Repository navigation
Human Impact: HI:2.0.2 value definitions and DT_HI:1.0.0 disagree in four cells #1227
Description
While implementing the deployer decision model I noticed that the two representations of Human Impact on the reference page appear to differ. I'd like to know which is normative.
The ssvc:HI:2.0.2 value definitions and the generated decision table ssvc:DT_HI:1.0.0 give different results for these combinations:
| Safety Impact | Mission Impact | HI:2.0.2 definitions | DT_HI:1.0.0 |
|---|---|---|---|
| Marginal | Degraded | Medium | Low |
| Marginal | MEF Support Crippled | Medium | Low |
| Marginal | MEF Failure | High | Medium |
| Critical | Degraded | High | Medium |
| Critical | MEF Failure | no matching condition | High |
The last row appears to be undefined in the prose. Low requires Safety Negligible. Medium requires (Negligible AND MEF Failure) or (Marginal AND Degraded/Crippled). High requires (Critical AND Degraded/Crippled) or (Marginal AND MEF Failure). Very High requires Catastrophic or Mission Failure. Safety Critical with Mission MEF Failure satisfies none of these.
I checked the Dryad calculator's standalone Human Impact model and it agrees with the decision table in all four cases.
One consequence, if the decision table is normative: Safety Impact Marginal produces the same Human Impact as Negligible for every Mission Impact value, so the four-value decision point behaves as three values for deployers. I wanted to check whether that is intended.
Which representation should implementers follow?
Activity
sei-vsarvepalli commentedon Aug 14, 2026
@sandakelum97 - there is a related issue #1218 - your observation is right. The prose language needs an update.
sei-vsarvepalli commentedon Sep 14, 2026
Hello @sandakelum97
Are you able to provide a PR to fix the prose language and also the JSON description as seen in #1218 ? That will be appreciated very much. The logic in the JSON decision table is correct.
ahouseholder commentedon Sep 14, 2026
Cross-reference comment: Any PR that closes #1218 should also close #1227. See #1218 (comment)
sandakelum97 commentedon Sep 19, 2026
@sei-vsarvepalli Yes, happy to. I'll open a PR against the value definitions in src/ssvc/decision_points/ssvc/human_impact.py so the
prose matches DT_HI:1.0.0, with the regenerated data/ files included, and reference both #1218 and #1227.
sandakelum97 commentedon Sep 19, 2026
@sei-vsarvepalli PR submitted: #1254. It corrects the HI:2.0.2 value definitions in src/ssvc/decision_points/ssvc/human_impact.py to match
DT_HI:1.0.0, and includes the regenerated data/ files. It's written to close both #1218 and #1227.