Quality Standards for Interactive Fiction¶

Craft guidance for evaluating and maintaining quality in interactive fiction through validation bars covering structure, clarity, safety, and coherence.

Overview¶

Quality assurance in interactive fiction differs from linear narrative because of branching paths, state dependencies, and player agency. These bars adapt industry QA practices to IF-specific concerns.

Industry grounding:

Technical QA: Structural validation (ChoiceScript's quicktest, Twine link checking)
Content QA: Narrative consistency, beta-testing feedback
Accessibility QA: WCAG compliance, inclusive design
Domain-specific: Some bars (Canon, Research Posture) apply particularly to fact-based or world-rich fiction

Bar Categories¶

Category	Bars	Focus
Technical	Integrity, Reachability	Does it work?
Clarity	Comprehension, Style	Is it understandable?
Content	Safety, Accessibility	Is it responsible?
Consistency	Canon, Spoiler	Does it cohere?
Research	Research Posture	Is it grounded?

Bar 1: Integrity¶

Definition: Structural completeness—all pieces fit together correctly.

What Integrity Checks¶

References resolve: Every link points to existing content
Fields present: Required data exists and is valid
Schema compliance: Artifacts match their type definitions
No orphans: All content connects to the structure

Common Integrity Failures¶

Issue	Example	Fix
Dangling reference	Choice leads to nonexistent passage	Create missing passage or redirect
Missing required field	Passage without title	Add required content
Type mismatch	String where number expected	Correct data type
Circular dependency	A requires B requires A	Restructure dependencies

Automated Validation¶

Integrity can be largely automated:

Schema validation catches type/field errors
Graph traversal finds dangling references
ChoiceScript's quicktest detects missing labels and syntax errors
Twine validates link targets

Bar 2: Reachability¶

Definition: Player access—all content can be reached through valid play.

What Reachability Checks¶

Forward reachability: Can players reach every passage from the start?
Gate obtainability: Can every gated path be unlocked?
No dead ends: Do all paths eventually lead somewhere meaningful?
Recovery possible: Can players recover from mistakes?

Common Reachability Failures¶

Issue	Example	Fix
Unreachable passage	No path leads to lighthouse scene	Add route or remove passage
Unobtainable gate	Need item that doesn't exist	Create obtainment path
Dead end	Choice leads nowhere	Add continuation
Permanent lock-in	Early choice blocks critical content	Add alternate paths

Testing Reachability¶

Automated: Graph traversal, randomtest (ChoiceScript runs random playthroughs)
Manual: Systematic branch coverage during playtesting
Hybrid: Automated detection, human verification of edge cases

Bar 3: Comprehension¶

Definition: Player understanding—readers grasp what's happening and what choices mean.

What Comprehension Checks¶

Plot clarity: Can players summarize what happened?
Choice distinction: Are options meaningfully different?
Motivation clarity: Do players understand why things happen?
Spatial/temporal logic: Is setting and timeline clear?

Common Comprehension Failures¶

Issue	Example	Fix
Unclear stakes	Players don't know what's at risk	Establish consequences
Synonym choices	"Go left" / "Head left" / "Turn left"	Differentiate intent
Missing context	Action assumes knowledge player lacks	Provide setup
Confusion cascade	One unclear element creates more	Clarify root cause

Detecting Comprehension Issues¶

Comprehension requires human testing:

Post-play summary: "What was the story about?"
Choice explanation: "Why did you pick that option?"
Confusion moments: Watch for re-reading, hesitation
Unexpected outcomes: "I thought that would..."

See Testing Interactive Fiction for methodology.

Bar 4: Style¶

Definition: Voice consistency—the work reads as if one author wrote it.

What Style Checks¶

Register stability: Does formality stay consistent?
Vocabulary coherence: Are word choices appropriate throughout?
Tone maintenance: Does mood match across scenes?
Period accuracy: Historical settings avoid anachronisms?

Common Style Failures¶

Issue	Example	Fix
Register shift	"My liege" then "What's up"	Maintain appropriate register
Anachronism	Modern slang in Victorian setting	Use period-appropriate language
Tone break	Comedic scene in tragedy without purpose	Align with overall tone
Voice drift	Character sounds different across scenes	Review for consistency

Style Validation¶

A consistent B+ voice beats inconsistent A+ fragments. Readers notice jarring shifts even if individual passages are excellent.

