newms-writing-style
GitHub用于撰写或润色《New Media & Society》期刊稿件,面向跨学科读者。规范约8000字篇幅、SAGE哈佛引用及摘要格式。通过概念优先、解释术语、避免行话和中心主义假设,提升文章清晰度与可读性,确保贡献前置。
触发场景
安装
npx skills add brycewang-stanford/Awesome-Journal-Skills --skill newms-writing-style -g -y
SKILL.md
Frontmatter
{
"name": "newms-writing-style",
"description": "Use when drafting or polishing the prose of a New Media & Society (NM&S) manuscript for an interdisciplinary readership within the ~8,000-word target, using SAGE Harvard referencing. Shapes the writing; it does not generate findings or design the study."
}
Writing Style (newms-writing-style)
NM&S prose must reach an interdisciplinary audience — sociologists, media and communication scholars, STS researchers, political scientists, and humanists read the same issue — inside a strict ~8,000-word target (all text including notes, references, tables, and charts). Write so a reader two fields away follows the argument, glosses your jargon, and reaches the contribution early.
When to trigger
- Drafting the introduction or revising for clarity, voice, and length
- A reviewer said the writing is jargon-heavy, technical-only, or buries the contribution
- You are over the word target and need to cut without losing the argument
The NM&S introduction arc
By the end of the introduction, a reader should have:
- The question, stated plainly for a non-specialist.
- The argument — a named concept or mechanism that travels (
newms-theory-building). - The evidence — what method adjudicates the leading rival reading (
newms-research-design). - Why it matters to new media and society broadly — the interdisciplinary stakes, with scope.
Front-load the contribution; do not make the reader reach the discussion to learn what you claim.
House-style essentials
- SAGE Harvard referencing (author-date in text, alphabetical reference list). Format consistently; reference managers help but verify the output. Confirm the current style on the live guidelines.
- Unstructured abstract ~150 words stating purpose, approach, and findings; ≥4 keywords.
- Gloss travelling concepts. Define "affordance," "datafication," "context collapse," "platformization" the first time — a cross-field reader should not have to know the lineage to follow you.
- Plain, precise voice. Avoid both engineering jargon and untethered theory-speak; name the platform but lead with the concept.
- International readership. Don't assume US/Anglophone defaults; gloss country-specific institutions.
Writing for the interdisciplinary reader
| Trap | Fix |
|---|---|
| Subfield jargon undefined | name it, then gloss it in a clause |
| Technical method dump | lead with what it shows, detail the method second |
| Theory-speak with no referent | tie every abstract claim to a concrete digital-media practice |
| US-centric assumptions | gloss institutions; engage non-Anglophone work |
| Contribution buried in discussion | move it into the first page |
Concept-first prose map
Every major paragraph should move from the portable idea to the platform case, not the other way around. Use this map when revising an introduction, theory section, or discussion.
| Sentence job | Strong NM&S version | Weak version |
|---|---|---|
| Opening problem | names a social/media problem a cross-field reader recognizes | starts with one platform feature |
| Concept sentence | defines the travelling concept in ordinary language | drops a keyword with no gloss |
| Site sentence | explains why this platform/community/data source reveals the concept | treats the site as self-evidently important |
| Evidence sentence | states what the method can adjudicate and what it cannot | lists data without the warrant |
| Stakes sentence | says what changes for media/social theory, policy, or practice | says the topic is "timely" |
If a paragraph begins with platform detail, ask whether the concept can come first. If the answer is no, the paragraph may be descriptive background rather than argument.
Cross-field readability pass
Run a "two fields away" edit before submission.
- Replace lineage-only phrases with a gloss: do not rely on readers already knowing a debate.
- Translate platform mechanics into social processes: "recommendation rank" becomes visibility, gatekeeping, labor control, attention allocation, or public formation where appropriate.
- Keep methods visible but subordinate to the claim: lead with what the design shows, then name the interview, ethnography, content analysis, computational, or mixed-method procedure.
- Flag local institutions, regulations, and platform cultures that need one sentence of context for international readers.
- Cut citation strings that only signal belonging to a subfield; keep the citations that carry the conceptual contrast or empirical warrant.
Managing the word target
The ~8,000-word limit is firm — submissions more than a few hundred words over are not considered. Tactics: cut over-signposted roadmaps; turn scattered quotes into one quote-to-claim table; trim literature that doesn't bear on the gap; move ancillary material to (clearly limited) supplements; remember tables and references count toward the total.
Word-budget triage
When the draft is over target, cut by function rather than by sentence-level trimming.
| Section pressure | Keep | Cut or move |
|---|---|---|
| Introduction too long | question, concept, evidence, broad stakes | extended roadmap, platform history not needed for the claim |
| Literature review too long | debates that create the conceptual gap | adjacent literatures not used in the argument |
| Methods too long | sampling/data logic and validity warrant | software minutiae better suited to supplement |
| Findings too long | exhibit or quote that advances the mechanism | repeated descriptive examples |
| Discussion too long | theory implication, scope condition, limitation | generic future-research boilerplate |
| References too long | sources actually used to build the contrast | ornamental citation clusters |
After each cut, check whether the intro arc still contains question, argument, evidence, and broad stakes. A shorter paper that loses the portable concept is not a better NM&S paper.
Claim-calibration edit
NM&S prose should not overclaim from platform-specific evidence. Calibrate verbs to evidence strength.
| Evidence base | Safe verb | Risky verb |
|---|---|---|
| qualitative or ethnographic pattern | shows, illustrates, elaborates, reveals | proves, measures universally |
| content/discourse analysis | maps, identifies, traces | demonstrates individual behavior |
| computational trace data | detects, estimates, classifies, models | captures social reality directly |
| mixed methods | triangulates, integrates, explains | settles all mechanisms |
| single platform/country | suggests, specifies, bounds | generalizes globally |
Worked micro-example (illustrative)
Before (buries it): "This paper examines couriers' experiences with platform algorithms using interviews."
After (front-loads): "Why do gig couriers discipline themselves to an algorithm they cannot see? We
argue opaque ranking produces *datafied control* — anticipatory self-discipline — and show it through
digital ethnography that separates control from mere gaming of the system."
Calibration anchors
- Two fields away is the test. If a reader outside your subfield can't follow the intro, revise.
- The ~8,000 target is all-in. Notes, references, and tables count; write lean from the start.
- Lead with the concept, name the platform second. The idea is the contribution; the app is the site.
Anti-patterns
- An introduction that reaches the contribution only in the discussion
- Undefined travelling concepts assumed known to all readers
- Engineering or theory jargon with no concrete referent
- Over-signposted "Section 2 reviews… Section 3 describes…" roadmaps eating the word budget
- Treating the ~8,000-word target as excluding references/tables (it does not)
Output format
【Intro arc】question / argument / evidence / broad stakes all present? [Y/N each]
【Jargon glossed】travelling concepts defined on first use? [Y/N]
【Referencing】SAGE Harvard, consistent? [Y/N]
【Length】within ~8,000 words incl. notes/refs/tables? [Y/N — current count]
【Cross-field pass】concept first / platform second / local context glossed? [Y/N]
【Cut plan】sections to cut or move before submission
【Claim calibration】verbs match evidence strength? [Y/N]
【Next】newms-transparency-and-data
Supplementary resources
../../resources/worked-examples/01-introduction.md— before→after NM&S introduction../../resources/official-source-map.md— word target, abstract, referencing facts
版本历史
- 1839142 当前 2026-07-05 14:07


