Streamr

Streamr Canvases are microservices that consume and act upon real-time data, defined in a visual drag-and-drop editor. Dashboards are collections of visualisation widgets extracted from Canvases. While they have proven to be useful tools in the ecosystem, maintaining and upgrading the tooling in future milestones requires considerable resources and steals focus from more fundamental efforts such as developing the Streamr Network itself.

Canvases have so far been a centralized service hosted by the Streamr core team to offer a cloud-like install-nothing experience. As the whole Streamr ecosystem moves towards decentralization as envisioned, Canvases can hardly continue their centrally hosted existence. In the original 2017 whitepaper, it was envisioned that Canvases could eventually run on decentralized computation frameworks developed by projects tackling that problem (such as Golem), but building such frameworks has proven to be more difficult than many imagined at the time, and none suitable for running Canvases are really available today.

The proposal is to remove the Canvas feature from the Streamr Core application and the associated API. The code will be archived into a fork for safekeeping and potential later use. An example of later use could be to relaunch the Canvas tooling at a later time as a self-hosted version which would connect to the decentralized Streamr Network for data.

The case against the proposal

Canvases have value as a tool to create simple automations and integrations based on data from Streamr streams, including simple centralized oracles that interact with Ethereum smart contracts. If the feature is removed, users will need to find other ways to accomplish what they need. Using alternative approaches may be harder than using Canvases, which are pretty user-friendly and approachable.

Streamr canvases in the Core app

The case for the proposal

Dropping Canvases will improve the team’s ability to focus on the essence of the project, the Streamr Network and its token economics, and speed up its delivery by eliminating some baggage that would otherwise need to be migrated to newer Network milestones.

Canvases can be used to build automation, visualisations, and oracles – but it’s unlikely to ever become the best tool for any of these tasks, as better, specialised tools and methods are available to most developers and often being used by them already. For example:

• Node-RED is a popular tool for creating data-driven automation workflows, and it already supports Streamr.
• Grafana is a common and flexible framework for visualisations, and it could easily ingest data from Streamr streams with a suitable plugin.
• Chainlink and API3 are frameworks focusing on connecting data to smart contracts, with capabilities that go well beyond the simple oracles that can be built with Canvases. Chainlink already supports Streamr, and Streamr is a founding partner in API3 with an integration planned.