See Voice Register Consistency for detailed guidance.

Bar 5: Safety¶

Definition: Harm prevention—sensitive content handled responsibly.

What Safety Checks¶

Content warnings present: Sensitive material flagged appropriately
No gratuitous content: Dark material serves story purpose
Harmful stereotypes avoided: Representations are thoughtful
Age-appropriate: Content matches target audience

Common Safety Failures¶

Issue	Example	Fix
Missing warning	Violence without content note	Add specific warning
Gratuitous darkness	Shock value without narrative purpose	Remove or justify
Harmful trope	Stereotyped character treatment	Revise representation
Audience mismatch	Adult content in middle-grade work	Adjust content or audience

Safety Review¶

Check content warnings against actual content
Evaluate purpose of sensitive material
Consider diverse reader perspectives
Consult sensitivity readers for relevant topics

Bar 6: Accessibility¶

Definition: Inclusive access—players with disabilities can engage with the work.

What Accessibility Checks¶

Screen reader compatibility: Text is properly structured
Keyboard navigation: All interactions work without mouse
Color independence: Meaning not conveyed by color alone
Reading level: Prose matches target audience capability
Timing flexibility: No time-limited interactions without options

Common Accessibility Failures¶

Issue	Example	Fix
Image-only content	Critical info in untagged image	Add alt text or text equivalent
Low contrast	Light gray text on white	Increase contrast ratio
Color-coded choices	Red=danger, green=safe with no other indicator	Add text/icon indicators
Rapid timing	Must click within 3 seconds	Allow timing adjustments

Accessibility Standards¶

WCAG 2.1 provides baseline guidance:

Perceivable: Content available to all senses
Operable: Interface works with various inputs
Understandable: Content and operation are clear
Robust: Works with assistive technologies

See Accessibility Guidelines for detailed standards.

Bar 7: Canon¶

Definition: World consistency—content matches established truth.

What Canon Checks¶

Fact alignment: Does content match the world bible?
Character consistency: Do characters behave as established?
Timeline coherence: Do events fit the chronology?
Rule compliance: Does content follow world rules?

Common Canon Failures¶

Issue	Example	Fix
Contradictory fact	Character dies in chapter 3, appears in chapter 5	Align with canon
Timeline error	Event happens before its cause	Reorder or clarify
Character break	Peaceful character initiates violence without reason	Justify or revise
Rule violation	Magic works differently than established	Follow world rules

Canon Validation¶

Cross-reference against world bible
Check timeline positioning
Verify character behavior against profiles
Track state dependencies across branches

Bar 8: Spoiler Hygiene¶

Definition: Player experience—discovery is preserved, secrets stay hidden.

What Spoiler Checks¶

No early reveals: Plot twists not exposed before their time
Reference safety: Codex/glossary doesn't spoil story
Gate text safety: Locked content hints don't reveal too much
Surface separation: Player-facing content is clean

Spoiler Classification¶

Level	Definition	Treatment
Minor	Background flavor	May appear with careful phrasing
Major	Affects understanding/strategy	Must be omitted or heavily obscured
Critical	Would ruin experience	Never appears, no exceptions

Common Spoiler Failures¶

Issue	Example	Fix
Codex spoiler	Entry reveals villain's identity	Remove from codex
Gate hint	Lock text reveals what's behind door	Obscure hint
Early foreshadowing	Too-obvious setup spoils twist	Increase subtlety or remove
Meta-knowledge	Player knows what character shouldn't	Separate knowledge layers

Bar 9: Research Posture¶

Definition: Factual grounding—claims about real-world facts are appropriately supported.

Note: This bar applies primarily to fiction engaging with real history, science, or current events. Pure fantasy may have minimal research requirements beyond internal consistency.

