skills/ai-film-pipeline-master/references/phase-01-story-generation/formats-kids.md

# Formats — children's and educational

Entry point into the database. Each kit lists technique ids; open the matching `cat-*.md` for the full card.

Children's formats are the only ones in this database where **the audience's developmental stage is a hard constraint, not a style choice.** A three-year-old cannot hold a delayed payoff across five minutes; a nine-year-old is insulted by one that arrives in thirty seconds. Every rule here follows from that, and the age band is the first decision, before premise.

One standing warning that applies to every kit in this file: **writing down is the universal failure.** Children detect condescension faster than adults and punish it by leaving. Simpler vocabulary, yes. Simpler feeling, never.

---

<!-- GENERATED:toc — do not edit by hand. Generated in the source package -->

## Contents

- [Preschool (2–5)](#preschool-25)
- [Kids' Song](#kids-song)
- [Children's Picture Book](#childrens-picture-book)
- [Educational and Explainer for Children](#educational-and-explainer-for-children)
- [Middle Grade and Older Children (8–12)](#middle-grade-and-older-children-812)
- [Early Reader and Chapter Book (5–9)](#early-reader-and-chapter-book-59)
- [Young Adult](#young-adult)
- [Kids' Non-Fiction](#kids-non-fiction)
- [Interactive and App-Based Children's Content](#interactive-and-app-based-childrens-content)
- [Kids' Content Across Translation and Dubbing](#kids-content-across-translation-and-dubbing)
- [Faith and Heritage Content for Children](#faith-and-heritage-content-for-children)
- [Coverage](#coverage)
- [Sources](#sources)

<!-- /GENERATED:toc -->

## Preschool (2–5)

**Constraints:** Attention runs in 3 to 7 minute units, often watched repeatedly and frequently with an adult in the room. Vocabulary is limited but comprehension is not — a preschooler understands far more than they can say. Cause and effect must be immediate and visible; anything delayed beyond a scene is lost. No irony, since irony requires holding two meanings at once. Repetition is not a concession, it is the mechanism: a child who can predict the next line is a child participating. Emotional content can be serious; the *resolution* must be reachable.

**Grammar:** One idea per episode, stated, demonstrated and restated. A three-part repeating structure — try, fail, try again differently, succeed — with the pattern established early enough that the child can anticipate it. Physical comedy reads at this age where verbal wit does not. The adult in the room is a second audience who can be served by one layer above the child's, but never at the child's expense. Direct address works and is native here in a way it is nowhere else.

**Palette:** 3–7 minutes · 1 idea · 3 attempts · a repeating refrain · immediate consequence · resolution the child could reach

**Rules:** One idea per episode, restated at the end | Consequence must follow the action immediately and visibly | Build a refrain and let the child say it before the character does | Physical before verbal | Keep the cast small and constant | The problem must be solvable by something the child could actually do | Never resolve by an adult arriving | Repetition is the method, not a compromise

**Examples:** preschool animation series, nursery-rhyme and song-led content, bedtime stories, first-concept videos

**Avoid:** Irony and sarcasm, which do not land and do land badly with parents. Peril the child cannot see the exit from. A lesson stated by an adult character, which converts the story into instruction. Jokes for the adult that cost the child the beat.

**The kit (ranked):**

1. **The Story Spine** (Structures) `struct.story_spine`
   Once upon a time, every day, one day, because of that. Strict causality is exactly what this age needs.
2. **Because of That** (Structures) `struct.spine_because`
   The one test that matters: if any link could be written "and then", a preschooler is lost.
3. **And Ever Since That Day** (Structures) `struct.spine_ever_since`
   The stem most often dropped and the one that tells the child the world is permanently different.
4. **The Rule of Three** (Comedy) `comedy.rule_of_three`
   Three attempts is the structure, not a joke device.
5. **Kids' Comedy** (Comedy) `comedy.kids_comedy`
   What actually makes a four-year-old laugh, which is not what makes an adult laugh at a four-year-old.
6. **Repetition with Variation** (Dialogue) `dial.repetition_variation`
   The refrain that returns with one word changed. The child hears the change and that is the story.
7. **The Kids' Register** (Voice) `voice.kids_register`
   Simple words, unsimplified feeling.
8. **Kid-Readable Subtext** (Subtext) `sub.kid_readable_subtext`
   Physical, shown three times, and never explained by an adult character.
9. **Dual Audience** (Subtext) `sub.dual_audience`
   The adult layer, placed so it costs the child nothing.
10. **The Repetition Turn** (Comedy) `comedy.repetition_turn`
    Stop at three or commit past seven. There is nothing in between.
11. **Physical on the Page** (Comedy) `comedy.physical_on_page`
    At this age the body carries the joke and the emotion both.
12. **Grief That Resolves** (Emotion) `emo.grief_that_resolves`
    The one place this is not a failure — a manageable shape is a duty of care here.
13. **The Explained Theme** (Theme) `theme.explained_theme`
    This card names a failure everywhere else in the database, and preschool is the exception that earns it: a two-year-old has no second pass and no inference budget, so the theme has to be said. Restatement is pedagogy here, not condescension.
14. **The Value Turn** (Scene) `scene.value_turn`
    One per episode, and it must be visible in behaviour rather than felt.

---

**The visual kit (ranked):**

1. **One Subject On A Clean Background** (Kids) `kids.clean_background_subject`
   A two-year-old cannot parse a busy frame and follow a story in it at the same time. The clean background is comprehension, and everything else here assumes it.
2. **Child Eye Level** (Kids) `kids.child_eye_level`
   At the child's height, not the adult's. A preschool piece shot from adult eye level is a piece about being looked down at, and they feel it before they can say it.
3. **No Fast Cuts** (Kids) `kids.no_fast_cuts`
   Attention runs in three to seven minute units and is destroyed rather than held by pace. The slow cut is the format's central discipline and the hardest one to keep.
4. **Wide, Clean, Countable** (Kids) `kids.wide_clean_count`
   Inside a teaching or counting beat the clean wide outranks the tight frame, because the child has to see the objects. This is the one place the vertical fill rule yields.
5. **Point and Name** (Kids) `kids.point_and_name`
   The oldest teaching shot there is: the thing, and a hand indicating it. It survives translation, it survives repetition, and it works at two years old.
6. **High Key, Soft** (Kids) `kids.high_key_soft`
   Watched repeatedly, often before sleep, frequently with an adult in the room. Hard light and deep shadow are doing something this audience did not ask for.
7. **Turn to Camera** (Kids) `kids.turn_to_camera`
   Direct address, and the silence after the question. The length of that silence is the teaching, and it is longer than an adult editor believes.
8. **Colour-Coded Characters** (Kids) `kids.colour_coded_characters`
   Vocabulary is limited and names are hard. Colour is how a preschool audience tracks who is who, and it holds across every episode.
9. **Emotion Face Insert** (Kids) `kids.emotion_face_insert`
   The feeling stated as a frame, because this audience is still learning to read faces and the piece is part of how they learn.
10. **Safe Framing** (Kids) `kids.safe_framing`
   Ranked last and non-negotiable: nothing frightening enters frame from outside it, and nothing arrives faster than the child can follow.

## Kids' Song

**Constraints:** 90 seconds to 3 minutes, built to be sung back, often watched hundreds of times by the same child. The melody carries the memory; the lyric must fit it exactly or the child will not be able to join in. Frequently produced in several languages from one melody, which means the lyric must survive translation without losing its stress pattern. Actions and gestures are part of the text, not decoration.

**Grammar:** The chorus is the product. Verses teach, the chorus consolidates, and the child's participation happens on the chorus, so it must be short, stressed simply, and contain the thing to be remembered. Counting, listing and sequencing structures are native — they give the song a spine a child can hold. A gesture attached to a line doubles retention. The final chorus should differ by exactly one element, so the song ends rather than stops — **but on a song built to be sung back, that one element is never a word the child sings.** See the precedence note below the rules.

**Palette:** 90s–3min · 2–4 verses · 1 short chorus · 1 gesture per line · a countable or listable spine · one change in the last chorus, outside the sung words if the child sings it

**Rules:** Write the chorus first and make it singable by a four-year-old | Match stress to melody before meaning, always | Attach a gesture to every line the child should remember | Use a countable or sequential spine — days, numbers, animals, steps | Keep the chorus under eight words if it is meant to be sung back | Change one thing in the last chorus so the song ends — the arrangement, the voice, the tempo, or a tag after it; not a sung word, if the child is meant to sing it | If it will be translated, check the stress pattern survives before the melody is locked

**Yields to:** `end.last_chorus` in [`cat-end-2.md`](cat-end-2.md) — **in the case where the song is built to be sung back by the child**, which is this kit's own stated premise. That card holds the firmer version of the rule: the sung words of the last chorus stay identical, because a child who has spent two minutes learning a line and meets a different one on the last pass stops singing and listens, and participation is what the ending is for. The song still has to end rather than stop, so put the change where the child does not have to follow it — strip or swell the arrangement, hand the same words to a new voice, slow the final pass, add a gesture, or add a spoken tag *after* the chorus finishes. The kit's rule as written, and its **Avoid** line about ending on the same chorus, apply unchanged to a kids' song the child **listens to** rather than sings back: a lullaby, a story song, a song closing an episode. There, mutate the word. `lyric.kids_repetition_contract` and `lyric.repeatable_chorus` are the supporting cards, and `lyric.congregational_lyric` rules out `lyric.refrain_mutation` outright for anything a room sings together.

**Examples:** alphabet and counting songs, days-of-the-week songs, action and movement songs, bedtime songs, faith and heritage songs for children

**Avoid:** A chorus too long to be sung back. Abstractions in the verse — a child cannot hold them. Rhymes forced at the cost of stress, which is immediately audible when sung. Ending on the same chorus as the one before it *with nothing at all changed around it* — on a sung-back song the words stay and the arrangement, voice, tempo or a tag after the chorus carries the ending; on a song the child only listens to, change the word. See the precedence note under **Rules**, and `end.last_chorus`.

