A Local Format Adaptation Solves a Problem. Change the Core Bible—or Keep a Local Version?
A Local Format Adaptation Improves a Rule. Change the Core Bible—or Keep a Local Version?
The paragraph is the least valuable thing the local team produced.
A territory tries something, it works, and the sentence describing it goes into the shared bible. Now every producer reads a rule written for a broadcast they have never seen, on a set they don't have, inside a runtime they weren't given. Some will try it. Others will wonder what's wrong with them when it doesn't work.
What a format owner actually has to decide is which of three things to hand over: a new default, an option that names the conditions under which it applies, or a bounded local solution nobody else has to carry. Before any of those, two pieces of work: separate the change from the claim that it worked, then find what has to travel with it.
Separate the change from the claim that it worked
The case below is invented. It is not drawn from a production I've observed or a document I've read, and it should not be treated as a record of any real territory's experience.
An invented studio game format: two teams accumulate points across four scored tasks; after the second, the leading team answers one question — bank the total, locking it, or keep it live through the two remaining scored tasks, where it can be lost. Core rule text, one paragraph. The local edition is commissioned at 52:30 against a core slot of 44:00. Its task supply runs to five tasks per episode: four are scored, and the fifth is a reserve used when a scored task fails in the studio.
The problem the local team set out to fix was not a feeling. Teams arriving at the choice answered it without understanding that the live option puts the total at risk. In the briefing they asked which answer was the safe one. On camera, the host restated the rule — sometimes three times — before a team answered.
Their replacement: an unscored rehearsal turn. Before the live choice, the team plays the same choice once on the reserve task, with its outcome shown on a board that reads, on screen and to the participants, as not counting. Then the live choice runs as written.
Write both versions of the rule as a participant would hear them, before judging anything. If the local version can't be stated as a rule — if it only exists as a habit the local host has — you have a practice, not a candidate for the bible.
Now sort the evidence, because three different kinds of statement are usually sitting in the same email.
Observation. The floor producer reports that the host restated the choice on camera less often after the change. That is a real observation with a real author, and the author introduced the change. It tells you someone noticed a difference. It does not tell you why.
Measurement. The rehearsal ran between five and seven minutes of studio time in nine of the twelve episodes. The edition is 8:30 longer than the core. Nobody estimated this; it was timed. Measurements like this can carry a condition — "it fits" — but not a cause.
Constraint explanation. The local team offers two reasons the problem existed: their everyday verb for the live option carries a safe connotation, and the core wording was translated literally. This is their hypothesis about the defect. It is worth more than the other two items, because it is the only one that points at something portable.
Not in the ledger: any claim that the local edition performed better. No matched comparison exists. Before and after differ in task bank, cast, and transmission week, and the count that looks like a result was kept by the person who made the change. Two broadcasts are not a documented development history. Watching the local edition and the core edition side by side tells you what two broadcasts did; it does not tell you a rule's lineage or a team's reasoning. A rule can be in the paperwork and dropped on the day, which is why the record matters more than the tape.
Then list the conditions that changed at the same time as the rule. If you can't list them, you can't price the change. In this case: the commissioned runtime, the existence of the reserve task, the local verb, and the host's willingness to explain rather than compress.
Trace the dependencies that must travel with it
The rehearsal turn is not one decision. It is five, and each one connects to something a receiving team may not have.
Task supply. The rehearsal needs material that is neither scored nor about to be played. Remove the reserve task and you have two bad options: drop the rehearsal, or improvise it on the material the team is about to play for points. The second one hands the contestants the task. The rehearsal's mechanism depends on the reserve; the reserve exists for a different purpose, which the local production quietly spent.
Episode timing. Five to seven minutes of studio time, plus edit consequences. The live choice moves later in the back half, the break position moves with it, and whatever sat behind the choice gets compressed. The 52:30 slot is the enabling condition, and in this invented case it was commissioned for reasons that had nothing to do with this rule.
