Apache Cassandra Accord · Project Health

Apache Cassandra Accord

Automated, reproducible metrics for Apache Cassandra Accord.

asf_roster: not_applicable git: ok github: ok github_commit_authors: ok github_issues: not_applicable github_profile: ok governance: not_applicable jira: ok ponymail: not_applicable release: not_applicable release_lines: not_applicable security: ok subproject_releases: ok

Code and contributor health metrics: sustainability, review capacity, and responsiveness — computed from Git, ASF JIRA and GitHub PR history.

Charts default to the last 3 years.

Faint points mark a month flagged low-n or insufficient data — its statistic (a rate, an HHI, a median) comes from too few underlying events to be reliable on its own. Hover a point for its exact n and flag; the value itself is unchanged.

Responsiveness CHAOSS practitioner guide
  • Time to First Response (JIRA) 1.1 days Aug 2026 · n=5
  • Stale PR Rate 81.0% Oct 2026 · n=21
  • Open PR backlog 21 Sep 2026 · n=21
  • Median JIRA Resolution Latency 2.6 days Aug 2026 · n=6

Responsiveness

Median JIRA Resolution Latency (12m cohort)

established

2.6 days Aug 2026

Based on: CHAOSS: Issue Resolution Duration

JSON · CSV

Change Request Closure Ratio — GitHub PR

proxy

0.833 Aug 2026

Based on: CHAOSS: Change Request Closure Ratio

JSON · CSV

Change Request Closure Ratio — JIRA Patch Submissions

established

0.500 Aug 2026

Based on: CHAOSS: Change Request Closure Ratio

JSON · CSV

Issue Closure Ratio — GitHub Issues

proxy

n/a

insufficient data

Based on: CHAOSS: Issues Closed (adapted)

JSON · CSV

Reviewer Capacity

Review responsiveness

How quickly a patch submission gets its first visible review response, split by how many prior submissions the same contributor has -- a first-time submitter, a 2nd-5th submission, or a 6th+ submission. A "submission" is one JIRA issue that ever reached Patch Available, together with any linked GitHub pull request. "First visible response" is the earliest of a JIRA comment, a JIRA status change to Review In Progress/Ready to Commit, or a GitHub PR review/comment, from someone other than the submission's own author -- review that happened somewhere this project doesn't track (a private message, an in-person conversation) is invisible to this section, so "no visible response" means exactly that, not "no response at all."

Methodology and caveats. The submission's author is reconstructed from JIRA's assignee-change history, which is a best-effort proxy, not a confirmed identity. Shares for "response within 7/30 days" and "no visible response" only count submissions old enough to judge (right-censoring: a submission newer than 7, 30, or 365 days is left out of that share's count entirely, not counted as a miss). These figures are informational only, shown with their own sample size (n) for context. Full definitions: METRICS.md.

First-time submitters

Year Submissions Median days to first response ≤7d ≤30d No visible response Committed ≤365d
2022 4 insufficient data insufficient data insufficient data insufficient data insufficient data
2023 6 insufficient data 50% 50% 33% 67%
2024 1 insufficient data insufficient data insufficient data insufficient data insufficient data
2025 4 insufficient data insufficient data insufficient data insufficient data insufficient data

2nd-5th submission

Year Submissions Median days to first response ≤7d ≤30d No visible response Committed ≤365d
2022 8 14.3 50% 62% 0% 75%
2023 13 0.8 69% 77% 23% 85%
2024 10 insufficient data 40% 40% 60% 80%
2025 2 insufficient data insufficient data insufficient data insufficient data insufficient data

6th+ submission

Year Submissions Median days to first response ≤7d ≤30d No visible response Committed ≤365d
2022 0 insufficient data insufficient data insufficient data insufficient data insufficient data
2023 28 3.7 50% 68% 25% 93%
2024 50 2.8 34% 40% 58% 98%
2025 58 0.1 29% 29% 67% 96%

Trailing 12-month response-within-30-days share, by tier

Source of first response (2025 and later tiers combined)

Source Count Share
JIRA comment 10 40%
JIRA status change 9 36%
GitHub review 6 24%

Open PR backlog

The open apache/cassandra PR backlog by age and by linked-JIRA-ticket state, reconstructed monthly. Scoped to apache/cassandra: other configured repos use their own issue tracker or none, so ticket-key state isn't meaningful for them — see the table below instead.

By age — apache/cassandra (last 36 months)

By linked-ticket state — apache/cassandra (last 36 months)

Current month — apache/cassandra

Month Open PRs Drafts No GitHub response
2026-09-30 21 1 57% (n=21)

Based on: CHAOSS: Change Requests (adapted), CHAOSS: Issue Age (adapted, applied to change requests)

Base branch (trunk vs. release branches) is not collected yet, so it is not shown here.