> **If this song becomes a video: a teaching song illustrates its lyric on purpose.**
> A kids' song that teaches colours, numbers, letters, days of the week or animal names takes the
> **Music Video — Teaching Song** kit in
> [`../phase-05-song-generation/formats-music.md`](../phase-05-song-generation/formats-music.md#music-video-teaching-song),
> which exists because the Narrative kit's rules against literal illustration are backwards here.
> A kids' song that tells a *story* rather than teaching a set — a bedtime narrative, a fable set to
> music — stays on **Music Video — Narrative**, where a picture set against the lyric is available
> and often better.
>
> The supporting craft is already in this phase: `lyric.on_screen_lyric_text`,
> `expo.kids_repetition_ladder` and `lyric.kids_repetition_contract`, all tagged to Kids Song in
> [`by-format-kids.md`](by-format-kids.md).

**The kit (ranked):**

1. **Repetition with Variation** (Dialogue) `dial.repetition_variation`
   The engine of every kids' song that works. The refrain returns; one word moves.
2. **The Rule of Three** (Comedy) `comedy.rule_of_three`
   Three of anything is the count a small child can hold and predict.
3. **Write for the Ear** (Voice) `voice.write_for_the_ear`
   This is a sung format with no rewind and a listener who cannot read.
4. **The Kids' Register** (Voice) `voice.kids_register`
   The vocabulary ceiling, and the feeling floor that must not be lowered with it.
5. **The Incantatory Register** (Voice) `voice.incantatory_register`
   A chorus meant to be chanted by a room of children is an incantation, and should be built as one.
6. **The Congregational Lyric** (Lyric) `lyric.congregational_lyric`
   The grammar that makes a song joinable, and the constraints that decide whether a room of
   untrained voices can actually get through it: one octave of range, one unhurried breath per
   phrase, symmetrical syllable counts, and a plural or a second person the whole room can honestly
   mean. Written for worship and communal singing, and it names kids' song and preschool in its own
   **Best in** line. A classroom of four-year-olds is a congregation for this purpose.
   **Not `voice.collective_we`.** That card, despite its name and its first-person plural, is a
   *narrator*: an undefined group — a town, a class, a generation — that observes and judges
   together, that must be *capable of being wrong as a group*, and whose effect is deliberately
   unsettling. Its own **Best in** line reads novel, short story, VO and documentary, with no
   children's format in it, and its **Avoid when** rules out anything needing interiority or a
   personal arc. It is the "we" of *we agreed, afterwards, that we had always known* — not the "we"
   of *we clap our hands*. It is a real card and stays in the database; it is simply not the
   sing-along card, and ranking it here taught the wrong lesson to anyone who opened it.
7. **The Specific Detail** (Emotion) `emo.specific_detail`
   One concrete image per verse. Children remember the particular and lose the general.
8. **The Callback** (Endings) `end.callback`
   The last chorus carrying what the verses put into it.
9. **The Thematic Motif** (Theme) `theme.thematic_motif`
   A returning image or gesture across verses, which is how a song acquires a shape.
10. **Sentence Rhythm** (Voice) `voice.sentence_rhythm`
    Prosody. The single most common reason a kids' lyric cannot be sung.
11. **Kids' Comedy** (Comedy) `comedy.kids_comedy`
    Surprise, exaggeration and physical reversal — not wordplay.
12. **The Iterative Habit** (Time) `time.iterative_habit`
    "Every day we…" — the native tense of the counting and routine song.

---

**The visual kit (ranked):**

1. **Singalong Text** (Kids) `kids.singalong_text`
   Built to be sung back, which means the words are on screen and timed to the syllable, not to the bar.
2. **Everyone Dances** (Kids) `kids.everyone_dances`
   The whole body stays in frame at child eye height. This outranks the tight-framing pressure of every vertical surface it ships on.
3. **Repetition Structure** (Kids) `kids.repetition_structure`
   Watched hundreds of times by the same child. The visual returns are the point rather than a limitation, and they are designed rather than reused.
4. **Call and Response Cut** (Kids) `kids.call_response_cut`
   The cut that leaves room for the answer. Cut on the beat instead and the child is being sung at rather than sung with.
5. **Clap Cue** (Kids) `kids.clap_cue`
   The physical instruction, shown. A song that asks for participation without showing the action gets watched instead of joined.
6. **Colour-Coded Characters** (Kids) `kids.colour_coded_characters`
   Across a series of songs this is how a child knows the cast before they know a single name.
7. **High Key, Soft** (Kids) `kids.high_key_soft`
   The register the format is for. A kids' song lit dramatically is a kids' song that will not be played again at bedtime.
8. **Gentle Push on the Singer** (Kids) `kids.gentle_push_singer`
   The only camera move the form needs, on the chorus, slowly. Anything faster competes with the melody for the same attention.
9. **Mirror Move** (Kids) `kids.mirror_move`
   Actions reversed so the child copying them gets it right. It is the single most-skipped rule in kids' music video and the one parents notice.
10. **Wave Goodbye** (Kids) `kids.wave_goodbye`
   The ending the format has, and the beat where the clean wide beats the close-up because the gesture has to be copyable.

## Children's Picture Book

**Constraints:** 32 pages, conventionally 500 to 800 words, though function should govern rather than the count. Read aloud by an adult to a child, usually many times. The illustrations carry half the narrative and must not be described by the text — what the picture says, the words must not. The page turn is a structural device, not a layout accident. Main character usually a child or an animal; usually third person.

**Grammar:** Built in spreads, so the story has roughly 14 to 16 beats and each one must survive a page turn. The turn is the format's cliffhanger: the strongest technique in the form is a line that raises a question on the left and answers it after the turn. Text and image should carry different information, and the gap between them is where the humour and the meaning live. Because it is read aloud repeatedly, the language must reward the adult on the tenth reading as well as the child on the first.

**Palette:** 32 pages · 14–16 spreads · 500–800 words · 1 question per page turn · text and picture never saying the same thing

**Rules:** Write to the page turn, not to the paragraph | Never describe what the illustration will show | Raise a question on one spread and answer it after the turn | Read it aloud at every draft; the page lies | Keep the cast countable on one hand | Make the ending visible in a picture, not stated in a line | Reward the adult reader without charging the child for it

**Examples:** classic and contemporary picture books, concept books, bedtime books, heritage and folk tales retold for children

**Avoid:** Text that narrates the illustration. A rhyme scheme maintained at the cost of sense — unrhymed is better than forced. A moral stated in the last line. More than a handful of named characters.

**The kit (ranked):**

1. **The Story Spine** (Structures) `struct.story_spine`
   Eight stems map almost exactly onto a 32-page structure.
2. **Kishōtenketsu** (Structures) `struct.kishotenketsu`
   The four-part turn, which is the native shape of a great many picture books and requires no antagonist.
3. **The Specific Detail** (Emotion) `emo.specific_detail`
   With 700 words, every noun has to earn its place.
4. **Repetition with Variation** (Dialogue) `dial.repetition_variation`
   The refrain the child joins in on by the third reading.
5. **The Rule of Three** (Comedy) `comedy.rule_of_three`
   Three spreads of attempt, then the turn.
6. **Kid-Readable Subtext** (Subtext) `sub.kid_readable_subtext`
   Carried in the picture, not the text — which is what the format is for.
7. **Dual Audience** (Subtext) `sub.dual_audience`
   The adult is reading this for the twentieth time. Give them something.
8. **The Final Image** (Endings) `end.final_image`
   A picture book ends on a picture. The last spread is the thesis.
9. **Opening Image, Answered** (Openings) `open.opening_image_answered`
   First spread against last spread is the whole arc, and a child can read it.
10. **The Object That Carries It** (Emotion) `emo.object_carrier`
    An object survives across spreads where an abstraction cannot.
11. **The Kids' Register** (Voice) `voice.kids_register`
    Read aloud, so the words are also sounds.
12. **Sentence Rhythm** (Voice) `voice.sentence_rhythm`
    The adult reading aloud is performing. Write something performable.
13. **Voice of a Place** (Voice) `voice.voice_of_a_place`
    For heritage and folk material, where the setting is the subject.

---

**The visual kit (ranked):**

1. **Ink and Line** (Medium) `medium.ink_line`
   Thirty-two pages illustrated by one hand. The line is the book's identity and it is declared before the first spread, not discovered on page twelve.
2. **Watercolour** (Medium) `medium.watercolour`
   Named here as the register the form defaults to, which is both why it works and why choosing it needs to be a decision rather than a habit.
3. **Paper Changes the Image** (Print) `print.paper_changes_the_image`
   The book is the object. Uncoated stock eats contrast and warms every colour, and an illustration approved on a screen has not been approved.
4. **Bleed and Trim** (Print) `print.bleed_and_trim`
   A spread crossing the gutter loses its centre to the binding. It is the constraint that most changes how a picture-book image is composed.
5. **Negative Space** (Composition) `comp.negative_space`
   Where the text goes. A picture book illustrated edge to edge is a picture book the words have to be printed on top of.
6. **Type Over Image** (Print) `print.type_over_image`
   Read aloud by an adult to a child who is looking at the picture. The text must be legible against the illustration under a bedside lamp.
7. **Colour-Coded Characters** (Kids) `kids.colour_coded_characters`
   A pre-reader tracks the cast by colour and shape across thirty-two pages, exactly as they do on screen.
8. **CMYK Gamut** (Print) `print.cmyk_gamut`
   The saturated blues and greens that look right on a monitor are the ones that print dull. Illustrate in the gamut the book will be printed in.
9. **Rule of Thirds** (Composition) `comp.rule_of_thirds`
   Applied to the spread rather than the page, because the reader sees two pages at once and never one.
10. **Proof Before the Run** (Print) `print.proof_before_run`
   Ranked last because it is the last chance. A colour error on a screen is a note; the same error after a print run is the edition.

