- Requirements gathering process and tools
- Use case development process and tools
- Prioritization process and tools
- XSEDE1 Final KPIs and metrics (see XSEDE 1 Final Report KPIs and Metrics Appendix Worksheet)
- Potential next activities for Venkat:
Requirements gathering process and tools
CEE (Kelly, Maytal) have agreed to create and maintain a JIRA project space for end user requirements / user needs.
They will populate with things they heard from their users, and others in other XSEDE program areas will be encouraged to do so as well.
The JIRA project space will probably have informal statements of user needs, not quite as detailed as we would like for our “use cases” but enough to get started with. Key details recorded will be where the user need came from. Who reported, when, etc?
We need this done ASAP as it is critical for our team’s success on KPIs and metrics (and staying on track!)
ACTION: JP will ask Kelly & Maytal for an estimate on when it will be ready. If it is longer than a couple of weeks, he will escalate with management.
ACTION: In the meantime, we are keeping track of new items in a Google doc (here) and hope to move it into JIRA ASAP.
Use case development process and tools
ACTION: RACD will propose a new lightweight use case format that CEE and ECSS can use to guide their reporting if they are able.
For groups that are not able to use the use case format, we will draft use cases as needed and review with them to make sure we got the idea correctly.
We will maintain a use case repository based on XSEDE1’s repository and keep it up-to-date. We assume that most user needs will involve tweaks to existing use cases, with few new use cases added.
Jim expresses skepticism that RACD will be able to maintain a big set of use cases given our small funding.
Prioritization process and tools
In theory, UREP is supposed to prioritize work that XCI does based on use cases and/or user needs that we present to them. None of us are sure that they have the bandwidth to do this on a routine basis. Depends on how much input & throughput we end up having. If we have to wait for prioritization for everything, we will have idle cycles on a regular basis and things won't happen very quickly.
We recommend that if we have a staff resource in XCI that isn’t fully occupied by the current priorities and can’t be used for others, we use it for whatever makes the most sense to us.
For example, if none of the current prioritized use cases from UREP are things that Venkat is appropriate for, and we have items that he could do that haven’t been prioritized yet, we go ahead and get him started on those while we wait for the next round of prioritization. In this case, it will probably be appropriate for him to work on either 2-factor authentication Phase 2 or an InCommon identity provider for XSEDE, even though those haven’t yet been prioritized by the UREP.
- Lee Liming will propose which use cases we should prepare CDPs for
- Jim Basney and JP Navarro will review and suggest changes
- Once we agree on a list JP Navarro will circulate it for external review/approval
- JP Navarro to investigate the discrepancy between the metric to prepare 7 CDPs and to deliver 10 components to production