Deep Battle Cards E–H
E — Retention vs Disposition vs Legal Hold
Retention: normal keep duration.
Disposition: authorized end-of-life action.
Legal Hold/Preservation: suspends normal destruction for litigation-relevant information.
Inputs: operational/legal/regulatory/fiscal/historical requirements plus litigation status.
Outputs: retention schedule, approved disposition or hold order.
Use: “how long?” → Retention; “what at end?” → Disposition; “do not destroy despite schedule” → Legal Hold.
Trap: legal hold is not merely a new normal retention period; it is an override/exception.
Memory hook: Keep / End / Stop the end.
Source: pp. 300–306.
F — Identification vs Preservation vs Collection vs Processing vs Review
Identification: locate matter scope, sources and custodians.
Preservation: prevent destruction/alteration.
Collection: acquire defensibly.
Processing: prepare/reduce/de-dupe/search.
Review: determine responsiveness/privilege.
Common confusion: Review vs Analysis—Review makes responsiveness/privilege decisions; Analysis interprets facts and relationships.
Scenario: huge duplicate collected set → Processing, not Review.
Exam trap: preservation must happen before routine deletion destroys evidence.
Memory hook: Find / Freeze / Fetch / Prepare / Decide.
Source: EDRM section, approx. pp. 301–306.
G — DMS vs Records System vs CMS vs ECM
DMS: document capture/version/workflow/rights/retrieval.
Records System: formal record classification/retention/disposition/legal controls.
CMS: content creation/publication/reuse.
ECM: enterprise coordination of content-management capabilities/processes.
Use: check-in/versioning → DMS; retention/disposition/legal → Records; web/content lifecycle → CMS; enterprise umbrella → ECM.
Scenario: platform versions files well but cannot apply retention schedules → Records capability is missing.
Trap: ECM ≠ one shared drive or repository, and technology deployment ≠ functioning governance/adoption.
Memory hook: Documents / Records / Publish / Enterprise umbrella.
Source: tools sections, approx. pp. 315–323.
H — Precision vs Recall
Precision = relevant retrieved ÷ all retrieved.
Recall = relevant retrieved ÷ all relevant that exist.
Use: irrelevant noise → Precision weakness; missing relevant material → Recall weakness.
Scenario: 80 relevant + 20 irrelevant retrieved, with 100 relevant total → Precision 80%; Recall 80%.
Trap: a search can have high precision and low recall at the same time.
Memory hook: Precision = purity; Recall = reach.
Source: pp. 323–327.