## Educational and Explainer for Children

**Constraints:** One concept per piece, 2 to 8 minutes, often watched in a classroom or alongside a parent, and frequently required to align with a curriculum point. The learner must be able to state what they learned afterwards — that is the actual success measure, not watch time. Misconceptions are the real obstacle: a child who already believes something wrong will not simply overwrite it.

**Grammar:** Name the misconception, show it failing, then supply the correct model — in that order, because a correct model delivered to a child who holds the wrong one is filed alongside it rather than replacing it. Demonstration beats explanation at every age and overwhelmingly at this one. Every abstraction needs a physical anchor. A question posed and left open for a few seconds before answering produces far better retention than the same fact stated.

**Palette:** 1 concept · 1 named misconception · 1 demonstration · 1 restatement · a question held open before each answer

**Rules:** One concept per piece, no exceptions | Name the wrong version before giving the right one | Demonstrate before you explain | Anchor every abstraction to something physical | Ask the question and leave a gap before answering it | Restate the concept at the end in different words | Say what it is *not*, which is often the faster route | Never let the fun cost the clarity

**Examples:** science and maths explainers for children, language-learning videos, heritage and history for young audiences, safety and health content

**Avoid:** Two concepts. A correct explanation with no wrong version addressed, which leaves the misconception intact. Entertainment that buries the point. Assessment-style language, which converts a video into homework.

**The kit (ranked):**

1. **Exposition as Correction** (Dialogue) `dial.exposition_correction`
   State the wrong version, then correct it. The single most important technique in this format.
2. **The Stated Promise** (Openings) `open.stated_promise`
   Say what the child will know by the end. Being on the nose is correct here.
3. **Rule by Demonstration** (Exposition) `expo.rule_by_demonstration`
   Show the rule working before naming it.
4. **The Explainer Scaffold** (Exposition) `expo.explainer_scaffold`
   The structural spine of the format.
5. **The Familiar Anchor** (Exposition) `expo.familiar_anchor`
   One new term at a time, tied to something already known.
6. **The Jargon Threshold** (Exposition) `expo.jargon_threshold`
   How many new words a piece can carry before comprehension collapses.
7. **The Explained Theme** (Theme) `theme.explained_theme`
   A named failure elsewhere, and a requirement here: an explainer is judged on whether the child can repeat the point afterwards, so the point is stated. Restatement is the method, not a lapse.
8. **The Kids' Register** (Voice) `voice.kids_register`
   Simple words, unsimplified respect.
9. **The Expert Who Explains Badly** (Dialogue) `dial.expert_explains_badly`
   Let the first explanation fail and be asked again. It models the child's own confusion and licenses it.
10. **The Handoff Ending** (Endings) `end.handoff_ending`
    One nameable thing to go and do or notice.
11. **Authority by Specificity** (Voice) `voice.authority_by_specificity`
    A number and a real example, not a claim of expertise.
12. **The Rule of Three** (Comedy) `comedy.rule_of_three`
    Three examples of a concept is the count that makes it stick.

---

**The visual kit (ranked):**

1. **One Subject On A Clean Background** (Kids) `kids.clean_background_subject`
   One concept per piece means one thing in frame at a time. A busy frame makes the child choose what to attend to, and they will choose wrong.
2. **Point and Name** (Kids) `kids.point_and_name`
   The teaching shot. Two to eight minutes of explanation is a sequence of these with the connective tissue removed.
3. **Build It, Do Not Generate It** (Data) `dataviz.build_do_not_generate`
   A diagram in children's educational work is an assertion a child will believe. Assembled from real values it is a diagram; generated it is plausible nonsense a teacher has to unteach.
4. **Object Build** (Kids) `kids.object_build`
   The thing assembled on screen, piece by piece. It is how a process is taught to someone who cannot hold four steps in their head at once.
5. **Letter and Number Insert** (Kids) `kids.letter_number_insert`
   Shown at the moment it is said, large, held. Frequently required to align to a curriculum, and this is the shot the curriculum is checking for.
6. **Wide, Clean, Countable** (Kids) `kids.wide_clean_count`
   Inside a counting beat the objects must be countable in frame. This overrides every instinct to punch in on the action.
7. **Never Colour Alone** (Access) `access.not_colour_alone`
   Watched in a classroom, which means around thirty children, which means at least two who will not read a colour-coded diagram.
8. **Segment Title Card** (Kids) `kids.segment_title_card`
   Structure made visible. A child following a lesson needs to know a new part has started, and a teacher needs somewhere to pause.
9. **No Fast Cuts** (Kids) `kids.no_fast_cuts`
   Comprehension over energy, always, in this format. The cut is slower than the edit wants and slower again than the platform rewards.
10. **High Key, Soft** (Kids) `kids.high_key_soft`
   Even, unthreatening light, because the piece is asking the child to concentrate rather than to feel something.

## Middle Grade and Older Children (8–12)

**Constraints:** Longer attention, genuine capacity for irony, and a hard resistance to being talked down to. Protagonists are conventionally at the top of the target age or slightly above it. Peril, unfairness and loss are permitted; despair without exit is not. Adults in the story should be limited, absent or unhelpful — the reader needs the child to solve it. This is the age at which a reader first notices and resents a lesson.

**Grammar:** A real plot with real stakes, run by the child characters. Humour is load-bearing and can be verbal now. The emotional register can be serious and the resolution can cost something, but the ending must leave agency with the young protagonist. Friendship and belonging are the dominant thematic engines. Subtext works at this age; it is the first band where a reader can hold two meanings.

**Palette:** child solves it · adults limited · verbal humour available · real cost at the end · agency retained

**Rules:** The child solves it, always | Remove or limit the useful adults, plausibly | Peril yes, despair no | Humour is structural, not decoration | Subtext is available now — use it | Never state the lesson | Let the ending cost something, and let the protagonist choose to pay it

**Examples:** middle-grade adventure and mystery, animated series for older children, family films, school stories

**Avoid:** An adult resolving the plot. A stated moral, which this reader will name and resent. Sanitised stakes, which read as a lie. Protagonists younger than the reader.

**The kit (ranked):**

1. **The Positive Change Arc** (Arcs) `arc.positive_change`
   The dominant arc at this age, and the one the reader is rehearsing.
2. **Want Against Need** (Character) `char.desire_made_physical`
   Concrete want, visible and losable.
3. **The Story Circle** (Structures) `struct.story_circle`
   Eight stations, proportional rather than page-numbered — it fits any length.
4. **The Cost Paid** (Endings) `end.cost_paid`
   This reader can take a real cost and notices when one is withheld.
5. **The Beloved Antagonist** (Antagonist) `anta.beloved`
   Opposition from someone who wants their good. The truest conflict at this age.
6. **Kid-Readable Subtext** (Subtext) `sub.kid_readable_subtext`
   Still the rule for the youngest of this band, loosening upward.
7. **Chosen Family** (Relationships) `rel.chosen_family`
   The band's dominant relationship structure.
8. **The Group Secret** (Relationships) `rel.group_secret`
   A shared secret is the fastest route to a group that feels real.
9. **The Ensemble Grid** (Relationships) `rel.ensemble_as_system`
   Build the friends as a system of differences.
10. **Age-Specific Logic** (Character) `char.age_specific_logic`
    Write from this age's own stakes, not from an adult's memory of them.
11. **The Outsider** (Relationships) `rel.outsider_in_the_group`
    The reader's most reliable point of entry.
12. **Comic Escalation** (Comedy) `comedy.escalation_spiral`
    Verbal and situational humour both work now.
13. **The Value Turn** (Scene) `scene.value_turn`
    Full strength, no longer needing to be visible in behaviour alone.

---

**The visual kit (ranked):**

**This format has no visual kit, and that is the finished answer rather than a gap.** Middle grade is prose. There is no frame to compose, no light to motivate and no cut to place, and a ranked list of visual cards here would be a list nobody could use.

**What it does ship is one image**, and that image is real work: the cover, which is the single largest factor in whether an eight-to-twelve-year-old picks the book up. That is Phase 7 work under the poster and key-art craft, opened for that one image, and it is subject to the same print constraints as any other cover - `print.paper_changes_the_image`, `print.cmyk_gamut` and `print.proof_before_run` all apply. Where the book carries chapter-head illustrations, the picture-book kit above governs them.

## Early Reader and Chapter Book (5–9)

**Constraints:** This is the band where the *reader* is the constraint, not the listener. Early readers, for roughly five to seven, run from about 300 words at the simplest level to 1,500 or so at the top, usually in 32 pages, and are sold against a publisher's own levelling system with controlled vocabulary, controlled sentence length and heavy repetition built into the brief. Early chapter books, for roughly six to nine, run about 4,000 to 15,000 words as short illustrated novels with an illustration on most spreads. Full chapter books and the lower middle-grade shelf above them run roughly 25,000 to 50,000, with a commercial sweet spot near 35,000. Chapters are short by design because the unit of achievement is finishing one. Almost everything in this band is sold as a series, so book one has to establish a returnable premise and a cast that will still be usable in book nine. Illustration is a hard budget line and the ratio falls sharply as the band rises.

**Grammar:** The child is doing two things at once — decoding the words and following the story — and the second collapses the moment the first gets hard. So sentence length is a story instrument here in a way it is nowhere else: a long sentence costs comprehension, and the writer spends it deliberately. Repetition is structural rather than stylistic, because a returning phrase is a rest for a reader who is working. The chapter is the real unit of pace: it ends on a small question and it ends before the child gets tired, and the sense of having finished something is itself the reward that brings them back tomorrow. Concrete nouns and strong verbs carry everything, because an abstraction is both harder to decode and harder to picture. Humour is the engine at this age and it is still mostly physical, tipping verbal at the top of the band. And the protagonist must want something the child can hold in one hand — a lost thing, a wanted thing, a thing that must not be found out — because a want that cannot be pictured cannot be followed across four chapters by a reader who stops every night.

