Release v0.11.2: unified plain community data

Merge Stack Exchange, GitHub, and GitLab records into a plain thread/post dataset with row-level source and license provenance.

README.md

@@ -1,52 +1,70 @@
 ---
-configs:
-  - config_name: threads
-    data_files:
-      - split: train
-        path: data/threads/train.parquet
-  - config_name: posts
-    data_files:
-      - split: train
-        path: data/posts/train.parquet
-tags:
-  - community-qa
-  - software-engineering
-  - github
-  - gitlab
-  - ai
+pretty_name: Community Data
+license: other
 ---
 
-# Community QA Lab — Plain Thread and Post Dataset (v0.11.1)
+# Community Data
 
-This release contains **425,362 source threads** and **1,464,177 source posts** from **704 GitHub and GitLab projects** (698 GitHub repositories and 6 GitLab projects). It is a downloadable dataset of public software-project conversations, issues, discussions, pull requests, and replies collected through 2026-09-30.
+An ordinary, source-attributed collection of public community threads and posts from Stack Exchange, GitHub, and GitLab. The tables contain the original thread and post text with source IDs, links, timestamps, authors where available, and row-level license information.
 
-## Download
+Release: **v0.11.2**  
+Rows: **489,208 threads** and **1,671,299 posts**.
 
-The package has two configurations. Each has a single `train` split containing the complete table; it is not a train/evaluation benchmark split.
+## Included sources
+
+- `github`: 469,944 threads; 1,620,060 posts
+- `gitlab`: 1,743 threads; 1,489 posts
+- `stackexchange`: 17,521 threads; 49,750 posts
+
+The source mix spans Stack Exchange questions and answers across 48 communities, GitHub issues, pull requests and discussions, and GitLab issues and merge requests. The `repository` column contains a repository path for GitHub/GitLab and the community hostname for Stack Exchange.
+
+## Files
+
+- `data/threads/train.parquet`: one canonical row per source thread.
+- `data/posts/train.parquet`: one row per question, answer, issue body, comment, or discussion post retained from a source thread.
+- `manifest.json`: exact row counts, file hashes, source counts, input snapshots, deduplication rules, and publication exclusions.
+
+This is a plain text dataset. It contains no retrieval chunks, embeddings, or task-specific ranking/preference labels.
+
+## Load the tables
 
 ```python
-from datasets import load_dataset
+from huggingface_hub import hf_hub_download
+import pyarrow.parquet as pq
+
+threads_path = hf_hub_download(
+    repo_id="aluha501/community-data",
+    repo_type="dataset",
+    filename="data/threads/train.parquet",
+)
+threads = pq.read_table(threads_path).to_pandas()
 
-repo_id = "aluha501/community-qa-lab"
-threads = load_dataset(repo_id, "threads")["train"]
-posts = load_dataset(repo_id, "posts")["train"]
+posts_path = hf_hub_download(
+    repo_id="aluha501/community-data",
+    repo_type="dataset",
+    filename="data/posts/train.parquet",
+)
+posts = pq.read_table(posts_path).to_pandas()
 ```
 
-You can also download the Parquet files directly from the Files and versions page.
+## Schema
+
+### Threads
+
+`thread_uid` is the stable dataset-wide ID. `source_thread_uid` and `thread_id` retain the source ID; `source_platform`, `source_instance`, `source_kind`, and `repository` identify where it came from. `title`, `body`, and `tags` contain the source text and tags. `source_url`, `author_name`, `author_url`, `created_at`, and `updated_at` retain source metadata. `license` and `license_evidence_url` describe the applicable source contribution license. `provenance_status`, `content_hash`, `source_snapshot_at`, and `source_version` record provenance and snapshot details.
 
