L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Développement · SKILL.md

socc-supplementary

socc-supplementary

Use when deciding what belongs in an ACM SoCC paper body versus its released artifact and appendices, covering the acmart 12-page (full) or 6-page (short) budget with unlimited references, the rule that decision-critical evidence stays inside the reviewed pages, dual-anonymous supplementary material, and how to split a cloud-systems measurement paper betwee…

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentation de la Skill

GitHub

name: socc-supplementary description: Use when deciding what belongs in an ACM SoCC paper body versus its released artifact and appendices, covering the acmart 12-page (full) or 6-page (short) budget with unlimited references, the rule that decision-critical evidence stays inside the reviewed pages, dual-anonymous supplementary material, and how to split a cloud-systems measurement paper between body and package.

SoCC Supplementary

Use this when assembling SoCC supplementary material. The governing rule is simple and strict: the paper must be judgeable from the reviewed pages alone. The released artifact and any appendix support the paper; they do not hold the argument. Reviewers read the artifact at their discretion, so anything the decision depends on — the headline throughput, tail, and cost results — lives in the body.

What goes where

ContentBody (within page budget)Artifact / appendix
The cloud problem and the contributionYes—
Core mechanism or measurement designYesFull config grids, extra parameters
Headline throughput / tail / cost resultsYesFull result tables, secondary regimes
The key baseline comparisonThe result + baseline named and tunedThe full sweep and raw logs
Testbed / trace descriptionSummary + key parametersThe complete node list, full trace spec
Raw logs, workload generators, replay scripts—Yes
Reproduction instructionsA pointerThe README and run scripts

If a reviewer would need to open the artifact to know whether to accept — for instance to find the p99 or the cost breakdown — the paper is mis-partitioned; move that evidence into the body.

The page-budget discipline

SoCC's acmart budget (full: 12 pages of text and figures; short: 6 pages), with unlimited references, counts figures, tables, and any in-paper appendix. Consequences:

  • Keep the tail and cost results and the primary baseline comparison in the body — they are decision-critical at a cloud venue.
  • Long configuration sweeps, per-node breakdowns, and additional workload regimes go to the artifact with explicit forward references.
  • The unlimited-reference allowance is for the bibliography only; it does not let body text or decision-critical figures escape the 12/6-page limit.
  • Do not use the artifact to smuggle in a result that would not fit — an argument that only closes with material outside the reviewed pages reads as unreviewable.

Dual-anonymous supplementary rules

[No identity]   no authors, affiliations, acknowledgements, grants, cluster names, provider hints,
                or repository owners anywhere in the archive or appendix
[Anonymized links] the availability link points at an anonymizing host, not a personal/corporate repo
[Scrubbed traces]  released traces carry no identifying provenance (org name, internal IDs)
[Clean archive]  no .git history, .DS_Store, credentials, caches, or large irrelevant files
[Opens clean]    verify the archive unpacks and the README orients a reader in one minute on a
                fresh machine

Appendix architecture (when the template allows in-paper appendices)

  • Order appendix sections to mirror the result order; reviewers navigate by claim, not page number.
  • Keep each appendix section referenced from the body at least once — orphaned material is invisible under discretionary review.
  • Confirm on the current call whether in-paper appendices count against the 12/6-page budget; this varies and changes the split.

Vignette: splitting a cloud-systems measurement paper

A paper measuring a scheduler on a replayed trace: the body keeps the problem, the mechanism, the headline throughput/p99/cost results with tuned baselines, the key trade-off figure, and a limitations paragraph; the artifact holds the replay harness, the (anonymized) trace, the full config sweep, the per-node logs, and the analysis scripts. Nothing decision-critical lives only in the artifact, because artifact inspection is discretionary and the reviewers judge from the pages.

Output format

[Supplement status] ready / needs fixes / not ready
[Partition check] anything decision-critical (tail/cost/baseline) outside the body? <none / move: what>
[Page budget] body within 12/6 pages? references-only unlimited allowance respected? yes/no
[Anonymity] archive + traces clean of identity + metadata? passed/issues
[Body dependency] <what a reviewer can decide without opening the artifact>
Skills enregistrées