**Palette:** 300–1,500 words for a levelled early reader · 4,000–15,000 for an early chapter book · 25,000–50,000 at the top of the band · chapters of 400–1,200 words · a question at every chapter end · one want, physical and losable · a cast of 3–5 · a returnable premise

**Rules:** Set the band and its word ceiling before the premise | Keep the sentences short and spend a long one on purpose | Make the want a thing the child could point at | End every chapter on a small question and end it early | Build a refrain the child can read correctly the second time and enjoy the third | Use concrete nouns and strong verbs; an adjective is a decoding cost | Physical comedy first, verbal comedy as the band rises | Give the recurring cast a premise that generates book nine, not just book one | Never let an adult solve it | Let the child be the one who notices | Read every page aloud at the child's reading pace, not at yours

**Examples:** *Frog and Toad*; *Elephant & Piggie*; *Mercy Watson*; *Magic Tree House*; *The Bad Guys*; *Dog Man*; *Ivy and Bean*

**Avoid:** Picture-book language stretched over chapters, which reads as a long story with nothing to hold onto. Vocabulary chosen for beauty over decodability in a levelled book, which fails the brief no matter how good the sentence is. Chapters that end on a full stop instead of a question. A protagonist whose want is a feeling. Series premises that resolve in book one. And an ending delivered by a parent or teacher, which this reader recognises instantly as the story being taken away from them.

**The kit (ranked):**

1. **The Kids' Register** (Voice) `voice.kids_register`
   Simple words, unsimplified feeling — and here the simple words are a publisher's specification, not a preference.
2. **The Story Spine** (Structures) `struct.story_spine`
   Eight stems, strict causality, and a shape short enough that a reader who stops for a night can pick it back up.
3. **Concrete Nouns Over Abstract** (Rhythm) `rhythm.concrete_nouns`
   At this band an abstraction costs twice: harder to decode, impossible to picture.
4. **The Over-Taught Rule** (Exposition) `expo.over_taught_rule`
   Named failure everywhere else, and ranked here because the levelled early reader is the exception that proves it: controlled-vocabulary books are *required* to restate the same rule, word and phrase far past the point that would be a defect for any older reader. Use the card in reverse — it tells you exactly which repetitions are doing work and which have simply gone on.
5. **Repetition with Variation** (Dialogue) `dial.repetition_variation`
   The refrain that returns with one word moved. For a reader who is working, a familiar line is a place to rest.
6. **Sentence Length as Tempo** (Rhythm) `rhythm.length_as_tempo`
   The one instrument that is simultaneously a pacing choice and a comprehension cost. Spend the long sentence on the turn.
7. **The Chapter as a Rhythm Unit** (Rhythm) `rhythm.chapter_as_unit`
   The chapter is the child's unit of achievement. End it early, end it on a question, and let finishing be the reward.
8. **Desire Made Physical** (Character) `char.desire_made_physical`
   A want you can point at survives four chapters and a week of bedtimes. A want that is a feeling does not.
9. **The Kids' Run** (Sequence) `seq.kids_run`
   The three-or-four-chapter unit of attempt and escalation that an early chapter book is actually built from.
10. **The Rule of Three** (Comedy) `comedy.rule_of_three`
    Still the structure, not a joke device — but at this age the third attempt can now surprise rather than simply succeed.
11. **Physical Comedy Written for the Page** (Comedy) `comedy.physical_on_page`
    The body carries the laugh, and it carries it in a picture the illustrator will draw. Write the action, not the punchline.
12. **Age-Specific Logic** (Character) `char.age_specific_logic`
    A seven-year-old's stakes are real stakes. Write from inside them, not from an adult's memory of them.
13. **The Value Turn** (Scene) `scene.value_turn`
    One per chapter, small, and visible in what somebody does rather than in what they feel.
14. **Bounded Threat** (Antagonist) `anta.bounded_threat`
    Real danger with a visible edge. This band can take fear; it cannot take fear with no floor under it.
15. **The Specific Detail** (Emotion) `emo.specific_detail`
    With a controlled vocabulary the particular has to come from choosing the right ordinary noun, not from a better adjective.
16. **The Ending a Child Can Carry** (Endings) `end.child_carry_ending`
    The last page is read at bedtime and then slept on. It must be something a child can take into the dark.

---

**The visual kit (ranked):**

1. **Ink and Line** (Medium) `medium.ink_line`
   The band where illustration is doing reading support rather than decoration. A clear line at small reproduction size is the whole requirement.
2. **One Subject On A Clean Background** (Kids) `kids.clean_background_subject`
   The reader is the constraint here, not the listener. A crowded illustration competes with the sentence the child is decoding.
3. **Type Over Image** (Print) `print.type_over_image`
   Early-reader type is large and the illustration has to make room for it. This is a layout decision taken before the drawing, not after.
4. **Negative Space** (Composition) `comp.negative_space`
   Gives a child who has just finished a line somewhere to rest before the next one, which is a real function at this reading age.
5. **Paper Changes the Image** (Print) `print.paper_changes_the_image`
   Chapter books print on cheap uncoated stock. Line survives it and delicate tone does not, which is why the line is ranked first.
6. **Colour-Coded Characters** (Kids) `kids.colour_coded_characters`
   Where the book is in colour, the cast is tracked by colour by a reader who is still working at the level of the word.
7. **Bleed and Trim** (Print) `print.bleed_and_trim`
   A small trim size and a tight gutter. An illustration composed for a picture-book spread does not survive being put into this format.
8. **CMYK Gamut** (Print) `print.cmyk_gamut`
   Two-colour and limited-palette printing is common in this band, which makes the gamut a design constraint rather than a production note.
9. **Rule of Thirds** (Composition) `comp.rule_of_thirds`
   On the page rather than the spread, because at this trim size the two pages are rarely read as one image.
10. **Proof Before the Run** (Print) `print.proof_before_run`
   Line work that has thickened on press is line work that has stopped being readable at this size.

## Young Adult

**Constraints:** Roughly 55,000 to 80,000 words for contemporary YA, with fantasy and science fiction running to 90,000 or 100,000 and anything above that needing a structural reason. Upper middle grade sits just below at roughly 45,000 to 65,000 and is a genuinely different category: same emotional seriousness, different permissions. Protagonists are conventionally 15 to 18. The immediate readership is not only teenagers — a large share of the buying audience is adult — but the book is not written for them and is ruined by trying. Gatekeeping is real and specific: school and library purchasing, challenge lists, and retailer categorisation all constrain what can be on the page, and those pressures have been moving year to year rather than decade to decade (noted 2026-09). First person present tense is close to a house style for the category, which makes choosing it a decision rather than a default. Series structure is commercially expected and artistically optional.

**Grammar:** The category's subject is **the first time a person is responsible for who they are becoming**, which is why the arc is almost always built on a lie the protagonist believes about themselves and why the climax is a choice rather than an escape. The adults are not removed as they are in middle grade — they are present, fallible, and part of the problem, and the protagonist's discovery is usually that an adult they trusted was wrong rather than absent. Interiority is the category's whole engine and its whole trap: the narration has to be genuinely inside a specific person's head, which means it must be allowed to be unfair, embarrassing and wrong. Shame, not sadness, is the characteristic emotion — the fear of being seen as you actually are — and it is what makes the secret such a reliable structural device here. The reader will forgive almost anything except being taught: the moment the book's own opinion becomes visible over the protagonist's shoulder, this reader closes it. And the ending can cost: this is the first band where the protagonist may pay and the payment is the point.

**Palette:** 55,000–80,000 words contemporary · 90,000–100,000 speculative · protagonist 15–18 · 1 believed lie · 1 secret with a social cost · adults present and wrong · a choice at the climax, not an escape · a cost the protagonist chooses to pay

**Rules:** Decide the category boundary — upper middle grade or YA — before the first chapter, because it governs every permission | Build the arc on a lie the protagonist believes about themselves | Make the adults present, fallible and implicated, not absent | Let the narration be unfair, petty and wrong, because a fair narrator is an adult narrator | Write shame rather than sadness | Give the secret a social cost, not just a private one | Never let the book's opinion become visible over the protagonist's shoulder | End on a choice the protagonist makes, and let it cost | Treat first person present as a decision, not a default | Write the friendships with the same seriousness as the romance, because that is how the reader lives

**Examples:** *The Hate U Give*; *Speak*; *We Were Liars*; *The Perks of Being a Wallflower*; *Long Way Down*; *Six of Crows*

**Avoid:** The moral delivered in the last chapter, which is the one failure this reader will name out loud in a review. Adults who are simply absent, which is a middle-grade solution applied to a category that needs them in the room. A narrator sanded smooth into likeability, which removes the only thing that made the interiority worth reading. Trauma used as backstory texture and never paid for. Romance carrying structural weight that the friendships were better placed to carry. And an ending that undoes the cost, which teaches the reader that none of it counted.

**The kit (ranked):**

1. **First Person Central** (POV) `pov.first_person_central`
   The category's near-default, and therefore the choice most in need of being made on purpose. If it is the right voice, commit to its unfairness.
2. **The Lie** (Arcs) `arc.the_lie`
   What the protagonist believes about themselves at page one. In this category it is the spine, not a component.
3. **Age-Specific Logic** (Character) `char.age_specific_logic`
   Sixteen has its own scale of consequence. Written from adult hindsight it flattens into nostalgia and the reader can feel it.
4. **The Tacked-On Moral** (Theme) `theme.tacked_on_moral`
   Named failure, not a recommendation, and ranked this high because it is the defect this category is most prone to and least forgiving of. Read it as the last pass over the final chapter.
5. **Shame and Humiliation** (Emotion) `emo.shame`
   The category's characteristic feeling. Not sadness — the specific fear of being seen accurately.
6. **The Secret** (Character) `char.the_secret`
   Give it a social cost as well as a private one, and the plot will generate itself.
7. **The Cost Paid** (Endings) `end.cost_paid`
   This reader has a very good detector for a cost that was announced and then refunded.
8. **Chosen Family** (Relationships) `rel.chosen_family`
   The band's central relationship structure, and usually the one carrying more weight than the romance it is standing behind.
9. **Show the Realisation, Not the Lesson** (Theme) `theme.realisation_not_lesson`
   The difference between a book a sixteen-year-old recommends and one they were assigned.
10. **The Outsider in the Group** (Relationships) `rel.outsider_in_the_group`
    The reader's entry point, and the position most of them are occupying while reading.
11. **Present-Tense Narration** (Time) `time.present_tense`
    Near-standard here, which is exactly why it should be chosen rather than inherited. It buys immediacy and it costs retrospect.
12. **Holding the Unlikeable Lead** (Character) `char.unlikeable_lead`
    Permission to write a narrator who is wrong, and the technique that keeps the reader with her while she is.
13. **Asymmetric Investment** (Relationships) `rel.asymmetric_investment`
    One person further in than the other. The engine of this category's romances and half of its friendships.
14. **What They Think Is Normal** (Character) `char.normal_baseline`
    The fastest way to establish a household, a school or a faith without a paragraph of explanation, and the mechanism by which a reader recognises their own.
15. **The Three Conditions** (Subtext) `sub.three_conditions`
    Subtext works at full strength here. This card says when it will actually land rather than read as evasion.
16. **The Open Ending** (Endings) `end.open_ending`
    Permitted, and often correct — provided the question left open is the one the book asked, not one it failed to answer.

