Overview of NASA AIST Evaluations
This session will give a broad overview of the first round of technology assessments coordinated by ESIP for NASA AIST, including stepping through the evaluation process as such:
- AIST Selects Appropriate Candidates
- Evaluators from Earth Sciences and Informatics Community
- Evaluators Collaborate with AIST PIs on Test Plan
- Evaluators Carry Out Testpaln
- ESIP Coordinates Communication over Slack, OSF, Email, Telecons
- Evaluators Submit Final Report Content
- ESIP Edits Final Report
We will also discuss evaluator feedback, lessons learned and next steps.
-
Independent Technology Assessment within the ESIP Testbed
-
AIST
-
A technology Readiness Level (TRL) is used to assess project maturity (internal only)
-
ESIP/AIST Collaboration:
-
ESIP providing independent assessment of AIST project TRL
-
Identifying opportunities/roadblocks for projects
-
Evaluation Goals
-
Achieve consistency, traceability and defensibility of evaluation results
-
Be recognized as comprehensive and fair
-
provide a valuable experience for PIs and project evaluators
-
-
-
Evaluation Components
-
Milestone Completions Review
-
TRL Objective Completion Review
-
Reporting
-
Evaluator Information
-
Evaluation Components
-
Test Plan
-
Infusion Potential
-
Other
-
Submit
-
-
-
Timeline: this is her "workflow" slide too, but the timeline is important
-
Workflow
-
AIST Selects Projects
-
ESIP Selects Evaluators
-
ESIP solicits suggestions from PI
-
ESIP reaches out to community
-
Telecons
-
-
Evaluator/AIST PIs create Test Plan
-
Access Restrictions Software Readiness
-
-
Evaluators carry out testing plan - did this seem to work well?
-
standardization on this may be difficult
-
strong human component
-
Checklist for software development best practices
-
Supportability, Portability, Testability, Accessibility, Community, Governance, Licensing, Copyright, Installability, Buildability, Learnability, Documentation, Understandability, Friendliness
-
In addition: code structure, modules, the quality of the codes, end user evaluation
-
-
Would love to hear from John Graybeal about his process for developing assessment; suggestions/ideas for the next round? With the Disaster Cluster project, for example?
-
Key to developing this was initial Software Sustainability Institute
-
Future work could be in defining further details, brining this up to date - or perhaps providing a simpler version?
-
Nice thing about the way its set up now is that the spreadsheet is interactive: if you are evaluating a TRL 6 project, more questions are exposed than if you evaluate a TRL 3 project.
-
-
Possible training track for ESIP people who want to grow in their capabilities to design good software/evaluate good software
-
Science Software Cluster collaboration, perhaps?
-
Perhaps they evaluate the TEF
-
-
-
Peter Fox & Chris Lynnes have done "Infusion Readiness" evaluations - are these artifacts relevant for further TEF efforts?
-
-
Evaluators fill out evaluation structure
-
Evaluators submit final report Content
-
ESIP edits and submits reports to AIST
-
-
FeedBack from evaluators
-
Lessons learned
-
Outlook: Provide the Earth sciences community with a novel, needed evaluation framework to improve technology development.
-
More on the Evaluation Spreadsheet
-
Official TRL checklist is different than John's spreadsheet criteria
-
Anne Wilson has some good ideas from a science software development perspective, to include in an evaluation.
-
Lots of interest in seeing John's spreadsheet - perhaps this is a good example to bring to the "Art of Critique" concept?
-
Spreadsheet also becomes a potential 'vetted checklist' that can help project managers/software leaders check against 'industry' software standards
-
TEF for the TEF?
-
-