Search Product strategy
This page outlines the strategy, and goals of the Search Product team over the next year or so.
For context on the mission, vision and guiding principles of Search, see the top-level search strategy page.
Quicklinks:
Where we are now
Sourcegraph is the leading code search product in the market, used by tens of thousands of developers across enterprises of all sizes. It has advanced, novel search capabilities, such as structural search and diff/commit search, as well as capabilities that extend search, such as Code monitors and Notebooks.
We’ve launched Notebooks in public beta. You can create and share Notebooks with a free Sourcegraph cloud account. Read more about Notebooks in the docs. This quarter will also see a Code monitors release featuring a 10,000 repository limit on monitors (a 200x increase), major peformance improvements, multiple actions per monitor, Slack integration, and webhook support.
What’s next and why
Goals
Sourcegraph needs to continue to differentiate its core platform beyond code search. Search will always power our platform, but to remain competitive we need to become more than the world’s best code search. We must provide compelling solutions for the most pressing use cases of our customers. Our goal is to ensure the base Sourcegraph platform (currently Search, Notebooks, and Code monitors) continues to become more compelling and differentiated and supports future innovation from our team and across Sourcegraph.
Expand the core platform functionality with products and features that address our core use cases. Sourcegraph needs to continue to differentiate its core platform beyond code search. Search will always power our platform, but to remain competitive we need to become more than the world’s best code search. We must provide compelling solutions for the most pressing use cases of our customers. Our goal is to ensure the base Sourcegraph platform (currently Search, Notebooks, and Code monitors) continues to become more compelling and differentiated and supports future innovation from our team and across Sourcegraph.
- Support customers in onboarding new developers and sharing best practices, via Notebooks
- Enable more connectivity to Sourcegraph via webhooks and Slack integration
- Continue to improve our core platform experience by continuing to invest in the existing products and features, performance, and experimental R&D efforts that leverage our core search competency.
Details
Notebooks achieve product market fit and enable individual developers and teams to easily document codebases and share best practices across an organization by enhancing Notebooks with sharing and duplication functionality, novel Notebook creation tools, and useful templates. Developers can already embed search results, code snippets and more, in a markdown-based notebook that creates more robust documentation. For example, by using structural search to persistently capture a function and its contents, even if it’s moved in the codebase.
Code monitor integrations will provide the ability to integrate Code monitors with any service that supports webhooks, and will enable multiple actions to be triggered by a Code monitor, including Slack notifications. We’ve also lifted the current fifty repository limit on monitors to 1,000 repositories, soon to be over 10,000.
Core platform investments like Notebooks and Code monitoring integrations are how we plan to position Sourcegraph as the code search platform. We need to express the power of our code search through multiple avenues so it’s clear why best-in-class code search is a prerequisite for fostering engineering capability. Extending and integrating the core platform products and features to servve our use cases will help the market see both how essential great code search is, and that it’s a part of Sourcegraph’s value proposition, not its only value proposition.
What we’re not working on and why
While we have R&D efforts in various areas, we have no implementation plans for machine learning-based ranking or relevance improvements. ML applications in code search remain a challenge. If we were to invest in this area it would likely be closer to semantic searching than a product like GitHub Copilot. While there’s a lot of value in high quality semantic code search, there are also many unique applications that can be unlocked by staying focused on–and continuing to extend–our distinctive approach to code search.
Related use cases
This section lists use cases that are related to this product team, along with the specific relevant features.