Skip to content

Force-fit Netflix into the BIQ picker on the current question bank - #72

Open
tig wants to merge 10 commits into
mainfrom
force-fit-netflix
Open

tig wants to merge 10 commits into
mainfrom
force-fit-netflix

Conversation

@tig

@tig tig commented Oct 9, 2026 •

Copy link
Copy Markdown
Member

This is a force fit, on purpose, so Netflix shows up in the BIQ picker. It does not add question text, example packs, or a schema change. The borrowed questions do not ask the Netflix practices. That gap is the strain.

The principles side is kindel/principles#90 (branch force-fit-netflix, company block 10000). Source of the words: https://www.jobs.netflix.com/culture. The proposed questions for Tig to write are in that pull request, not in this bank.

What changed

scripts/sync_from_principles.py added the netflix company shell: set title "Culture Memo," the first-sentence preamble, twenty principles with empty questions, facets stamped from data/facets.json, and aliases from the principle terms plus the facet label. .kindel/last-sync.txt records principles commit 588aad05a93ac59c137fcb4b9872f132f7832d3a.

The diff adds sixteen keys to facetQuestions in data/questions.json. The sync added thirteen. Later edits added bold-bet, let-them-decide, and ask-who-disagrees. Each key lists existing question ids that already have Junior, Senior, and Exec packs. No new id was minted. The seven Netflix principles on existing facets were left to inherit the donor list. Adding ids on those facets would have required every company on the facet to carry them. Three Netflix facets have no facetQuestions key: for-its-audience, after-a-setback, and cut-the-extra-step.

Facet Borrowed from
dream-team Hire and Develop the Best: b5ec3e16, bf907bc5, d55f9793, b7f9d574, 92a3ef89
selflessness Ownership: fc2a4107, 41603be8, 192b8a6a, 3de6ea90
creativity Invent and Simplify: 72fe8953, 2b30a0e6, a9e5b2d9, 73f36f42
bold-bet Think Big: a24ba9d7, ccc7a524
courage Have Backbone, Insist on the Highest Standards, Think Big: f5a5e3e0, 0ed453ee, c4c248f7, a24ba9d7, 82218229
resilience Deliver Results and Think Big: 85dc5cd5, 95d04d47, ccc7a524, efe1cec6
keeper-test Low performer, struggling teammate, promotion: bf907bc5, 0f0c4692, d59e1646, 9d996460
people-over-process Bias for Action and Frugality: 0788c8eb, 4bac1a15, 0e07e86c, e89fa18f
context-not-control Earn Trust and Bias for Action: cf2dba5d, 0788c8eb, db715723, 6ba06d3b
informed-captain Ownership, Are Right A Lot, Have Backbone: 3de6ea90, 86851599, 4d7cb33a, fc2a4107
farming-for-dissent Are Right A Lot and Have Backbone: c418917e, f5a5e3e0, 86851599, 0ed453ee
almost-no-rules Frugality and Bias for Action: 0e07e86c, 0788c8eb, efd390d4, 6d1129ab
uncomfortably-exciting Think Big and Bias for Action: a24ba9d7, ccc7a524, 09c7af08, eb5a224e, 98441249
artistic-expression Think Big and Invent and Simplify: 78ffe24a, eb5a224e, 73f36f42, 2b30a0e6
let-them-decide Bias for Action: 0788c8eb
ask-who-disagrees Are Right, A Lot: c418917e

Where the force fit strains

  1. item_for in scripts/sync_from_principles.py was not changed, so howTo says "Pick two or three questions per assigned Culture Memo." kind is lp because the set name does not end in Value or Factor. Both are in data/questions.json. "Culture Memo" is our label. The page heading is "The Best Work of Our Lives."

  2. Stored questions on every Netflix principle are empty. That is what the sync does. A new facet resolves only through facetQuestions. The ids above are the nearest existing questions, and they ask a neighboring behavior. The keeper test questions ask about a low performer. They do not ask "would you fight to keep them?" Context not control asks about deciding without a manager. It does not ask about giving context and leaving the call with the team. Artistic expression asks about a novel approach. It does not ask about supporting a slate you find objectionable. Almost no rules asks about doing the work with fewer resources. It does not ask about "Act in Netflix's best interests."

  3. Principles on an existing facet inherit that facet's whole donor list. Judgment inherits Are Right, A Lot. Candor inherits Earn Trust, including questions that are not about feedback. Great and Always Better inherits the better-every-day donor. Inclusion inherits GitLab belonging. There is no per-principle filter in this bank.

