I pushed a change draft for Issue #46 : Provide support for graphs and graph traversals The change draft is: Documents/ChangeDrafts/sarif-v2.0-issue-46-graphs-nested-traversals.docx I will move its adoption at the next TC meeting. This version includes Michael’s suggesting for identifying nested traversals with an id, rather than embedding them. This draft also includes a few dozen wording changes that tighten up the text. Most notably, when I want to refer to an ancestor object (such as the result object that ultimately contains a message ), I now refer to the “containing result object”. Previously, I’d been inconsistent in how I expressed this. Michael and David approved of this formulation. Thanks, Larry