Solution validation: Can we use Issues as Incidents?
UPDATE: During the process of preparing mocks and building the discussion guide, the ~"group::health" determined that current GitLab issues (as of May 2020) are not suitable long-term solutions for incidents. This research cycle was terminated early, before interviews, but still considered successful as the overarching goals of the research were achieved.
What's this issue all about? (Background and context)
A critical part of the Triage workflow is when a responder determines that an Alert is critical and that it requires immediate remediation. At this point, the responder creates an Incident to coordinate the response, track progress of the incident and fix, and generally centralize activity and findings.
In %13.0 the ~"group::health" team is adding an Alert as a first-class object to GitLab. We need to determine what "incidents" will be. We have the opportunity to use GitLab Issues as incidents. The purpose of this round of solution validation is to validate if GitLab issues can be leveraged as Incidents versus instantiating a new first-class object called Incidents.
What are the overarching goals for the research?
- Identify if issues can be used as incidents
- Identify what features issues are lacking that are needed for incidents
What hypotheses and/or assumptions do you have?
- Issues do not contain critical features desired on incidents
- Issues contain features that are not critical to triage of IT incidents and these features clutter the interface
- Indicating incidents with labels makes it hard to discern the different between an Issue and Incident
What research questions are you trying to answer?
- How do we need to augment issues to use them as incidents? Once we understand what changes we would need to make, we will assess feasibility of these changes based on the constraints of the issue object. If the changes are impossible, we will create a new object.
What persona, persona segment, or customer type experiences the problem most acutely?
DevOps Engineers
What business decisions will be made based on this information?
- Can we use issues for incidents or do we need to create a new object?
What, if any, relevant prior research already exists?
Who will be leading the research?
Amelia Bauerly and Sarah Waldner
What timescales do you have in mind for the research?
13.1
Relevant links (script, prototype, notes, etc.)
Edited by Sarah Waldner