  4. Sync copies the facet label into aliases. "are right a lot" finds Netflix Judgment. "earn trust" finds Candor. "belonging" finds Inclusion. "have backbone, disagree and commit" finds Disagree Then Commit. "better every day" finds Great and Always Better. The alias is how the facet is named, not how Netflix names the principle. See the alias lists in data/questions.json.

  5. The preamble in data/questions.json is one sentence. The memo's sentence about four core principles cannot be stored without an em dash sentence in front of it. The picker introduces a twenty-principle set whose preamble does not say four.

  6. Practices the memo states and this bank cannot ask, because there is no question text and no pack: the keeper test, context not control as giving context, farming for dissent and whose opinion weighs more, artistic expression and the slate, the five-word expenses policy, the Dream Team standard, the informed captain, extraordinary candor, top of market pay, and improving the culture rather than preserving it. The principles pull request lists the questions to write. They are not in this diff.

Porridge is unchanged. Its picker reads the principles index. Singles stay 404 until a porridge sync.

Add the Netflix company shell from the principles force-fit and point the thirteen new facets at existing question ids. No new question text. The borrowed questions do not ask the Netflix practices.
@tig

tig commented Oct 9, 2026

Copy link
Copy Markdown
Member Author

@codex review

@chatgpt-codex-connector

Copy link
Copy Markdown

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🟢 Approval recommended

The intentional force-fit is documented, and all mapped IDs resolve to complete example packs.

0 open findings

What changed in this PR

Adds Netflix to the BIQ picker using existing facet-based questions and example packs.

Changes:

  • Adds 20 Netflix principles and 13 facet mappings.
  • Records the synchronized principles commit and manual mappings.
File Description
data/​questions.json Adds Netflix and borrowed question mappings.
.kindel/​last-sync.txt Records the upstream sync details.

🧠 Review effort: Balanced


💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

@tig
tig marked this pull request as ready for review October 9, 2026 23:32
@tig tig self-assigned this Oct 9, 2026
@chatgpt-codex-connector

Copy link
Copy Markdown

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

The principles record now sits on help-anyway for the time-to-help row, so the picker should inherit that facet's questions too.
@tig
tig requested a balanced review from Copilot October 10, 2026 00:08
@tig

tig commented Oct 10, 2026

Copy link
Copy Markdown
Member Author

@codex review

@chatgpt-codex-connector

Copy link
Copy Markdown

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🔵 Needs a closer look

Netflix is incorrectly marked as examples-disabled despite using existing packed questions.

0 open findings

Previously missed (1)

In code that hasn't changed since last review

Medium severity Enable Netflix examples to match its packed question IDs

data/​questions.json:4011

Netflix resolves every principle to existing packed question IDs, so leaving the company examples-disabled conflicts with the bank contract (README.md:76-78) and causes consumers such as scripts/check_question_ids.js:79-94 to skip its example validation. The current browser happens to expose inherited IDs anyway, which makes the metadata disagree with the UI. Mark this company enabled now that its borrowed packs are present.

🧠 Review effort: Balanced

@tig

tig commented Oct 10, 2026

Copy link
Copy Markdown
Member Author

examples stays false. The shell stores no questions of its own, same as Toyota and Blue Origin. The Examples button uses the packs on the inherited questions. The flag is about whether this company's own stored questions carry packs, and Netflix has none stored.

@tig
tig requested a balanced review from Copilot October 10, 2026 00:37
@tig

tig commented Oct 10, 2026

Copy link
Copy Markdown
Member Author

@codex review

@chatgpt-codex-connector

Copy link
Copy Markdown

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🔵 Needs a closer look

The deliberate semantic gaps between Netflix practices and borrowed questions require final human judgment.

0 open findings

🧠 Review effort: Balanced

@tig

tig commented Oct 10, 2026

Copy link
Copy Markdown
Member Author

The gap is the force fit. Borrowed questions are the existing bank, mapped by facet. New question text stays in the principles pull request for Tig, because a new question needs example packs this change does not add. examples stays false for the same reason as Toyota and Blue Origin: Netflix stores no questions of its own.

@tig

tig commented Oct 10, 2026

Copy link
Copy Markdown
Member Author

The facet lists follow the principles composition. Dream team also inherits build-extraordinary-teams and raise-the-bar. Selflessness also inherits stay-humble. Creativity also inherits listens-with-an-open-mind. Courage also inherits earn-trust. Uncomfortably exciting also inherits courage. No new question text.