Casting relationships. The core format leaves the arithmetic unspoken; the host's habit of compressing is a performance choice territories may have cast for. The rehearsal asks that host to teach before asking them to preside. If the format's tension depends on the host withholding, the change revises the part being cast, not only the page being shot.
Set elements. The rehearsal needs a visual state that reads as not counting — a separate board, a card, a graphic. Without it, an audience scores the rehearsal and the live choice arrives muddled.
Supporting instructions. The participant briefing, the host's rule card, the scoring sheet, the editor's notes on what may be shown, and the rule text itself. The rehearsal is a performance; whether it counts as a play is decided in the rules, and if the rules don't say, the briefing and the scoreboard will disagree on air.
Which downstream decisions stay independent? The four scored tasks, the scoring values, the contestant casting, the final reveal, the prizing. Say so explicitly in the handoff, because a receiving team reading a list of dependencies will assume the list is longer than it is and start redesigning things that were never implicated.
Keep the authorship attached. The local team's contribution is the mechanism and the rewrite of the rule in their own language; the format owner's contribution is the rule text and the reserve convention. Whoever invented the turn should write the rationale paragraph, because they are the only people who know what it was fixing. Laundering their reason into a general improvement claim is how a specific fix becomes a vague one.
Compare a default, a conditional option, and an exception
A 2021 post from The Format People announcing a revised and expanded bible template describes format bibles as transferable production know-how — a document that separates elements a format requires from elements a territory may adapt — and notes that later adaptations can inform subsequent editions. That is a template author's description of his own template, not an independently inspected bible and not a licensing standard; the members-only template wasn't available here. It is useful for one narrow point: treating a local variant as a candidate for the next edition is an established position among practitioners, not an eccentric one. It says nothing about whether any particular change should be inherited.
That judgment is yours, and each route has a test.
Making it the default. A core revision has to be right across the intended operating contexts, not just the one where it was invented. Ask how many current and anticipated territories share the enabling conditions. If most run shorter than the core, a default that requires more time is a change to the product every team must absorb, including teams that never had the comprehension problem. If most productions don't hold a fifth task, the default quietly asks them to buy one.
Offering it as a conditional option. A conditional option has to name when it applies and what else changes with it. "Use the rehearsal turn where the runtime allows" is a default wearing a caveat. The workable version reads: where the commissioned runtime exceeds the core slot by at least eight minutes and where the supply includes one unscored reserve task, the following change together — briefing paragraph, on-screen unscored state, host's rule card, back-half timing plan — and the adopting team accepts the loss of its reserve. A territory that can meet one condition and not the other cannot adopt it, which is a feature.
Keeping a local exception. The exception preserves a bounded solution without asking anyone to inherit it. It costs less now and more later: the reasoning sits in one annex, and when that team's staff turns over, the reasoning goes with them. The underlying defect also keeps getting patched territory by territory, separately, forever.
Weigh the learning against the version complexity. Every option you add is a variant to maintain, a row in the comparison table, and a question in every handoff, and it multiplies against the number of territories you support. If the evidence for a change is one producer's count in one territory, a default is an expensive way to find out you were wrong.
Here is the move that matters most in this case. The rehearsal is a mechanism for buying time to state the consequence twice, in different words, while the stakes are real. If the defect is wording, then the wording is the transferable element and the mechanism is the expensive workaround. The local team never tested the wording alone, so nobody knows whether the rehearsal is necessary — but the wording fix has no runtime dependency, no reserve task, and no graphics state. It is a different change with a different footprint, and bundling it with the rehearsal would hide that.
So I'd choose the exception, with one proposal traveling separately. The rehearsal stays local, documented and bounded, with its conditions and its costs written down. The wording change goes upstream as a proposed revision to the core rule, marked as untested, for trial in the territory that has the problem and review before it reaches anyone else. Do not bundle them. The rehearsal diagnosed a wording defect; the wording fix may make the rehearsal unnecessary, and running them together would make that impossible to find out.
