The three design-automation flagships were left out of ee78e30 as
electronics rather than computer science. That was the wrong cut: all three
carry system-level and embedded-software tracks, and a publishing guide that
omits a field flagship is worse than one carrying a row the reader skips.
DATE resolves from WikiCFP (201644) and its entry matches the official call
exactly — abstract 13 Sep 2026, paper 20 Sep 2026, Dresden 22-24 Mar 2027 —
so it stays source: wikicfp and maintains itself.
ICCAD is an acronym collision of the kind the README warns about. WikiCFP's
"ICCAD 2026" is the IEEE/IFAC International Conference on Control, Automation
and Diagnosis, in Lisbon in July, and a search-based lookup recorded its dates
against the IEEE/ACM computer-aided design conference without complaint. It is
now wikicfp_id: false with a manual entry from iccad.com: San Jose, 8-12 Nov
2026, paper deadline 14 Apr 2026.
DAC has announced its 2027 edition — San Jose, 11-14 Jul 2027 — but not the
call, so its manual entry carries event fields and no deadline. It renders with
a TBD paper deadline and sorts into the calendar's "Awaiting the next call"
table, which beats omitting the venue and losing the known dates. WikiCFP still
holds only DAC 2026, already run, so the ID is false.
CODES+ISSS remains untracked: ICORE C, and its system-synthesis scope is
covered by the ESWEEK entries already listed. The venues.yaml header comment
now records that, rather than the blanket EDA exclusion it claimed before.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Eight conferences and six journals, curated for the computer-science side of
the field. EDA and circuit-design venues are deliberately excluded — DAC, DATE,
ICCAD and CODES+ISSS are canonical for embedded systems but not for the kind of
work this site tracks — and venues.yaml carries that rationale as a header
comment so the next person doesn't "helpfully" add them back.
Two curation calls worth recording. SenSys, IPSN and IoTDI merged in 2026 into
one conference, the ACM/IEEE International Conference on Embedded Artificial
Intelligence and Sensing Systems; it is tracked under the SenSys acronym, which
is what ICORE and DBLP still use, and IPSN and IoTDI are not listed separately.
RTCSA and ICCPS were dropped — the first is CORE B with no announced 2027
edition, the second carries no CORE rank at all.
Deadlines are all source: manual, verified against the official CFPs. Only EWSN
has a WikiCFP entry for its current edition; every other series there stops at
an edition that has already been held, so the rest are wikicfp_id: false. ECRTS
and LCTES have no announced next edition and sit under missing:, which renders
as TBD. With the stale-deadline handling from 5b06f48 this reads correctly:
only RTAS shows an open deadline, and the rest land in the elapsed-rounds table
rather than advertising passed dates as upcoming.
The topic also exposed three generator bugs that cloud-edge never hit:
_journals_section_body() read scimago_* straight off the venues.yaml entry, so
the journals index showed a dash for every journal without inline values, while
the journal pages resolved the same numbers from rankings/scimago.csv. Both now
share a _journal_ranks() helper, and gen_section_indexes() takes the SCImago
table to feed it. cloud-edge never saw this because all seven of its journals
carry inline values.
The SCImago CSV writes SJR with a decimal comma (0,864). Nothing consumed those
values before, so the loader now normalises them to a dot and CSV-derived
numbers render like the inline ones.
gen_digests() now writes digests/_index.md. Hugo only auto-creates a section
page when the section has a listable page, and a topic with no digests yet
holds nothing but build.list: never stubs, so the Digests nav link 404'd. It
preserves any existing body, leaving cloud-edge's file unchanged.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The calendar's deadline table rendered every recorded submission cycle, so
elapsed rounds sorted to the top and read as the next thing due — 13 of its 17
rows were already in the past. Underneath that, the fetcher preserves
`source: manual` entries forever without ever checking whether they still
describe a future deadline, and 11 of 13 cloud-edge venues are manual.
Preserving them is correct: WikiCFP cannot express multi-cycle venues and does
not carry most systems conferences at all, so re-fetching would replace
researched rounds with worse data or nothing. What was missing is a signal.