---

**The visual kit (ranked):**

**This format has no visual kit, and that is the finished answer rather than a gap.** Young adult is prose, at fifty-five to a hundred thousand words. Nothing in it is composed, lit or cut.

**What it does ship is one image**, and that image is real work: the cover, and in this category it is closer to a poster campaign than to a book jacket - it is designed to be recognised on a phone in a feed as much as on a shelf. Phase 7's key-art craft governs it, `access.legible_at_size` decides whether the title survives thumbnail size, and the print cards decide whether the colour survives the press.

## Kids' Non-Fiction

**Constraints:** A non-fiction picture book runs roughly 800 to 1,200 words of main text, longer than its fiction equivalent because it is bought partly as a classroom object; back matter — author's note, glossary, timeline, sources, further reading — is not counted in that word count and is routinely longer than the book. The school and library market is a substantial share of sales, which means a curriculum hook is a commercial requirement and not a nicety. Older bands run to full narrative non-fiction of 25,000 to 60,000 words. Facts must be checkable and sourced, and the audience includes teachers, librarians and specialists who will check them. Photographs and archive images carry rights costs that shape what can be included. And a young reader's misconception is the real obstacle: a correct explanation handed to a child who already believes something else gets filed beside the wrong belief rather than replacing it.

**Grammar:** The form's central move is **narrative technique applied to material that must stay true** — scenes, a subject with a want, tension, a shape — without inventing anything, which is a much harder constraint than fiction's. So the writer's structural decisions are made in what to select and where to stand, because those are the only freedoms left. Name the wrong version before supplying the right one, or the right one has nowhere to go. Every abstraction needs a physical anchor a child has actually touched, and one new term at a time, defined by the sentence it stands in. The back matter is where the honest complications go — the dispute, the source that disagrees, the thing nobody knows — and putting them there rather than in the main text is what lets the main text stay a story. Authority comes from a specific number and a real example, never from a claim to expertise, because a child has no way to evaluate the claim and every way to evaluate the example. And the ending hands over: one nameable thing to go and look at, do or ask about.

**Palette:** 800–1,200 words of main text · back matter uncounted and often longer · 1 subject with a want · 1 named misconception · 3–5 new terms across the whole book · 1 curriculum hook stated for the catalogue · 1 handoff at the end

**Rules:** Pick one question the book answers and refuse the second | Name the wrong version before giving the right one | Anchor every abstraction to something the child has physically handled | Introduce one new term at a time, defined by its context | Put the dispute, the doubt and the sources in the back matter, not in the running text | Build authority from a number and a real example | Give the subject a want, because a life without one is a list of dates | Check every fact against a source you could name to a librarian | Say plainly where the evidence runs out | End by handing the child something to go and do

**Examples:** *Bomb: The Race to Build—and Steal—the World's Most Dangerous Weapon*; the young-readers editions of *Hidden Figures* and *The Boy Who Harnessed the Wind*; *Fantastically Great Women Who Changed the World*; the *Horrible Histories* and *Horrible Science* lines

**Avoid:** Invented dialogue and invented interiority presented as fact, which is the line between narrative non-fiction and a novel and which reviewers in this category police closely. Facts with no shape, which is a list and not a book. A correct explanation delivered without naming the misconception it is supposed to displace. Jargon introduced three terms at a time. A back matter that is a bibliography and nothing else, when it could have been the best part of the book. And the tone that treats a child as an audience to be entertained into learning rather than a person who wanted to know.

**The kit (ranked):**

1. **The Explainer Scaffold** (Exposition) `expo.explainer_scaffold`
   The structural spine. Question, wrong model, demonstration, correct model, restatement, handoff.
2. **Exposition as Correction** (Dialogue) `dial.exposition_correction`
   Name the wrong version first. Without it the right version is filed beside the misconception instead of replacing it.
3. **Familiar Anchor, One Strange Word** (Exposition) `expo.familiar_anchor`
   One new term, tied to something already in the child's hands. This is the whole of vocabulary pacing at this age.
4. **Rule by Demonstration** (Exposition) `expo.rule_by_demonstration`
   Show it working before naming it, because a named rule that has not been seen working is a word to memorise.
5. **Authority by Specificity** (Voice) `voice.authority_by_specificity`
   A real number and a real example. A child cannot assess a credential and can absolutely assess a detail.
6. **Theme in Non-Fiction** (Theme) `theme.theme_in_nonfiction`
   The book has a claim whether or not you wrote one. Decide it, or the selection of facts will decide it for you.
7. **The Specific Detail** (Emotion) `emo.specific_detail`
   One concrete particular does more than five accurate generalities, and it is the thing the child repeats at dinner.
8. **The Jargon Threshold** (Exposition) `expo.jargon_threshold`
   How many new words a piece can carry before comprehension collapses. In this band it is a very small number.
9. **The Proper Noun Budget** (Exposition) `expo.proper_noun_budget`
   Names, places and dates are the commonest overload in children's history and biography.
10. **The Optional Appendix** (Exposition) `expo.optional_appendix`
    The back matter, by its proper name, and the single most under-used craft opportunity in the format.
11. **Contradiction Between Sources** (Information) `info.contradictory_sources`
    Children can hold "we are not sure" if you tell them plainly. Putting it in the back matter is how you tell them.
12. **Disputed History** (Information) `info.disputed_history`
    For any subject with more than one community's account, which in children's history is most of them.
13. **Heritage Specificity** (Exposition) `expo.heritage_specificity`
    The named object, the real place, the actual word. It is what separates a book about a culture from a book about a category.
14. **The Two-Touch Rule** (Exposition) `expo.two_touch_rule`
    Two encounters with a fact before it carries weight — and in a book that will be reread, the second touch can be the picture.
15. **The Kids' Register** (Voice) `voice.kids_register`
    Simple words, unsimplified respect. In non-fiction the respect is what makes the child believe the words.
16. **The Handoff Ending** (Endings) `end.handoff_ending`
    One nameable thing to go and do, notice or ask. The measure of this format is what happens after the book is shut.

---

**The visual kit (ranked):**

1. **Build It, Do Not Generate It** (Data) `dataviz.build_do_not_generate`
   Ranked first because this is the format that teaches facts. A generated diagram in a children's non-fiction book is a wrong fact printed in an edition.
2. **Pick the Chart from the Question** (Data) `dataviz.chart_from_the_question`
   Eight hundred to twelve hundred words of main text carrying real information. The graphic states the claim the sentence is making, or it is decoration.
3. **Point and Name** (Kids) `kids.point_and_name`
   The labelled diagram, which is this format's central image and the reason a child reads it more than once.
4. **Never Colour Alone** (Access) `access.not_colour_alone`
   A keyed map or a coded chart in a children's book is exactly the case this rule exists for, and print removes the option of fixing it later.
5. **Ink and Line** (Medium) `medium.ink_line`
   Non-fiction illustration is doing explanation. A line drawing shows a cross-section that a painting cannot.
6. **Type Over Image** (Print) `print.type_over_image`
   Labels, captions and callouts sit on the illustration here more than in any other children's format.
7. **CMYK Gamut** (Print) `print.cmyk_gamut`
   Scientific and natural-history colour is where gamut errors become factual errors: a bird printed the wrong blue is wrong, not just disappointing.
8. **Negative Space** (Composition) `comp.negative_space`
   Non-fiction spreads default to dense. The empty area is what lets a child find the one thing the page is actually about.
9. **Paper Changes the Image** (Print) `print.paper_changes_the_image`
   Coated stock for photographic plates, uncoated for line. The decision changes what can be drawn.
10. **Proof Before the Run** (Print) `print.proof_before_run`
   A non-fiction error survives in print for the life of the edition, and colour is a kind of fact in this category.

## Interactive and App-Based Children's Content

**Constraints:** Privacy law is the first constraint and it is not negotiable. Content directed at children under 13 in the United States requires verifiable parental consent before any personal data is collected, with separate consent for each third party the data reaches, and the definition of "directed to children" has been widened to take in app reviews, marketing and the ages of users on comparable apps; penalties are per violation and substantial. Apple's Kids Category requires an app be built for one of three bands — 5 and under, 6 to 8, 9 to 11 — with a privacy policy and no behavioural advertising. These rules changed in 2025 and again in 2026 and will change again: verify the current FTC rule text and the current store guidelines before designing anything that touches data, advertising or purchases (noted 2026-09). Beyond that: a small child holds the device, taps imprecisely, cannot read, may be pre-verbal, and will repeat the same interaction fifty times. Sessions are minutes long. A parent is often in the room and is always the purchaser.

