How the local session method works
The product treats each handoff decision as an explicit link, not a mutable spreadsheet cell.
Stable cue, changing script
A cue keeps one stable identity. A correction creates a numbered immutable revision. A take points to the exact revision read, so a later script change can reveal a stale selection without erasing history.
One current selection, every alternative retained
Selecting a take clears the prior current selection for that cue but leaves all alternatives and notes intact. Pickup responses form another explicit link rather than overwriting the original take.
Observed metadata, modest claims
The browser compares expected filenames with bounded WAV/AIFF header observations. It does not decode audio samples or infer whether a performance, file, or delivery is correct.
Worked example
Cue C017 in the sample reads: correction, the lantern station is beside row six, not row five. After the baseline freeze it was amended to say row seven, not row six, and the amendment created revision 2 with the reason recorded as a field map correction. The take that was selected for C017 still points at revision 1, because that is the text the performer actually read.
Nothing was overwritten and nothing was silently fixed. The take is still valid evidence of what was recorded, the cue is still the current script, and the mismatch between them is now a visible finding rather than an accident waiting to ship. A spreadsheet cell would have destroyed that distinction in one keystroke.
Method
Every consequential relationship is an explicit link rather than a mutable value. A cue keeps one stable identity; a correction appends a numbered immutable revision; a take points to the exact revision read; a selection points to one take; a pickup round points to a cue revision, a previous take and, later, a response take.
Because the links are explicit, a change anywhere upstream shows up downstream as a stale reference instead of disappearing. That is the whole design: preserve the ability to notice.
What this does not do
- It cannot tell you whether the newer script revision is the correct one.
- It cannot tell you whether a take is a good performance of the revision it references.
- Display labels are user-entered context and are not authenticated identities or signatures.
Do this in the app
Amend rather than edit, and let the stale marker do its job.
- Open station 02 Cues and amend the cue, giving a real reason.
- Read the takes station for selections that now reference a superseded revision.
- Decide per cue whether to re-record, re-select or accept the older read.
Open the local workspace at station 02 Cues