- Extract the cycle helpers into cycles.py so the fetcher and the generator
share one definition of which round a reader should act on. cycles_of()
reads an entry without folding flat fields into a cycles list, which would
otherwise rewrite every single-round entry the fetcher writes back.
- The calendar table lists open cycles only; a venue whose every round has
elapsed moves to an "Awaiting the next call" table instead of vanishing.
- pa-fetch-deadlines records those venues under a new stale: key and probes
the year-bumped CFP URL to say which ones already have a next edition
online. A candidate must return 200, mention the target year, and be newer
than any year the entry records — venues.yaml still points SC at sc24.
- README: Task 7 and a matching agent prompt for keeping deadlines current,
since nothing in this repo refreshes them on its own.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
PaperMod ships grey on grey, which read as flat and dim on pages that are
mostly tables. Both schemes are now redefinitions of PaperMod's own
variables, so the theme picks them up without further overrides, plus
--accent, --warm and --wash for the placements below. Every text pair
clears WCAG AA.
Venue tables drop the full 1px grid for horizontal rules, a tinted header
and an accent row hover; domain and tag code spans become chips. Warm
marks the one cell worth finding first in a headered table — the ICORE
rank, the SCImago quartile, the still-open cycle — and detail-page fact
tables are excluded, since bold there is a row label, not emphasis.
The FullCalendar variable mapping was dark-only, leaving light mode on
the library's stock chrome; it now maps once through theme variables and
covers both. Event colours were saturated with white text on top, so they
become five pale chips carrying explicit dark ink, legible on either
theme.
Ranks were also emitting '**A***', which closes the emphasis on the
second star and leaves the third loose — visible once ranks had their own
colour.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Two small presentation fixes.
The calendar page dropped its "Journal Special Issues" heading entirely when
no entry existed, which is the state since f132f6b removed the two unverified
CFPs. With only the conference table left, the page read as if journals
weren't tracked at all rather than as having no open call. The heading now
always renders, carrying a line that says so and why: these journals use
rolling submission, so special issue CFPs are their only dated deadlines.
In the digests list the intro paragraph sat flush against the first card.
PaperMod gives .post-entry a bottom margin only, and .md-content zeroes its
last child's, so nothing supplied the gap — the theme never hits this because
its own .md-content usage sits inside a .post-entry. Added a page-intro class
with a bottom margin matching the card-to-card gap, in the extended stylesheet
rather than the submodule.
Hugo isn't installed here, so the digests spacing is reasoned from the theme
CSS rather than rendered.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The two special_issues entries were proof-of-concept data and could not be
corroborated: no TPDS "Edge AI Systems" or open FGCS "Serverless Computing
at the Edge" call is findable, and both named real researchers as guest
editors. Removed rather than corrected — nothing verifiable exists to
replace them with. No open special issue was found at any tracked journal,
so the field stays absent until a real CFP appears.
Audited the rest of the data while there. All 39 candidate pools come from
DBLP and all 471 digested papers reconcile against them, so the paper set
is sound. Three defects surfaced:
- DBLP returns HTML-escaped text and fetch_papers never decoded it, so 136
entities sat in the cached pools and "Theodore Y. Ts'o" reached the
published ATC-2024 digest. Decoded at the fetch boundary and cleaned the
existing files through the YAML parser.
- SC event_dates recorded the exhibition window (Nov 17-19) instead of the
conference (Nov 15-20); EuroSys ran a day long (Apr 19-24 vs 19-23).
Both verified against the official sites, which also confirm every
SC, EuroSys and OSDI submission deadline is exact.
- TC-2025 digest carried a paraphrased title, now matching DBLP.
EuroSys only records its spring cycle; the fall cycle has no home in the
one-deadline-per-venue schema.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Hugo/PaperMod static site tracking 13 conferences and 7 journals for
edge and cloud systems research. Includes FullCalendar deadline view,
ICORE/SCImago rankings, DBLP paper digest pipeline, and Python
fetch/generate scripts. PaperMod added as a git submodule.