-## Tables
+### Posts
 
-- **`threads`**: one row per source conversation root. Includes source IDs, platform and repository, title and body, timestamps, tags, URL, author attribution, and license information.
-- **`posts`**: one row per source post or reply. Join `posts.thread_uid` to `threads.thread_uid`; `parent_post_id` preserves reply relationships. Rows include the post body, role, timestamps, source URLs, author attribution, and license information.
+`post_uid` is the stable dataset-wide ID and `source_post_uid` is the source-side ID. `thread_uid` joins a post to its thread; `parent_post_id` identifies its parent where the source provides one. `post_role` distinguishes questions, answers, issues, pull requests, discussions, comments, and other source roles. `body`, timestamps, author and URL fields retain post-level data; `thread_source_url` provides the source thread link when a source does not expose a direct link for an individual post. `accepted` is populated where a source has an accepted-answer signal; it is null where that concept does not apply. `reaction_count` contains the source score or reaction count when available. `license` and `license_evidence_url` are per-post.
 
-Both tables retain source provenance. `license` and `license_evidence_url` are provided per row because the corpus combines projects under different upstream licenses. Check those fields and follow the applicable source license before reusing individual records.
+## Attribution and licenses
 
-## Inclusion and deduplication
+This is a mixed-source dataset, so **there is no single license for the whole collection**. Each text row carries its source `license`, source URL, author attribution when available, and `license_evidence_url`. Follow the license on every row you use and preserve its attribution requirements. Stack Exchange licenses are taken from the API's per-post `content_license` field; the corresponding post timeline is linked for license evidence. GitHub and GitLab rows retain the repository-level license and evidence recorded during collection.
 
-The corpus includes records that pass the project’s text-safety filter and have `redistributable_with_attribution` provenance. Posts are included only when their parent thread is included. Repeated records with the same source ID are collapsed across snapshots, retaining the latest `updated_at` version. Similar text with different source IDs is retained.
+## Scope and exclusions
 
-This is an ordinary thread/post dataset. It contains no retrieval chunks, embeddings, vector index, generated preference pairs, ranking views, or synthetic quality labels.
+The corpus merges exact source IDs across the available Stack Exchange API snapshots and normalized GitHub/GitLab snapshots. Posts without text are omitted; a thread whose source item has a title but no body remains available in the threads table. Exact duplicates are resolved using the latest source update timestamp, with later snapshots breaking timestamp ties. Near-duplicate text is not removed. The included data is a snapshot, not a live feed.
 
-## Files and manifest
+The v0.2.0 Stack Exchange baseline is represented where the later API snapshots retained verifiable source metadata: 3,267/3,409 question IDs and 8,344/8,677 answer IDs. The remaining 142 questions and 333 answers are omitted because 139 question threads lack verifiable API attribution and 3 threads are held by the safety filter.
 
-The release contains `data/threads/train.parquet`, `data/posts/train.parquet`, this dataset card, and `manifest.json`. The manifest records row counts, checksums, source snapshots, and filtering rules.
+The Discourse pilot is not included because its current records are metadata-only and text redistribution permission was not verified. Rows without an eligible source license/attribution or rows held by the existing safety filter are excluded; the counts are in `manifest.json`.

data/posts/train.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0738eec67b233e4b4556dc4e8bfcfb8c084eda7daec111ac7dcde89377572ef5
-size 646828274
+oid sha256:fb958568de3aa656482bcadb558e9958c81189f4d97885a31ae7f9446d366a4d
+size 749265897

data/threads/train.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8e9f0a0edfeab232a1b01b10fb406e850dc72fe28d5f7664fbea40d46db7c03c
-size 342605598
+oid sha256:b5ca54eb433974c9585c52c984a3ca90fc90720ca6c86b5fa1fbe083ab704e38
+size 388851113