Hand over an identifiable operating version
A recipient needs to be able to answer four questions without asking you: what am I being asked to do, what else has to change if I do it, who approved it, and what happens if I don't. Everything else in the note serves those four.
Show what each recipient receives — rule text as it will be spoken, rationale, dependencies, applicable conditions, approval status, author, and whether existing teams are being asked to change or the variant applies only to later transfers. Two distinctions do most of the work here. A proposed revision is not an approved edition, and a document update is not authority. Whether a team is permitted to run a variant is settled by the agreements in place, not by the edition number on a bible; nothing in an editorial handover grants or withholds that. If the licensing side needs a decision, it is a separate conversation with the people who hold it.
A specimen, invented like the rest:
Local variant LV-04 — local-only. Change: unscored rehearsal of the bank/live choice, played on the reserve task before the live choice. Solved: teams answered the live choice without understanding that the live option puts the total at risk; the host was restating the rule on camera. Requires: commissioned runtime at least 8 minutes above the core slot; one unscored reserve task available on the day; an on-screen state that reads as not counting; briefing paragraph 12 replaced; back-half timing plan re-cut. Does not require: any change to the four scored tasks, the scoring values, contestant casting, or the final reveal. Status: local exception. Not part of the shared edition. Other teams are not asked to change. Origin of change: local producer, with rationale written by the originating team. Related: proposal P-07, plain-language rewrite of the core rule text, no runtime dependency. Submitted for review; not trialed; not approved.
The failure case is worth writing into the same note, because it is the one a receiving team is most likely to walk into. A version that copies the rule text without the extra time or the reserve task will either rehearse on the material the team is about to play for points, or truncate the rehearsal until the host delivers one clause instead of two. The second reinstates the exact condition the change was invented to fix, and it will look like the change didn't work.
Read the note back and count what it asks of the reader. It should say what they get, what it costs, which version they are holding, and that nobody has been asked to move on evidence they haven't seen. When the evidence supports only a local solution, the shared edition should say so plainly. A bible that implies a universal improvement on one territory's floor count is promising something nobody has tested.
Frequently asked questions
What three routes are available for a local variation that seems to work?
A format owner can make it a new default, offer it as a conditional option that names when it applies and what else changes, or keep it as a bounded local exception. Each route has different costs in version complexity and transferability.
What kinds of evidence should be separated before revising a core bible?
Separate observation (the host restated the rule less often), measurement (the rehearsal ran five to seven minutes in nine of twelve episodes), and constraint explanation (a local verb carried a safe connotation and the core wording was translated literally). The first tells you someone noticed a difference, the second can support a condition like 'it fits', and the third is a hypothesis about a portable defect. No matched comparison exists, so the local edition's performance cannot be claimed.
What has to travel with the rehearsal turn if another team adopts it?
It requires an unscored reserve task, an on-screen state that reads as not counting, extra runtime—five to seven minutes of studio time plus edit consequences—changes to the briefing, the host's rule card, the scoring sheet and the editor's notes, and a re-cut back-half timing plan. It also revises the host's role from withholding to teaching and depends on the commissioned runtime.
Why not simply make the local rehearsal a default for every territory?
A default must work across intended operating contexts, not just the one where it was invented. This rehearsal requires at least eight minutes above the core slot and one unscored reserve task; territories that lack those conditions cannot adopt it, and a default would impose extra time or a fifth task on teams that never had the comprehension problem. The author instead keeps it as a local exception and sends a separate wording proposal upstream.
What separate change does the author suggest, and why not bundle it with the rehearsal?
The separate proposal is a plain-language rewrite of the core rule text, submitted for review but not trialed or approved. The rehearsal is a mechanism for buying time to state the consequence twice while stakes are real; if the underlying defect is wording, the wording fix has no runtime dependency, no reserve task and no graphics state. Bundling them would make it impossible to learn whether the wording fix alone is enough.