get the same software as our Zulip Cloud Standard customers. While Quill customers are left with a static data dump, you can continue using Zulip without interruption no matter what happens.
  • We’re building software that is easy to maintain, so it does not require a large team to keep the lights on. We have consistently emphasized high standards for codebase readability, code review, commit discipline, debuggability, automated testing, tooling, documentation, and all the other subtle details that together determine whether software is easy to understand, operate, and modify.
  • Zulip is built by a distributed community of developers from all around the world, with 74 people who have each contributed 100+ commits. Our paid team provides leadership for the project, but Zulip has a vibrant community far beyond the Kandra Labs team. Most new features are built by community contributors.

Organizing Zulip so that it will endure has required a great deal of creativity, focus, and disciplined work on the part of a thoughtful team. But our users tell us that Zulip lets them move faster and connect with each other better, manage hundreds of conversations at once, and collaborate in a way they’ve never seen before. And that makes the effort absolutely worth it.