**Grammar:** The child's verb set is **tap, drag, tilt, and say**, and every narrative idea has to be expressible in one of them or it does not exist. Interaction is the reward, which means the story is paced by the child's hands and not by the writer — a screen the child will not touch is a screen the story stops on. Repetition is the format's currency rather than its risk: the same tap producing the same delight is the entire design of a good preschool app, and the craft is in how much variation sits under the surface of the identical action. There is no fail state in the ordinary sense; the equivalent of failure is a nudge, and it has to come from a character, warmly, within a few seconds, because a stuck child leaves. The instruction is diegetic or it is not followed: a character asking is followed, an on-screen instruction is not, and a written one is invisible. And there are two customers in the room — the child playing and the adult paying — which is the one place a dual-audience layer is a commercial requirement.

**Palette:** age band declared as 5-and-under, 6–8 or 9–11 · 3–8 minute sessions · 1 verb per interaction · a nudge within 5–8 seconds of inaction · no reading required below 6 · every instruction spoken by a character · an adult-facing layer that costs the child nothing

**Rules:** Settle the privacy and store constraints before the premise, because they will rewrite it | Express every story beat in a verb a small child can perform | Let the interaction be the reward, not the gate before the reward | Write the nudge as a character being kind, and fire it early | Assume nothing is read below six and little is read below eight | Make the repeated tap identical on the surface and varied underneath | Give the parent a layer, and never charge the child for it | Design for imprecise fingers and accidental taps | Never put a purchase, a link or a form where a small child can reach it | Write the ending so the session can be stopped at any point without it feeling taken away

**Examples:** the *Toca Boca* and *Sago Mini* catalogues; *Khan Academy Kids*; *Endless Alphabet*; *Nighty Night*; interactive read-along and sticker-book apps built around a single preschool character

**Avoid:** A story that needs the child to read. Instruction delivered by text or by a disembodied narrator instead of by a character. A gate the child cannot pass and no nudge behind it. Advertising or a purchase flow inside reach of a five-year-old, which is a legal problem before it is a taste problem. Adult jokes that cost the child the beat. And the design that treats repetition as a failure and keeps changing the response, which is precisely the opposite of what a three-year-old came for.

**The kit (ranked):**

1. **Pacing Without Control of Pacing** (Interactive) `inter.pacing_without_control`
   The child sets the tempo, and the child will spend four minutes on one duck. Build a story that survives it.
2. **Say It, See It, Ask For It** (Exposition) `expo.kids_repetition_ladder`
   The teaching ladder this format is built on, and the reason the same interaction can be offered three times and be three different things.
3. **The Repeated Beat** (Beat) `beat.repeated_beat`
   Named failure in every other format in this database, and ranked here because this one is the exception: a preschool app's core loop *is* a repeated beat, deliberately identical on the surface so that a small child can predict it and join in. The card still earns its place, because it tells you exactly what has to vary underneath for the repetition to stay alive rather than go dead.
4. **The Problem as Opponent in Preschool and Kids' Song** (Antagonist) `anta.preschool_problem`
   A shoe that will not go on, a puddle in the way. The opposition has to be something a child's own hands could work at.
5. **Accessible Interactive Prose** (Interactive) `inter.accessible_interaction`
   Touch targets, speech rate, no timing requirement, and a route through for a child who cannot read. Authoring decisions, not a retrofit.
6. **The Diegetic Tutorial** (Interactive) `inter.diegetic_tutorial`
   A character asking is followed; an instruction panel is not. There is no second option in this band.
7. **The Kids' Register** (Voice) `voice.kids_register`
   Simple words, unsimplified feeling, spoken aloud by a performer at a pace a three-year-old can follow.
8. **The Fail State as Narrative** (Interactive) `inter.fail_state_as_story`
   Here the fail state is a gentle nudge in character, delivered within seconds. Write it as warmth, never as a correction.
9. **Environmental Storytelling** (Interactive) `inter.environmental_storytelling`
   A child who cannot read still reads the room. The scene is the entire exposition budget.
10. **The Flavour Choice** (Interactive) `inter.flavour_choice`
    Which hat, which colour, which animal. It costs almost nothing and it is where a small child's sense of authorship comes from.
11. **The Two-Audience Layer** (Subtext) `sub.dual_audience`
    Two customers in the room. In this format the adult layer is not a bonus, it is why the app was bought.
12. **Personalisation Without Trespass** (Interactive) `inter.personalisation_limit`
    The creepiness line sits much nearer with a child's data, and the law sits nearer still. Use the child's first name and stop there.
13. **The State Flag** (Interactive) `inter.state_flag`
    One thing the app remembers — a name, a colour chosen last time — is worth more to a four-year-old than a branch.
14. **Voice-First Narrative** (Interactive) `inter.voice_first_narrative`
    Below reading age, speech is the only interface that is not a picture. It changes what the piece can be told in.
15. **The Resolvable Problem** (Endings) `end.resolvable_problem`
    A problem the child could imagine solving, resolved by somebody their size.
16. **Localising Conditional Text** (Interactive) `inter.conditional_localisation`
    Kids' apps ship in many languages and their strings are full of names and counts. The grammar has to survive all of them.

---

**The visual kit (ranked):**

1. **Safe Framing** (Kids) `kids.safe_framing`
   Ranked first because in an interactive piece the child controls the pace, and a frightening element they can hold on screen indefinitely is a different object from one that passes.
2. **One Subject On A Clean Background** (Kids) `kids.clean_background_subject`
   A tappable thing on a clean field. Ambiguity about what is interactive is the commonest design failure in children's apps.
3. **Point and Name** (Kids) `kids.point_and_name`
   The instruction, shown rather than spoken, because the child may have the sound off or may not read.
4. **Legible at the Size It Is Actually Watched** (Access) `access.legible_at_size`
   A tablet held by a five-year-old at fifteen centimetres, often in daylight, often at an angle.
5. **Colour-Coded Characters** (Kids) `kids.colour_coded_characters`
   Navigation for a pre-reader. Colour is the only wayfinding a child who cannot read a label can use.
6. **Never Colour Alone** (Access) `access.not_colour_alone`
   Colour navigation still needs a second cue - shape, position or sound - or part of the audience cannot use the app at all.
7. **High Key, Soft** (Kids) `kids.high_key_soft`
   Held close to the face for long stretches. This is a comfort decision as much as a style one.
8. **Emotion Face Insert** (Kids) `kids.emotion_face_insert`
   Feedback the child understands without words: the app responding to what they did, in a face.
9. **Surprise Box** (Kids) `kids.surprise_box`
   The reward beat. Designed once and reused, it is what makes a child return to an interactive piece rather than finish it.
10. **Accessible Interaction** (Interactive) `inter.accessible_interaction`
   A tap target sized for a child's hand, a path through the piece that does not require reading, and no time limit on a choice.

## Kids' Content Across Translation and Dubbing

**Constraints:** Children's content is translated more heavily than any other category and it is dubbed rather than subtitled below reading age, which changes the constraint completely. A dub is written to the picture: dialogue must fit the existing mouth movements — the open-close-pause patterns animators call mouth flaps — so the adapted line is bound by duration, by syllable count and by where the open vowels fall, and a literal translation almost never fits. That is why dubbed lines diverge from the source: the writer is solving a timing problem and a meaning problem at once, and when the timing is ignored the performer is forced to speak at a tempo no person uses. Songs are harder again: the melody is fixed, so the new lyric must match stress to the same notes while remaining singable by a child. Names, foods, jokes and culturally located references are replaced or kept by a decision that is made once and then has to hold across a hundred episodes. Above reading age, subtitles bring their own ceiling — a child reads more slowly than an adult, so the reading-rate limit is lower and the line count is lower with it.

**Grammar:** The unit of adaptation is **the line's duration, not its meaning**, and everything else is negotiated inside that. So the adapter's first pass is a timing pass — count the syllables, mark the open mouths, find where the character's lips close — and only then a writing pass. Rhythm survives translation better than wordplay does, which is why a refrain, a rhyme scheme and a repeated structure are the parts of a children's script most worth building deliberately: they are what will still be there in eleven languages. Jokes that depend on a word are lost and must be replaced with a joke in the same place doing the same job, which means the source writer's real gift to the adapter is a joke built on situation or on physical action. The song repetition contract is the sharpest case of all: a child's chorus is a promise that it will come back the same, and a translation that varies it to fit the rhyme has broken the one thing the child was holding. And the decisions that matter most are made before episode one — the name list, the register, the address forms — because they cannot be revisited at episode forty.

**Palette:** dubbed lines matched to syllable count and mouth closes · a lower reading rate for child subtitles · a locked name and term list before episode 1 · songs matched to stress, not to meaning · jokes rebuilt in place, not carried over · the refrain identical in every language

**Rules:** Do the timing pass before the writing pass | Match syllables and mouth closes, then find the meaning that fits | Build the comedy on situation and body so it can survive the crossing | Keep the refrain identical across every language, even when the rhyme suffers | Match stress to melody before sense in every sung line | Lock the name, term and address-form list before the first episode is adapted | Set child subtitles to a slower reading rate than adult ones and cut the line count to match | Write the source script with a translator in mind, and mark what may be changed | Replace a lost joke with a joke in the same place, not with an explanation | Check the adapted line by speaking it against the picture, every time

**Examples:** the multi-language dubs of *Peppa Pig* and *Bluey*; the long-running international dubs of *Pokémon* and *Miraculous*; the Studio Ghibli English adaptations; the international co-productions of *Sesame Street*

**Avoid:** A literal translation dropped onto a fixed mouth, which forces the performer into a tempo no person uses and which every child hears without being able to name. Wordplay carried across instead of rebuilt. A chorus altered between language versions to save a rhyme. Names changed at episode thirty that were kept at episode one. Subtitles for children timed at an adult reading rate. And the adaptation that smooths out everything specific in the source, leaving a competent script with nowhere in particular in it.

