POPL Submission
Audit against the cycle you are actually entering. The mechanics below come from the
POPL 2027 research-papers call, read 2026-07-08 through search renderings of
popl27.sigplan.org (direct fetches were blocked — see
resources/official-source-map.md). POPL 2027's cutoff was July 9, 2026, 11:59 PM
AoE, one day after that check: confirm the live dates page before assuming any window
is open.
The 2027-cycle mechanics
- One July deadline, no abstract deadline. POPL 2027 papers were due July 9,
2026 AoE, uploaded to
popl27.hotcrp.com. Nothing pre-registers the submission,
so a paper can materialize on deadline day — and so can your forgotten conflicts
list. A missed July means the next POPL is a year out (though PACMPL siblings
offer intermediate deadlines; see popl-workflow).
- 25 pages of text, bibliography excluded, in the PACMPL layout: acmart with the
acmsmall option, single column, 10 pt font on 12 pt line spacing, a 5.478 in by
7.884 in text block. The call names summary rejection as the consequence of a
shrunken font or widened block, and PACMPL takes no Word files.
- Full double-blind. Since POPL 2023 this replaced the older lightweight regime:
reviewers never see author names, and identities are revealed only after
conditional-acceptance decisions, only for the papers that made it.
Where theory papers deanonymize themselves
A semantics or verification paper leaks identity through different channels than an
ML paper: a Rocq/Coq development already public under your group's namespace, a
paper-named logic that only one lab builds on, "in our previous work [7]" phrasing,
and appendix PDFs generated with author metadata intact.
pdfinfo main.pdf | grep -iE 'author|producer'
pdftotext main.pdf - | grep -niE 'our (previous|earlier) (work|paper)|github|gitlab|zenodo|orcid' | head
tar tzf supplement.tar.gz | grep -E '\.git/|home/[a-z]+/|_CoqProject.*~' | head
Cite your own prior papers in the third person, scrub \author blocks and grant
lines, and route any proof development through an anonymized mirror rather than the
lab repository.
Severity table
| Failure |
Likely outcome under the 2027 call |
Can it be repaired after the deadline? |
| Text beyond 25 pages or altered format |
Summary rejection |
No |
| Word manuscript |
Not accepted by PACMPL tooling |
No |
| Author-identifying PDF or supplement |
Chair action against double-blind policy |
No |
| Same result under review elsewhere |
Dual-submission violation |
No |
| Main theorem stated without its assumptions |
Survives triage, dies in review |
Only before upload |
| Missing HotCRP conflicts/topics |
Assignment errors, wasted expert reviews |
Sometimes, at chair discretion |
Last-72-hours order of operations
- Freeze definitions and theorem numbering; late renumbering desyncs body, appendix,
and proof scripts.
- Rebuild the PDF from a clean clone so no stale lemma statement survives.
- Run the leak scan above on the paper and every supplementary archive.
- Fill HotCRP metadata — title, abstract, topics, conflicts — a day early; the form
is fast, but conflict lists take thought.
- Upload well before 11:59 PM AoE and re-download your own submission to verify the
PDF the committee will see.
Output format
[POPL submission audit] pass / blockers found
[Format] <25-page text count, acmsmall compliance, PDF build>
[Anonymity] <leak-scan findings across paper, appendix, proof archive>
[Policy] <dual submission, prior publication, conflicts>
[Remaining actions] <ordered, with owner and cutoff time in AoE>
Source: brycewang-stanford/Awesome-Journal-Skills → POPL-Skills/skills/popl-submission/SKILL.md
1---2name: popl-submission3description: Use when preparing a POPL submission for the single July HotCRP deadline — auditing the 25-pages-of-text acmsmall cap, summary-rejection format rules, full double-blind hygiene for theory papers with public proof repositories, dual-submission exposure, and the final-days upload order before the AoE cutoff.4---5
6
7# POPL Submission
8
9Audit against the cycle you are actually entering. The mechanics below come from the
10POPL 2027 research-papers call, read 2026-07-08 through search renderings of
11`popl27.sigplan.org` (direct fetches were blocked — see
12`resources/official-source-map.md`). POPL 2027's cutoff was July 9, 2026, 11:59 PM
13AoE, one day after that check: confirm the live dates page before assuming any window
14is open.
15
16## The 2027-cycle mechanics
17
18- **One July deadline, no abstract deadline.** POPL 2027 papers were due July 9,
19 2026 AoE, uploaded to `popl27.hotcrp.com`. Nothing pre-registers the submission,
20 so a paper can materialize on deadline day — and so can your forgotten conflicts
21 list. A missed July means the next POPL is a year out (though PACMPL siblings
22 offer intermediate deadlines; see `popl-workflow`).
23- **25 pages of text, bibliography excluded**, in the PACMPL layout: acmart with the
24 `acmsmall` option, single column, 10 pt font on 12 pt line spacing, a 5.478 in by
25 7.884 in text block. The call names **summary rejection** as the consequence of a
26 shrunken font or widened block, and PACMPL takes no Word files.
27- **Full double-blind.** Since POPL 2023 this replaced the older lightweight regime:
28 reviewers never see author names, and identities are revealed only after
29 conditional-acceptance decisions, only for the papers that made it.
30
31## Where theory papers deanonymize themselves
32
33A semantics or verification paper leaks identity through different channels than an
34ML paper: a Rocq/Coq development already public under your group's namespace, a
35paper-named logic that only one lab builds on, "in our previous work [7]" phrasing,
36and appendix PDFs generated with author metadata intact.
37
38```bash
39pdfinfo main.pdf | grep -iE 'author|producer'
40pdftotext main.pdf - | grep -niE 'our (previous|earlier) (work|paper)|github|gitlab|zenodo|orcid' | head
41tar tzf supplement.tar.gz | grep -E '\.git/|home/[a-z]+/|_CoqProject.*~' | head
42```
43
44Cite your own prior papers in the third person, scrub `\author` blocks and grant
45lines, and route any proof development through an anonymized mirror rather than the
46lab repository.
47
48## Severity table
49
50| Failure | Likely outcome under the 2027 call | Can it be repaired after the deadline? |
51|---|---|---|
52| Text beyond 25 pages or altered format | Summary rejection | No |
53| Word manuscript | Not accepted by PACMPL tooling | No |
54| Author-identifying PDF or supplement | Chair action against double-blind policy | No |
55| Same result under review elsewhere | Dual-submission violation | No |
56| Main theorem stated without its assumptions | Survives triage, dies in review | Only before upload |
57| Missing HotCRP conflicts/topics | Assignment errors, wasted expert reviews | Sometimes, at chair discretion |
58
59## Last-72-hours order of operations
60
611. Freeze definitions and theorem numbering; late renumbering desyncs body, appendix,
62 and proof scripts.
632. Rebuild the PDF from a clean clone so no stale lemma statement survives.
643. Run the leak scan above on the paper and every supplementary archive.
654. Fill HotCRP metadata — title, abstract, topics, conflicts — a day early; the form
66 is fast, but conflict lists take thought.
675. Upload well before 11:59 PM AoE and re-download your own submission to verify the
68 PDF the committee will see.
69
70## Output format
71
72```text
73[POPL submission audit] pass / blockers found
74[Format] <25-page text count, acmsmall compliance, PDF build>
75[Anonymity] <leak-scan findings across paper, appendix, proof archive>
76[Policy] <dual submission, prior publication, conflicts>
77[Remaining actions] <ordered, with owner and cutoff time in AoE>
78```
79
80---
81
82**Source:** [`brycewang-stanford/Awesome-Journal-Skills`](https://github.com/brycewang-stanford/Awesome-Journal-Skills) → `POPL-Skills/skills/popl-submission/SKILL.md`
thedixitjain