Posture Levels¶

Posture	Meaning	Surface Treatment
Corroborated	Multiple reliable sources agree	State directly
Plausible	Reasonable based on evidence	Soft hedge ("believed to be")
Disputed	Sources actively conflict	Present as in-world disagreement
Uncorroborated	No sources found	Neutral phrasing, assess risk

Risk Assessment for Uncorroborated Claims¶

Risk	Impact	Example
Low	Flavor detail	Color of tavern sign
Medium	Affects plot	How organization made decisions
High	Central premise	Medical procedure critical to plot

Research Application¶

Every real-world claim should have assessed posture
High-risk uncorroborated claims need attention before publication
Surface treatment (hedging language) matches posture level
See Historical Fiction for detailed methodology

Validation Approaches¶

Pre-Gate (Quick Check)¶

Fast validation for work-in-progress:

Schema compliance
Required fields present
Link/reference validation
Basic structural integrity

When to use: Before passing work to next stage, during active creation.

Full-Gate (Comprehensive)¶

Thorough validation before commitment:

All bars assessed
Cross-references verified
Human review for non-automatable aspects
Complete quality review

When to use: Before promoting to canon, before publication.

What Can Be Automated¶

Bar	Automation Level
Integrity	High — schema validation, link checking
Reachability	High — graph traversal, randomtest
Comprehension	Low — requires human testing
Style	Low — requires human judgment
Safety	Medium — keyword flagging, human review
Accessibility	Medium — automated checks + manual testing
Canon	Medium — cross-reference, human verification
Spoiler	Low — requires content understanding
Research	Low — requires source evaluation

Actionable Feedback¶

Quality feedback must be actionable. Vague criticism wastes time.

Bad Feedback¶

"The prose needs work."
"Something feels off about this scene."
"Style issues throughout."

Good Feedback¶

"Paragraph 3, sentence 2: 'okay' is anachronistic for 1850s setting. Replace with period-appropriate acknowledgment."
"Choice 2 and Choice 3 are near-synonyms—both involve 'investigating.' Differentiate the action or target."
"Canon conflict: Chapter 2 states the Guild was founded in Year 350, but this passage says Year 347. Verify and align."

Feedback Structure¶

Location: Where exactly is the issue?
Problem: What specifically is wrong?
Standard: Which quality bar is violated?
Fix: What specific change resolves it?

Bar Priority¶

When bars conflict, prioritize:

Safety — Never compromise on harm prevention
Accessibility — Inclusive access is non-negotiable
Integrity — Must be structurally sound
Canon — World consistency matters for coherence
Comprehension — Players must understand
Other bars — Balance based on context

Design Goals vs Quality Bars¶

Some aspects of IF quality are design goals, not validation bars:

Nonlinearity/Choice Meaningfulness:

Whether branches lead to meaningfully different experiences is a design decision. Linear IF and highly branching IF can both be high quality. Evaluating whether "choices matter" depends on authorial intent and genre expectations, not a universal standard.

Emotional Impact:

Whether the story moves readers emotionally cannot be validated—only tested through reader response.

Pacing Effectiveness:

Whether rhythm and flow work requires human judgment about subjective experience.

These aspects require playtesting and reader feedback, not validation bars.

Quick Reference¶

Bar	Focus	Key Question	Automation
Integrity	Structure	Does it hold together?	High
Reachability	Access	Can players get there?	High
Comprehension	Clarity	Do players understand?	Low
Style	Voice	Does it sound unified?	Low
Safety	Harm	Is content responsible?	Medium
Accessibility	Inclusion	Can everyone engage?	Medium
Canon	Truth	Does it match the world?	Medium
Spoiler	Discovery	Are secrets preserved?	Low
Research	Facts	Are claims supported?	Low

Research Basis¶

Key sources informing IF quality standards:

Concept	Source
ChoiceScript quicktest/randomtest	Choice of Games documentation
Content vs Technical QA	BioWare narrative QA practices
Playtest methodology	Emily Short, IF craft essays
Accessibility standards	WCAG 2.1 (W3C)
Comprehension testing	Usability research (Nielsen, Lewis)

ChoiceScript's automated testing tools (quicktest for syntax/structure, randomtest for reachability through random playthroughs) represent industry-standard IF validation. BioWare's distinction between Content QA (narrative consistency, character voice) and Technical QA (scripting, triggers, flags) informed the bar categorization.