@tig
tig requested a balanced review from Copilot October 10, 2026 01:14
@tig

tig commented Oct 10, 2026

Copy link
Copy Markdown
Member Author

@codex review

@chatgpt-codex-connector

Copy link
Copy Markdown

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

@tig
tig marked this pull request as ready for review October 10, 2026 03:48
@chatgpt-codex-connector

Copy link
Copy Markdown

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

@tig tig self-assigned this Oct 10, 2026
Earn the Trust of Others lists the shared facet, and the search alias is the facet label.
@tig
tig marked this pull request as draft October 10, 2026 03:51
@tig tig removed their assignment Oct 10, 2026
@tig

tig commented Oct 10, 2026

Copy link
Copy Markdown
Member Author

Principles 6eb82e8 put Blue Origin Earn the Trust of Others on ask-who-disagrees and removed the second quiet-room table from listens-with-an-open-mind.

This bank now lists that facet on principle 9009, with the search alias "ask who disagrees". Stored questions stay empty. The principle inherits Amazon's Are Right, A Lot questions through ask-who-disagrees, because Amazon is the first principle on that facet with a stored list, and it still inherits the listens-with-an-open-mind questions.

This pull request is a draft again, and it is unassigned, until this head is reviewed.

@tig
tig requested a balanced review from Copilot October 10, 2026 03:51
@tig

tig commented Oct 10, 2026

Copy link
Copy Markdown
Member Author

@codex review

@chatgpt-codex-connector

Copy link
Copy Markdown

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🟢 Approval recommended

All Netflix principles resolve to valid questions with complete Junior, Senior, and Exec example packs.

0 open findings

🧠 Review effort: Balanced

Blue Origin Deliver Results and Resilience list after-a-setback. Invent and Simplify and People Over Process list cut-the-extra-step.
@tig

tig commented Oct 10, 2026

Copy link
Copy Markdown
Member Author

Principles 472f647 added two shared facets.

after-a-setback is on Blue Origin Deliver Results and Netflix Resilience. Search alias: "after a setback". Neither principle has stored questions, and neither does anyone else on that facet, so this adds no donor list. They keep the questions they already inherit.

cut-the-extra-step is on Amazon, Dawn, and generic Invent and Simplify, Blue Origin Technical Ambition and Simplicity, and Netflix People Over Process. Search alias: "cut the extra step". People Over Process now also inherits Amazon's Invent and Simplify questions, because Amazon is the first principle on that facet with a stored list.

This pull request stays a draft, and it stays unassigned, until principles is ready and this head has a clean review.

@tig
tig requested a balanced review from Copilot October 10, 2026 04:11
@tig

tig commented Oct 10, 2026

Copy link
Copy Markdown
Member Author

@codex review

@chatgpt-codex-connector

Copy link
Copy Markdown

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🔵 Needs a closer look

The PR description documents thirteen new facet mappings, while the diff adds sixteen.

0 open findings

Previously missed (1)

In code that hasn't changed since last review

Low severity Update documentation to list all sixteen facetQuestions mappings

data/​questions.json:122

The PR description says facetQuestions gained thirteen keys and its table lists only those thirteen, but this diff adds sixteen. bold-bet, let-them-decide, and ask-who-disagrees are additional mappings. Update the description and mapping table so the documented scope matches the current diff.

🧠 Review effort: Balanced

@tig

tig commented Oct 10, 2026

Copy link
Copy Markdown
Member Author

The table was short. The diff adds sixteen facetQuestions keys, and the description listed thirteen. bold-bet, let-them-decide, and ask-who-disagrees are in the table now, with the ids already in data/questions.json. Think Big, Bias for Action, and Are Right, A Lot. No code change. for-its-audience, after-a-setback, and cut-the-extra-step still have no facetQuestions key.

@tig
tig requested a balanced review from Copilot October 10, 2026 04:37

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🟢 Approval recommended

The implementation is consistent, with only a minor sync-report omission noted.

1 open finding

🧠 Review effort: Balanced

Comment thread .kindel/last-sync.txt
@tig
tig requested a balanced review from Copilot October 10, 2026 04:50

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🟢 Approval recommended

The mappings, aliases, inheritance behavior, and example-pack coverage are internally consistent and match the stated force-fit design.

0 open findings

1 resolved since last review

🧠 Review effort: Balanced

@tig
tig marked this pull request as ready for review October 10, 2026 05:02
@tig tig self-assigned this Oct 10, 2026
@chatgpt-codex-connector

Copy link
Copy Markdown

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants