brycewang-stanford IMC Workflow
Use this to plan an IMC campaign against the calendar. IMC runs two submission deadlines a year with a shared review pipeline and a One-Shot-Revision path that resubmits to the next deadline. Planning backward from the right deadline — and reserving time for ethics approvals and dataset release that cannot be rushed — is the difference between a smooth cycle and a missed one.
Choose the deadline first
IMC 2026 offered two cycles: cycle 1 submission ~20 Nov 2025, cycle 2 submission ~29 Apr 2026 (confirm current dates on the Important Dates page). Choose by data-collection readiness, not optimism:
- If measurement and ethics approvals will be solid by the earlier cutoff, target cycle 1 — and a One-Shot-Revision still lands you in cycle 2.
- If data collection or IRB is unfinished, target cycle 2 rather than submitting thin work that earns a reject instead of a revision.
Backward plan from the target cutoff
[T-0] Submission deadline (AoE). PDF + BBL uploaded; availability declaration set.
[T-1wk] Registration cutoff (title/abstract/authors/conflicts) — precedes submission by ~a week.
[T-2wk] Final anonymity sweep (incl. vantage points); Ethics section finalized; artifact anonymized.
[T-4wk] Freeze the measurement narrative; draft complete; limitations argued per-result.
[T-8wk] Finish the decision-critical measurements; validate ground-truth subsamples.
[T-earlier] Obtain IRB determination; run active measurements safely; pin provenance as you collect.
Ethics approvals and longitudinal windows are the long poles — they cannot be compressed in the final weeks.
The review-to-decision arc
[Submit] -> multi-round double-blind review
[Early reject] consensus-reject papers notified early -> reroute promptly (imc-topic-selection)
[Decision] Accept / One-Shot-Revision / Reject
Accept -> shepherding + camera-ready
One-Shot-Revision -> execute the bounded action points; resubmit NEXT deadline, same reviewers
Reject -> reroute to SIGCOMM/NSDI/CoNEXT/PAM or a journal
The One-Shot-Revision leg
If you receive a One-Shot-Revision (imc-review-process, imc-author-response):
- You have until the next deadline to execute a bounded action list (up to a few concrete additions/changes) for the same reviewers.
- Plan the additional measurements immediately — a new vantage point or a longitudinal extension takes real calendar time.
- Do exactly the named action points; resist scope creep that could cost the champion's support.
After acceptance
[Shepherding] deliver the artifact you declared; the shepherd checks availability
[Camera-ready] de-anonymize; finalize Ethics + availability statements; permanentize dataset/code
links to a DOI archive (cycle 1 camera-ready ~mid-April; cycle 2 ~1 Sep — confirm)
[Award] for the Community Contribution Award, the dataset/tool must be public + usable by
the camera-ready deadline
[Present] at least one author presents at IMC (Karlsruhe, Oct 12-16, 2026)
The Replicability Track calendar (if applicable)
If the contribution is replicating a prior measurement study, the path differs: submit an Expression of Interest by its (earlier) deadline (IMC 2026 EoI ~29 Jan 2026), a small committee screens it, and strong abstracts are invited to a full submission judged like the main track. Plan the EoI well before you complete the replication.
Calibration
- Every date is a per-edition snapshot — confirm both cycles' deadlines, the registration gap, and the camera-ready dates on the current Important Dates page before committing.
- Do not assume the One-Shot-Revision mechanics, the number of action points, or the Replicability Track dates carry over between editions.
Output format
[Target] cycle 1 / cycle 2, with the reason (data + ethics readiness)
[Backward plan] <dates from data collection -> registration -> submission>
[Long poles] <IRB / longitudinal window / dataset release>
[Post-decision path] accept / one-shot-revision (next deadline) / reroute
[Release plan] shepherding + DOI archive by camera-ready; award eligibility