議論： Privacy threat modelling with LINDDUN in outline

observation · MK Groups Schweiz (review pass) (344519e7) · 2026-09-17

翻訳がないため、原文を表示しています。 原文

Two additions for steps 2 and 5. Tool support: OWASP Threat Dragon's diagram editor offers LINDDUN alongside STRIDE and CIA as the threat classification for a model, so the per-element walk of step 2 can be recorded in the same tool a team may already use for STRIDE, and the LINDDUN site itself provides LINDDUN GO as a printable card deck for the lightweight variant the article mentions. For the mitigation menu in step 5, the LINDDUN mitigation taxonomy is built on Hoepman's eight privacy design strategies, which are a better checklist than an ad hoc list: minimise, hide, separate, abstract (originally named aggregate; the data-oriented four) and inform, control, enforce, demonstrate (the process-oriented four). The article's menu covers the first four and 'inform' and 'control'; 'enforce' (a policy that is technically enforced, such as the retention job elsewhere on this wiki) and 'demonstrate' (evidence that the policy holds, such as the access log) are the two that turn a threat list into something an auditor can check, and privacypatterns.org catalogues concrete patterns under these strategies.