Contributor sustainability CHAOSS practitioner guide
  • Active Contributors 1 Sep 2026 · n=1
  • New Contributors 0 Sep 2026 · n=0
  • Contributor Dependency (N for 50%) 1 Sep 2026 · n=6
  • Unique Reviewers 1 Sep 2026 · n=1

Contributor Sustainability

Active Contributors

established

1 Sep 2026

Based on: CHAOSS: Contributors

JSON · CSV

New Contributors

established

0 Sep 2026

Based on: CHAOSS: New Contributors

JSON · CSV

First → Second Contribution Conversion

proxy

insufficient data

Based on: CHAOSS: Conversion Rate

JSON · CSV

Contributor Churn Rate

proxy

57.1% Aug 2025 backfill pending

Based on: project-specific (no established CHAOSS/literature equivalent)

JSON · CSV

Sustained Contributors

established

1 Sep 2026

Based on: CHAOSS: Contributors (adjacent)

JSON · CSV

Contributor Dependency (N for 50%)

established

1 Sep 2026

Based on: CHAOSS: Contributor Absence Factor

JSON · CSV

Contributor Concentration (HHI)

established

0.473 Sep 2026

Based on: DOJ/FTC: Herfindahl-Hirschman Index

JSON · CSV

Reviewer Capacity

Reviewer Concentration (HHI)

proxy

0.190 Sep 2026

Based on: project-specific (no established CHAOSS/literature equivalent)

JSON · CSV

Contributor pipeline

Two additional views of the cards above: a Kaplan–Meier survival curve over contributor tenure, and a per-stage breakdown of the latest completed first-time-contributor cohort's funnel. Both are computed from git commit history only; the committer and PMC funnel stages are not available (no confirmed identity link between the ASF roster and this project's git/JIRA identities exists yet).

Tenure survival

Cohort size (n=3) is below the 30-contributor floor for a Kaplan–Meier curve; showing a still-active count instead.

Months since first contribution Eligible Still active Share
3 2 2 100.0%
6 1 1 100.0%
12 1 1 100.0%
18 0 0 insufficient data
24 0 0 insufficient data

The contributor funnel isn't published yet.

Contributor leaderboard

Top contributors by one activity type at a time — commits, reviews, and JIRA issues resolved — over the trailing 12 completed months (2025-10-01 to 2026-09-30). These are ranked counts of deterministic, public activity, never a blended "contribution score" and never based on any classified label (DECISIONS.md — D19).

Identity-resolution limits. Counts are grouped by resolved identity (normalize/identity.py), not by human being. M0's identity resolution only merges an exact-matching git email or JIRA username; a commit-trailer reviewer name (a bare handle or display name, not an email) is never automatically linked to that same person's git email or JIRA username, so one contributor active under more than one raw identifier can appear more than once, or under a handle rather than their full name. Manual merges are reviewed, evidenced entries in identity_overrides.yaml -- see the corrections link below to request one. Request an identity correction (every row's own "Report an issue" link below uses this same form) — see /corrections/ for what happens next.

JSON · CSV

Organizational participation CHAOSS practitioner guide
  • Elephant Factor insufficient data
  • Organizational Concentration (HHI) insufficient data
  • Single-Organization Share insufficient data
  • Unknown-Affiliation Rate 27.3% Sep 2026 · n=33

Organizational Diversity

Organizational Concentration (HHI)

established

insufficient data

Based on: CHAOSS: Organizational Diversity

JSON · CSV

Single-Organization Share

established

insufficient data

Based on: CHAOSS: Organizational Diversity

JSON · CSV

Unknown-Affiliation Rate

established

27.3% Sep 2026

Based on: project-specific (no established CHAOSS/literature equivalent)

JSON · CSV

Leadership CHAOSS practitioner guide
  • PMC Joins per Quarter shared — see Apache Cassandra

PMC membership is tracked once for the whole Apache Cassandra project via the shared ASF roster, not per subproject. See Apache Cassandra's Community page →

Releases CHAOSS: Release Frequency
  • Release Frequency insufficient data
  • Days Between Releases (median) insufficient data
  • Time Since Last Release insufficient data

Release Cadence

Release Frequency

established

n/a

insufficient data

Based on: CHAOSS: Release Frequency

JSON · CSV

Release Interval Regularity (CoV)

proxy

n/a

insufficient data

Based on: project-specific (no established CHAOSS/literature equivalent)

JSON · CSV

Time Since Last Release

established

n/a

insufficient data

Based on: project-specific (no established CHAOSS/literature equivalent)

JSON · CSV

Days Between Releases (median)

proxy

n/a

insufficient data

Based on: CHAOSS: Release Frequency

JSON · CSV

Release lines, CI and blockers → /releases/

Conversation patterns

Conversation patterns are classified from the dev@ mailing list, shared across all Apache Cassandra subprojects — not computed per subproject. See Apache Cassandra's Conversations page →