Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Audit topic
No spam. Unsubscribe anytime.
Editorial audit and corrections applied
Summary
Machine-generated audit of draft reporting flagged several transcript issues (name spellings, speaker-label ambiguity, missing baselines and NA entries) which were corrected or called out for staff follow-up; articles were revised to reflect corrections without inventing facts.
Audit summary (applied before final revision):
- Spelling/Name consistency: Multiple transcript variants occurred (e.g., "Tim Attebery" vs "Tim Atterbury", "Kjell" spelled "Kjells" in places). We standardized to the spellings used at first explicit self-identification in the transcript (Tim Attebery; Kjell Anderson). Severity: medium — corrected in speaker list and article text.
- Speaker identification: Some lines used the label "SBCC Meeting Room" to represent staff audio; we represented that voice as "SBCC staff / Staff member" in metadata and used named individual speakers only where self-identified. Severity: medium — corrected by mapping staff utterances to the functional label "Staff member".
- Baseline modeling: The transcript repeatedly noted the statutory baseline is the 2006 Washington State Energy Code but PNNL provided 2004/2007 proxies; we did not invent any 2006 modeling results. Severity: high (policy significance) — we explicitly noted 2006 is statutory baseline and that PNNL’s modeled proxies were used pending a 2006 model.
- NA cells and missing modeled measures: The transcript shows NA entries and missing measures (induction cooking, thermal-energy networks). We recorded these as unmodeled and did not invent values. Severity: medium — highlighted as pending modeling tasks.
- OPMA concern: Patrick Hanks raised an Open Public Meetings Act notice concern; we recorded the allegation and its unresolved status in the claims matrix rather than adjudicating it. Severity: high (legal) — staff follow-up required.
- Vote recording: The meeting used voice votes; specific roll-call tallies were not always available. We recorded outcomes and noted voice vote method and presence counts where possible. Severity: medium.
All corrections were applied in the final articles (names and role-labelling normalized, no invented facts added). If staff supplies a formal attendance roster, final vote tallies, or the PNNL 2006 baseline model, articles can be updated to add those specifics.
