State of Safety Critical Rust Adoption Survey #258
Description
Purpose and Justification
While the RSCC has some representation from industry there are many hundreds more companies and thousands more engineers who could be using Rust for safety critical applications. They may have concerns, tooling or standards gaps, or needs of which we are not aware. The intention of this survey is to poll the greater safety critical community - regardless of their current language choices - and document their needs. We hope to identify gaps in standards, tooling, or knowledge that we can then help to fill with our partners and the Rust community at large.
While the Rust Foundation has been conducting excellent surveys, for the same reason that the Safety Critical Consortium was needed a more targeted survey is also needed. Certified tooling, standards, and processes like formal methods aren't relevant to the greater Rust community and haven't been addressed by existing surveys.
In short: the goal is get this survey in front of as many software-adjacent engineers in safety critical industries as possible to identify ways that the consortium can help enable Rust adoption.
Timing
It would be helpful to have at least early results by the next consortium meeting. Assuming ~1 week of data processing and report generation and an approximate 2 week survey period, we should be distributing it on or around April 21 2025. This leaves the next ~3.5 weeks for designing the questions and preparing the form. This seems doable.
Distribution
Existing Rust organizations and outlets - such as The Rust Foundation via This Week in Rust or The Embedded Rustacean - can help distribute the survey. However, it will be important to get the survey distributed beyond the Rust community and to the greater safety critical software industry. It would be helpful to have distribution via any industry groups with outreach, such as SAE.
Format
An online survey platform like SurveyMonkey or Google Forms should be a good fit. Some platforms support branching questions (i.e., your answer to a question may change the next set of questions). This could be utilized to keep the survey short and focused.
Questions
Most of the coordination work within the consortium ahead of releasing the survey will be deciding on the correct list of questions. I invite discussion on this GitHub issue, in subcommittee meetings, and on the Zulip about questions that members would be interested in asking. There will be dedicated survey design meetings later in the month to triage down to a set of questions that is broad yet keeps the length of the survey reasonable.
As mentioned above, some tools allow a branching set of questions. If there's strong interest in one set of questions for non-Rust users, one for current non-safety critical users, and one for current safety critical Rust users we could look into a tool that allows multiple paths.