**The kit (ranked):**

1. **Singable Translation** (Lyric) `lyric.singable_translation`
   The hardest craft in the category. Melody fixed, stress fixed, meaning negotiable, and a child who has to be able to sing it.
2. **What Actually Survives Translation** (Rhythm) `rhythm.translation_survivors`
   Rhythm, structure and image survive; wordplay does not. Knowing which is which is the whole planning decision.
3. **Building a Joke That Survives Translation** (Comedy) `comedy.translation_resistance`
   Situation and body cross borders. A pun is a hole in the script that somebody in another country has to fill.
4. **Writing for a Translator** (Rhythm) `rhythm.write_for_translator`
   Most of the adapter's problems are decided in the source script. Write it knowing eleven people will have to solve it.
5. **The Kids' Song Repetition Contract** (Lyric) `lyric.kids_repetition_contract`
   The child's chorus is a promise. In translation it is the one element that must not move, whatever it costs the rhyme.
6. **Parallel Syllable Lock Across Verses** (Lyric) `lyric.syllable_lock`
   The counting discipline a dub adapter is doing anyway. Applied to the source, it makes every later version cheaper.
7. **Writing for the Subtitle** (Dialogue) `dial.subtitle_constraint`
   Above reading age, and at a lower rate than an adult subtitle, with fewer lines and no dependence on reading speed.
8. **The Translated Sentence That Scans Wrong** (Rhythm) `rhythm.translationese_scan`
   Named failure, not a recommendation, and ranked here as the standing diagnostic of the whole category: a line that is accurate, grammatical and audibly not how anyone in the target language speaks. In a dub it is the commonest defect and the hardest to hear in your own work.
9. **The Kids' Register** (Voice) `voice.kids_register`
   Every target language has its own children's register, and it is not the source language's register translated.
10. **Rhythm in Translation** (Rhythm) `rhythm.translated_rhythm`
    The cadence of the new language, not the cadence of the old one carried across with the words.
11. **Writing a Claim That Survives Translation** (Theme) `theme.translation_survival`
    The thing the episode argues has to arrive intact in a language whose moral vocabulary is shaped differently.
12. **The Untranslatable Moral Word** (Theme) `theme.untranslatable_value`
    Duty, shame, honour, belonging. The place where children's content most often changes meaning without anyone noticing.
13. **The Name as Design** (Character) `char.name_as_design`
    Keep it or replace it, and decide before episode one — including whether it is pronounceable and singable in every target.
14. **Comedy Across Cultures** (Comedy) `comedy.cross_cultural_comedy`
    What is funny to a six-year-old varies less than adults expect, and what is permissible varies a great deal more.
15. **Localising Conditional Text** (Interactive) `inter.conditional_localisation`
    For anything with an app, a game or a name-insertion layer beside the show, which is now most children's properties.
16. **Arabic and Other Right-to-Left Rhythm** (Rhythm) `rhythm.arabic_rtl_rhythm`
    A worked case of a target language whose prosody, script direction and children's register all differ from the source at once.

---

**The visual kit (ranked):**

1. **Two Scripts on One Surface** (Text) `textimg.two_scripts_one_surface`
   Dubbed below reading age and subtitled above it, often both in the same release. When two scripts share a frame they are set for equal reading weight, not equal point size.
2. **Clean Plate Clause** (Text) `textimg.no_watermark_clause`
   Children's content is translated more heavily than any other category. Type burned into the picture is type that cannot be localised, and it is the single most expensive mistake in this format.
3. **Singalong Text** (Kids) `kids.singalong_text`
   Re-timed per language, because the syllables do not land where the original's did. It is re-authored rather than translated.
4. **Subtitle Size in the Frame** (Text) `textimg.subtitle_size_in_frame`
   A second script is set larger than its point-size match to read as equal, and children's subtitles are read by people still learning to read.
5. **Mirror Move** (Kids) `kids.mirror_move`
   Survives every language, which makes it one of the few teaching devices that needs no localisation at all.
6. **Point and Name** (Kids) `kids.point_and_name`
   The most translatable shot in children's work: the hand does the grammar and the word changes underneath it.
7. **Letter and Number Insert** (Kids) `kids.letter_number_insert`
   The shot that does not translate. An alphabet segment is re-made per language rather than re-voiced, and planning for that is the format's main structural decision.
8. **Text Safe Zone Planned in the Image** (Text) `textimg.text_safe_zone`
   A language that runs right to left, or one that needs thirty per cent more characters, needs the room designed in from the start.
9. **Legible at the Size It Is Actually Watched** (Access) `access.legible_at_size`
   Applies per script. Arabic and Syriac lose their dots and strokes before Latin loses its letterforms.
10. **Colour-Coded Characters** (Kids) `kids.colour_coded_characters`
   Carries the cast across every language version without a word, which is why it is worth deciding before the first localisation.

## Faith and Heritage Content for Children

**Constraints:** Commissioned, almost always, by a community institution rather than a broadcaster — a church, a school, a language foundation, a diaspora organisation — which means the client is a committee, the budget is small, the timeline runs to a festival or a school term, and the piece will be shown to the community before it is shown to anyone else. Doctrinal or historical accuracy is usually what the commissioning community cares about most: take it from the brief, then the skill, then a search (root `SKILL.md` §0), and record the choice. Heritage-language content carries an extra, structural constraint: the child's input in that language is thin compared with the abundance in the majority language, so the piece is competing directly with better-funded, faster, louder content the child already has. Sensitivity around sacred texts, images and music varies by tradition; take the tradition's own practice from the brief or a search. And the audience is doubled in a particular way: the child watches, the grandparent watches with them, and the grandparent is the reason it was made.

**Grammar:** The form's whole problem is that **it is teaching a child about something the child already belongs to**, and almost every failure comes from forgetting that. A story written as an introduction to a culture positions the child as a visitor to their own house; the register that is correct for explaining a tradition to outsiders is the wrong register entirely for a child inside it. The craft answer is to write from inside: assume the festival, the word, the grandmother's kitchen, and let the story be about something that happens in that world rather than about the world. Specificity carries the heritage — the named dish, the actual word, the particular tune — because a generic version teaches a child that their inheritance is a category. The sacred material is handled by keeping the story beside it rather than on top of it; a parable has a contract that lets a story make a claim, and using that contract is different from putting an argument in a holy mouth. Repetition, chant and the call-and-response of a congregational line are native here and are the mechanism by which a child acquires the language without being taught it. And the ending must be carryable: this material is watched at bedtime, on a feast day, in a language the child is still building.

**Palette:** 3–8 minutes, or 32 pages · 1 event inside the world, not a tour of it · 3–5 named specifics — a dish, a word, a tune, a place · 1 sung or chanted line the child can join · a grandparent-facing layer · doctrinal check by a named reviewer · an ending a child can take to bed

**Rules:** Write from inside the culture and assume the reader is already in it | Give the story an event, not a syllabus | Name the specific dish, word, tune and place, and never the generic version | Put the chant or the sung line where a child can join it | Keep the story beside the sacred text, not inside its mouth | Let the child solve it, and never let the elder resolve it for them | Write the adult layer for the grandparent who is in the room | Get the doctrinal and historical check from a named person before the draft, not after | Say the heritage word without translating it in the same breath, then make the next line make it clear | End on something a child can carry into the dark

**Examples:** *The Prince of Egypt*; *VeggieTales*; *Shalom Sesame*; *Bino and Fino*; the heritage-language children's catalogues produced by diaspora foundations and language schools

**Avoid:** The tone of a museum label. Making the child's own festival strange in order to explain it. A story that exists to deliver a doctrine, which children detect and resent at exactly the same age they detect any other lesson. Elders who arrive to solve the plot, which removes the child from their own story. Sanitising a tradition into a set of pleasant images with none of its actual difficulty in it. And the failure specific to this format, named plainly: **the piece that teaches a culture about itself in the register used for teaching outsiders about it** — a child watching their own inheritance introduced to them as though they were a visitor learns, very efficiently, that they are one.

**The kit (ranked):**

1. **The Culture as Lesson** (Theme) `theme.culture_as_lesson`
   Named failure, not a recommendation, and ranked first because it is the defining failure of this format and the one every commission drifts toward: the piece that explains a culture to the children inside it, in the register built for outsiders. Read it before the premise, not after the draft.
2. **Writing From Inside** (Theme) `theme.from_inside`
   The corrective, and the entire craft position. Assume the world; tell a story that happens in it.
3. **Heritage Specificity** (Exposition) `expo.heritage_specificity`
   The named dish, the actual word, the particular tune. A generic version teaches a child that their inheritance is a category.
4. **Oral and Heritage Narration** (POV) `pov.oral_narration`
   The tradition already has a way of telling this. Borrowing its voice is more accurate than translating its content.
5. **Subtext a Child Can Actually Read** (Subtext) `sub.kid_readable_subtext`
   Physical, repeated, and never explained by an adult character — which is also how it stops being a lesson.
6. **The Fable Contract** (Theme) `theme.fable_contract`
   The one contract that lets a children's story make a claim without becoming a sermon. Use it deliberately, and know when you have stepped outside it.
7. **Sacred Material and the Limits of Argument** (Theme) `theme.sacred_material`
   Where the story may stand beside the text, and where it may not stand inside it.
8. **The Congregational Lyric** (Lyric) `lyric.congregational_lyric`
   A line a room of children can get through together on the second hearing. This is how a heritage language actually enters a child.
9. **Writing the Believer** (Character) `char.writing_the_believer`
   Faith written as a way of seeing rather than as a plot position, which is the difference between a character and an illustration.
10. **The Liturgical Register** (Voice) `voice.liturgical_register`
    A real register with real rules, and a powerful contrast against the child's own speech when it is placed on purpose.
11. **The Diaspora Character** (Character) `char.diaspora_character`
    Two worlds held at once, which is the actual condition of most of this audience and almost never the subject of what they are given.
12. **The Voice of a Place or an Object** (Voice) `voice.voice_of_a_place`
    A village, a church, a kitchen, a river. The setting carries the inheritance more reliably than any character can state it.
13. **The Audience That Will Correct You** (Information) `info.heritage_audience_correction`
    Someone in the community knows the vestment, the date and the tune. Being right is a craft requirement here, not a virtue.
14. **The Community's Version of a Person** (Information) `info.communal_version`
    The saint, the ancestor, the founder as the community actually tells them — which is the version the child will hear at home.
15. **The Untranslatable Moral Word** (Theme) `theme.untranslatable_value`
    Say it in the heritage language, then let the next line make it clear. That sequence is how the word is inherited rather than glossed.
16. **The Ending a Child Can Carry** (Endings) `end.child_carry_ending`
    Watched at bedtime, on a feast day, in a language they are still building. The last image is what they take with them.

