WEBVTT

1
00:00:00.000 --> 00:00:08.800
This overview explains how source identities connect
to claims, narration, captions, scenes, and rendered outputs.

2
00:00:08.800 --> 00:00:16.200
A changed dependency can make downstream content stale
even when another part of the evidence remains unchanged.

3
00:00:16.500 --> 00:00:21.000
Use a fictional two-row table with values eight and twelve.

4
00:00:21.000 --> 00:00:26.167
The chosen summary is the arithmetic mean, so add the two values and divide

5
00:00:26.167 --> 00:00:28.167
by the number of rows.

6
00:00:28.167 --> 00:00:33.400
This simple example makes the evidence dependency explicit.

7
00:00:33.400 --> 00:00:35.333
The mean is ten.

8
00:00:35.333 --> 00:00:41.000
A claim card records that value with its input identities and calculation rule.

9
00:00:41.000 --> 00:00:45.833
A narration cue and a visual scene can then display the same claim.

10
00:00:45.833 --> 00:00:52.267
Each depends on the derived value rather than on a hand-typed unrelated number.

11
00:00:52.267 --> 00:00:55.067
Change the second row to sixteen.

12
00:00:55.067 --> 00:00:57.333
The mean becomes twelve.

13
00:00:57.333 --> 00:01:04.133
The old claim card, narration cue, and displayed number now refer to the previous source.

14
00:01:04.133 --> 00:01:07.700
Create successors for those dependent artifacts while preserving the

15
00:01:07.700 --> 00:01:11.667
earlier version and its provenance.

16
00:01:11.700 --> 00:01:14.500
Follow a dependency from source path,

17
00:01:14.500 --> 00:01:21.433
locator and hash to a claim, spoken cue, scene, chapter contract and output.

18
00:01:21.433 --> 00:01:28.533
Each transformation needs an identity so a changed
source can invalidate dependent work precisely.

19
00:01:28.533 --> 00:01:34.367
Imagine one retained CSV supporting a ratio in two episodes.

20
00:01:34.367 --> 00:01:40.100
Change that CSV while leaving an unrelated operator proof untouched.

21
00:01:40.100 --> 00:01:47.267
Both measured claims need revalidation; their
dependent scenes and chapters need new identities.

22
00:01:47.267 --> 00:01:53.733
The mathematical proof does not become
false because an unrelated timing file changed.

23
00:01:53.733 --> 00:01:59.200
Preserve the old outputs as historical artifacts and link their successors.

24
00:01:59.200 --> 00:02:05.267
A hash establishes an identity comparison,
not mathematical truth or approval.

25
00:02:05.267 --> 00:02:10.052
Content adoption and production authorization are separate records.

26
00:02:10.052 --> 00:02:10.067
Content adoption and production authorization are separate records.

27
00:02:10.333 --> 00:02:15.800
Suppose a file incorrectly states that the mean of eight and sixteen is eleven.

28
00:02:15.800 --> 00:02:19.500
Two exact copies can have matching hashes.

29
00:02:19.500 --> 00:02:26.300
The hash confirms byte identity, including the mistake; it does not evaluate the formula.

30
00:02:26.300 --> 00:02:32.433
Independently recomputing the arithmetic reveals twelve, so the claim of eleven is wrong.

31
00:02:32.433 --> 00:02:37.033
Provenance and correctness support each other but are different checks.

32
00:02:37.033 --> 00:02:42.133
Both are needed for a trustworthy reproducible explanation.

33
00:02:42.133 --> 00:02:46.833
Only artifacts that depend on the changed table need that data correction.

34
00:02:46.833 --> 00:02:50.867
An independent Boolean identity proof can remain valid.

35
00:02:50.867 --> 00:02:56.233
Track editorial adoption, production completion, and owner acceptance separately so that a

36
00:02:56.233 --> 00:03:01.764
generated file is not mistaken for an approved final master.

37
00:03:01.764 --> 00:03:01.767
generated file is not mistaken for an approved final master.

38
00:03:01.767 --> 00:03:06.800
Bind the rendered output to its actual inputs and tool versions.

39
00:03:06.800 --> 00:03:13.151
A matching hash establishes identity, not mathematical truth or approval.