manifest.json

@@ -1,83 +1,59 @@
 {
   "artifact_type": "community_qa_lab_plain_thread_post_corpus",
-  "created_at_utc": "2026-09-30T04:52:12.490355+00:00",
-  "dataset_version": "v0.11.1",
+  "created_at_utc": "2026-09-30T07:45:08.878067+00:00",
+  "dataset_id": "aluha501/community-data",
+  "dataset_version": "v0.11.2",
   "deduplication": {
     "near_duplicate_removal": false,
     "post_key": "source_platform + repository + post_uid (fallback post_id)",
     "thread_key": "source_platform + repository + thread_uid (fallback thread_id)",
     "winner": "latest updated_at; later normalized snapshot breaks timestamp ties"
   },
-  "description": "Canonical source threads and posts; no retrieval chunking or task-specific views.",
+  "description": "Plain, source-attributed community threads and posts from Stack Exchange, GitHub, and GitLab.",
   "dropped_rows": {
-    "posts_parent_thread_not_included": 12505,
-    "posts_safety_or_provenance": 8540,
-    "threads_safety_or_provenance": 4366
+    "posts_empty_text": 11070,
+    "posts_parent_thread_not_included": 13489,
+    "posts_safety_or_provenance": 9554,
+    "threads_safety_or_provenance": 4739
   },
   "eligible_rows_before_dedup": {
-    "posts:github:github_discussion": 213610,
-    "posts:github:github_issue": 1257146,
-    "posts:github:github_pull_request": 11000,
-    "posts:gitlab:gitlab_issue": 1149,
-    "posts:gitlab:gitlab_merge_request": 594,
-    "threads:github:github_discussion": 65767,
-    "threads:github:github_issue": 361551,
-    "threads:github:github_pull_request": 4124,
-    "threads:gitlab:gitlab_issue": 1149,
-    "threads:gitlab:gitlab_merge_request": 594
+    "posts:github:github_discussion": 238477,
+    "posts:github:github_issue": 1399902,
+    "posts:github:github_pull_request": 13621,
+    "posts:gitlab:gitlab_issue": 1287,
+    "posts:gitlab:gitlab_merge_request": 510,
+    "posts:stackexchange:stackexchange_answer": 38251,
+    "posts:stackexchange:stackexchange_question": 20003,
+    "threads:github:github_discussion": 71743,
+    "threads:github:github_issue": 403905,
+    "threads:github:github_pull_request": 5466,
+    "threads:gitlab:gitlab_issue": 1382,
+    "threads:gitlab:gitlab_merge_request": 743,
+    "threads:stackexchange:stackexchange_question": 20003
   },
+  "excluded_sources": [
+    {
+      "reason": "metadata_only; forum-specific text redistribution permission was not verified",
+      "snapshot": "v0.7.0-discourse"
+    }
+  ],
   "files": {
     "posts": {
       "path": "data/posts/train.parquet",
-      "rows": 1464177,
-      "sha256": "0738eec67b233e4b4556dc4e8bfcfb8c084eda7daec111ac7dcde89377572ef5"
+      "rows": 1671299,
+      "sha256": "fb958568de3aa656482bcadb558e9958c81189f4d97885a31ae7f9446d366a4d"
     },
     "threads": {
       "path": "data/threads/train.parquet",
-      "rows": 425362,
-      "sha256": "8e9f0a0edfeab232a1b01b10fb406e850dc72fe28d5f7664fbea40d46db7c03c"
+      "rows": 489208,
+      "sha256": "b5ca54eb433974c9585c52c984a3ca90fc90720ca6c86b5fa1fbe083ab704e38"
     }
   },
   "input_counts": {
-    "post_rows": 1504544,
-    "thread_rows": 437551
+    "post_rows": 1746164,
+    "thread_rows": 507981
   },
   "input_snapshots": [
-    {
-      "manifest_sha256": "e4d66baf70713764675eb8dba4ef5e2550129302e8216cd4110a50134e620a1d",
-      "output_version": "v0.7.0-multisource-experimental",
-      "posts": 24645,
-      "snapshot": "v0.9.0-github-modern-ai",
-      "threads": 6203