---

**The visual kit (ranked):**

1. **The Reference-First Rule for Heritage** (Reference) `ai.reference_first_heritage`
   Commissioned by a community institution, which means the audience includes people who know. Every historical or religious element is built from a reference or it is not built.
2. **Period-Truthful Body Language** (Pose) `pose.period_body_language`
   A modern gesture in a period scene is the detail an institutional audience sees first, and it is invisible to everyone who made the piece.
3. **One Subject On A Clean Background** (Kids) `kids.clean_background_subject`
   The teaching frame still governs. A child learning an unfamiliar word and an unfamiliar object at once needs the object alone in frame.
4. **Period-Truthful Lettering** (Text) `textimg.period_lettering`
   A script on screen in a heritage piece is a claim about a writing system. Getting it wrong in children's work teaches the error.
5. **Colour-Coded Characters** (Kids) `kids.colour_coded_characters`
   Works here as everywhere, and has to be checked against the period palette rather than chosen from the brand one.
6. **High Key, Soft** (Kids) `kids.high_key_soft`
   The register the commission is for. Dramatic light on religious material in children's content reads as fear rather than as reverence.
7. **An Invented Script That Cannot Be Misread** (Text) `textimg.invented_script`
   Where lettering is needed and no documented form exists, it is invented explicitly rather than approximated - an approximation is a wrong claim about a real script.
8. **Point and Name** (Kids) `kids.point_and_name`
   How an unfamiliar object gets a name. In heritage work it is also how the piece stays specific instead of generically ancient.
9. **Safe Framing** (Kids) `kids.safe_framing`
   Faith and heritage material carries events that are not for this audience at this age, and the framing decision is where that is settled.
10. **Never Colour Alone** (Access) `access.not_colour_alone`
   Applies to any map, family tree or timeline, which this category produces more of than any other children's format.

## Coverage

11 kits: Preschool, Kids' Song, Children's Picture Book, Educational and Explainer for Children, Middle Grade, Early Reader and Chapter Book, Young Adult, Kids' Non-Fiction, Interactive and App-Based Children's Content, Kids' Content Across Translation and Dubbing, Faith and Heritage Content for Children.

**One listed item turned out to be two, and one turned out to be one.** "Early reader and chapter book" is written as a single kit because the two share a reader, a levelling system and a commercial shelf, and the differences between them are numbers rather than craft. "Young adult" is written without a separate upper-middle-grade kit, because upper middle grade is not a different format — it is the same craft with a different permission set, which is stated inside the YA kit's constraints where a writer will actually meet it.

**The named failures ranked in this file, and why.** Children's formats are where this database's failure cards most often turn out to be local rules. `expo.over_taught_rule` is ranked in Early Reader and Chapter Book because a levelled book is required to over-teach. `beat.repeated_beat` is ranked in Interactive and App-Based Children's Content because a preschool app's core loop is one by design. `theme.tacked_on_moral` is ranked in Young Adult and `rhythm.translationese_scan` in Translation and Dubbing as standing diagnostics rather than as exceptions. `theme.culture_as_lesson` is ranked first in Faith and Heritage Content because it is that format's defining failure and naming it is the point.

**Card-shaped gaps found while writing these kits** — techniques these formats need that no card in the database currently covers:

- **Lip-sync and mouth-flap adaptation** — writing a line to a fixed duration, a syllable count and the positions where an animated mouth opens and closes. It is the central craft of dubbing children's animation and there is no card for it. `lyric.singable_translation` is the nearest thing and it solves a different fixed constraint.
- **Decodability and the levelled text** — writing to a controlled vocabulary, a sentence-length ceiling and a publisher's reading level. `voice.kids_register` covers the register; nothing covers the specification.
- **Back matter as craft** — the author's note, glossary, timeline and source list of a children's non-fiction book, which is uncounted, often longer than the text, and frequently the best writing in it. `expo.optional_appendix` is adjacent and is written for fiction.
- **The curriculum hook** — the single sentence that makes a children's non-fiction book buyable by a school, and the structural pressure it applies upstream.
- **The parent-facing purchase layer** — distinct from `sub.dual_audience`, which is about a joke the adult enjoys. This is the adult deciding whether to buy, and it shapes app and picture-book structure directly.

**Shelf life:** dated — review 2027-09. The craft is durable. The word counts, category boundaries, privacy law and app-store rules in the constraint blocks rest on 2026 practice, and the two that move fastest — children's privacy regulation and the Kids Category guidelines — say so in the text.

## Sources

Verified 2026-09-17.

- [How to Write a Children's Picture Book in 2026 — Reedsy](https://reedsy.com/blog/how-to-write-a-childrens-picture-book/)
- [Understanding Children's Book Age Group Categories — Mary Kole Editorial](https://www.marykole.com/childrens-book-age-group)
- [The Age Levels for Children's Books — Journey to Kidlit](https://journeytokidlit.com/age-levels-for-childrens-books/)
- [The Ultimate Guide on How to Write a Children's Picture Book — Journey to Kidlit](https://journeytokidlit.com/how-to-write-a-childrens-picture-book/)
- [Picture Book Word Count and Other Rules — Highlights Foundation](https://www.highlightsfoundation.org/2019/04/08/picture-book-word-count-and-other-rules-meant-to-be-broken/)
- [How to Write for Kids: 10 Tips from R.L. Stine — MasterClass](https://www.masterclass.com/articles/learn-how-to-write-for-kids-with-rl-stine)
- [Children's Book Age Groups Explained](https://getyourbookillustrations.com/understanding-childrens-books-age-groups/)
- [Middle grade literature — Wikipedia](https://en.wikipedia.org/wiki/Middle_grade_literature)
- [The Story Spine — Aerogramme Studio](https://www.aerogrammestudio.com/2013/03/22/the-story-spine-pixars-4th-rule-of-storytelling/)
- [How to Write Books for Early Readers — Mary Kole Editorial](https://www.marykole.com/books-for-early-readers)
- [Explaining Chapter Book Age Range — Mary Kole Editorial](https://www.marykole.com/chapter-book-age-range)
- [How to Write Early Chapter Books: Word Counts, Structure, and Series Tips — Erin Alladin](https://erin-alladin.com/2026/05/13/how-to-write-early-chapter-books-word-counts-structure-and-series-tips/)
- [Children's Book Word Count by Age 2026 — BookFoundry](https://bookfoundry.ai/blog/children-s-book-word-count-by-age-in-2026-your-complete-guide)
- [Word Count for Novels and Children's Books: The Definitive Post — Writer's Digest](https://www.writersdigest.com/whats-new/word-count-for-novels-and-childrens-books-the-definitive-post)
- [Decoding fiction age categories from middle grade to adult — Kristen Pham](https://kristenpham.com/2026/04/24/decoding-fiction-age-categories-from-middle-grade-to-adult/)
- [Writing Middle Grade vs. Young Adult — Amy Koerner](https://www.amykoerner.com/blog/middle-grade-vs-young-adult)
- [Word Count by Genre: What Literary Agents Want (2026 Guide) — The Manuscript Academy](https://manuscriptacademy.com/word-count-by-genre-literary-agents-2026)
- [How to Write Nonfiction Books for Kids — Mary Kole Editorial](https://www.marykole.com/nonfiction-books-for-kids)
- [Writers' Questions About Back Matter — Annette Whipple](https://www.annettewhipple.com/2022/01/writers-questions-about-back-matter.html)
- [Backmatter: When "The End" Isn't The End — Institute of Children's Literature](https://www.instituteforwriters.com/backmatter-when-the-end-isnt-the-end/)
- [A Deep-Dive Into the Types of Nonfiction Picture Books — Children's Book Mastery](https://childrensbookmastery.com/a-deep-dive-into-the-types-of-nonfiction-picture-books/)
- [Mouth Flaps — TV Tropes](https://tvtropes.org/pmwiki/pmwiki.php/Main/MouthFlaps)
- [Lip Lock — TV Tropes](https://tvtropes.org/pmwiki/pmwiki.php/Main/LipLock)
- [Lip-Sync Techniques for Perfect Animation Dubbing — DubNSub](https://dubnsub.com/lip-sync-techniques-for-perfect-animation-dubbing/)
- [COPPA Compliance: key requirements for 2026 — Usercentrics](https://usercentrics.com/us/knowledge-hub/coppa-compliance/)
- [New COPPA Rules 2026: Mobile App Compliance — Respectlytics](https://respectlytics.com/blog/coppa-rules-2026-mobile-app-compliance/)
- [A Guide to COPPA and Mobile Apps — iubenda](https://www.iubenda.com/en/blog/guide-coppa-mobile-apps/)
- [Navigating New Regulations for Children's Apps — MetaRouter](https://www.metarouter.io/post/navigating-new-regulations-for-childrens-apps)
- [Designing an interactive audio narrative for children — The Writing Platform](https://thewritingplatform.com/2021/04/designing-an-interactive-audio-narrative-for-children/)
- [Children's documentaries: distance and ethics in European storytelling about the wider world — Taylor & Francis](https://www.tandfonline.com/doi/full/10.1080/17482798.2021.1974502)
- [Religion in Children's Visual Media: Preschool Holiday Specials — Journal of Media and Religion](https://www.tandfonline.com/doi/abs/10.1080/15348423.2020.1812339)
- [Raising Children in the Diaspora: Messages Shape Identity — Cultural SEL](https://culturalsel.com/raising-children-diaspora-identity/)
- [Multimedia Input and Bilingual Children's Language Learning — PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC7438791/)
- [Connecting Children to Their Cultures and Communities — Community Playthings](https://www.communityplaythings.com/resources/articles/connecting-children-to-their-cultures-and-communities)