Clean Insights is Open Source Software. So while we have an idea of where we’re headed, we’d love to hear from you. We track most of our work on Gitlab, which is a great first stop to report an bug or request a feature.
- Easier onboarding
- Supporting data streams beyond user behavior
- Supporting more use cases
Here are some of the things we’re working on or want to work on soon:
- [Done Apple, Android] Support auto-tracking on non-JS versions.
- Our JS SDK supports “auto-tracking” which gives folks an easy on-ramp to privacy-preserving analytics by tracking the most commonly measured in-browser behavior, pageviews. We’d like to offer similarly easy-to-integrate options for other platforms with well-worn patterns. For example, we’d like our Android and Apple SDKs to track app-starts.
- Create a built-in consent UI
- Implementing a custom consent UI can be one of the time consuming parts of trying out Clean Insights. It’s a lot harder than doing what a lot of other packages suggest: nothing at all. While we believe a thoughtful consent experience is one of the most important aspects of an analytics integration, we want folks to have a simple option they can use while prototyping to testing.
- Make testing integrations easier
- Clean Insights batches data on device before sending it to the analytics server. While this reduces the opportunity for a server or observer to infer activity based on network traffic, it makes testing new or modified integrations difficult. We’d like to add a “report right now” mode that’s suitable for testing, but not for production use.
- One of our most commonly requested features surveys. Even when teams are collecting behavioral data, they also want important human context on why users are doing what they’re doing, or what kind of use case they have.
- Supporting new backends
- Today, Clean Insights is built to store data in Matomo via the Clean Insights Matomo Proxy. Matomo has a rich feature set and compliments the Clean Insights usecase well. But, simpler backends or ones more tailored to the Clean Insights use case might be easier to administer and to parse data out of. Among other open source candidates, we’re looking at counter.dev. We don’t expect to build our own backend from scratch at this time.
- Analytics add-ons for common frameworks (e.g. Python’s Django or Flask)
- We’d like to make it easy to measure common patterns within popular frameworks. For example, to collect privacy-preserving page-view analytics server-side (replete the usual CI staples of consent and transparency).
- Analytics add-ons for [package-manager-distributed] libraries.
- We’ve seen a pattern towards measuring usage CLI tools. And, as distributors of libraries ourselves at Guardian Project, we often wonder how folks are using our libraries out in the wild. This is an awfully tough use with a very privacy-concious audience, so we’re still working on exactly how we’d want to do this.
A more detailed list of issues can be found in each SDK’s repo. There’s no timeline attached to most of these; we prioritize our work based on the needs of folks implementing Clean Insights and their users.