-    },
-    {
-      "manifest_sha256": "48eac9ab6a5ffacec1fd5813c0fbdbceb885a8e0a4c9938773e916087b6cb4d6",
-      "output_version": "v0.7.0-multisource-experimental",
-      "posts": 107090,
-      "snapshot": "v0.9.4-github-modern-ai",
-      "threads": 30450
-    },
-    {
-      "manifest_sha256": "bb5775b7e7cda11c426d9f67ab5582c6904290f9e9e236bbd92d574e559af852",
-      "output_version": "v0.7.0-multisource-experimental",
-      "posts": 63564,
-      "snapshot": "v0.9.5-harmonized-modern-ai",
-      "threads": 16042
-    },
-    {
-      "manifest_sha256": "e09d63b9e5c0c902485e057a2f389faa1f103f8493e86e96c087e122233d10bb",
-      "output_version": "v0.9.9-github-expansion-experimental",
-      "posts": 90206,
-      "snapshot": "v0.9.9-github-expansion",
-      "threads": 21887
-    },
-    {
-      "manifest_sha256": "b8202ebb686d4a461fcd50ea6fdfa410f399bf0c6e55643b1aa3c0bcc29448ef",
-      "output_version": "v0.9.9-gitlab-expansion-experimental",
-      "posts": 1800,
-      "snapshot": "v0.9.9-gitlab-expansion",
-      "threads": 1800
-    },
     {
       "manifest_sha256": "70c4aa5e85445210dd28c005ae827ec91045c3c708cd8fc5e5ce89c1ec4e1d2d",
       "output_version": "v0.10.0-github-expansion-experimental",
@@ -175,34 +151,316 @@
       "posts": 146195,
       "snapshot": "v0.11.0-ai-expansion-wave-12",
       "threads": 40611
+    },
+    {
+      "manifest_sha256": "f6ddcb8ca28a30e44f9f8df9ed621bec121719d4fa54ff1131841aedd5ff4499",
+      "output_version": "v0.11.0-ai-expansion-wave-13-experimental",
+      "posts": 83299,
+      "snapshot": "v0.11.0-ai-expansion-wave-13",
+      "threads": 24825
+    },
+    {
+      "manifest_sha256": "fdcca70cbdbb073b074700616c5bfc722ae58c240ba9dcb2d85c3df25d3e7ab3",
+      "output_version": "v0.11.0-ai-expansion-wave-14-experimental",
+      "posts": 75533,
+      "snapshot": "v0.11.0-ai-expansion-wave-14",
+      "threads": 19898
+    },
+    {
+      "manifest_sha256": "99a2941c4085c7c9155561b426d207a820f3b5948f65d38adcf5f21d71b482a3",
+      "output_version": "v0.7.0-gitlab-retrieval-experimental",
+      "posts": 390,
+      "snapshot": "v0.7.0-gitlab",
+      "threads": 390
+    },
+    {
+      "manifest_sha256": "c86de923c531af5315a41a89a7bb83c8946b4161155109de70a396c021c27fcb",
+      "output_version": "v0.7.0-multisource-experimental",
+      "posts": 231,
+      "snapshot": "v0.7.0-multisource",
+      "threads": 29
+    },
+    {
+      "manifest_sha256": "e609257680d13f05eb15b37711d6757de554d5c87f77c1450bc7c3665143fe22",
+      "output_version": "v0.7.0-multisource-experimental",
+      "posts": 3537,
+      "snapshot": "v0.7.0-multisource-extended",
+      "threads": 904
+    },
+    {
+      "manifest_sha256": "8b4c7fa4e4c3e3fdd5477be3c484ca014ac53405fd93a33fa3d06a1fcc282c4a",
+      "output_version": "v0.7.0-multisource-experimental",
+      "posts": 963,
+      "snapshot": "v0.7.0-multisource-large",
+      "threads": 229
+    },
+    {
+      "manifest_sha256": "5a82a671365a6dfc3524309879645f7bc2968e37ae8063cdd47c2b222e958ab7",
+      "output_version": "v0.7.0-multisource-experimental",
+      "posts": 12449,
+      "snapshot": "v0.7.0-multisource-scale",
+      "threads": 3018
+    },
+    {
+      "manifest_sha256": "5b40644d184ea149909cb75392bb66f714d4aecce0953e998b56fa070ec04da2",
+      "output_version": "v0.7.0-multisource-experimental",
+      "posts": 6706,
+      "snapshot": "v0.7.0-multisource-xlarge",
+      "threads": 1058
+    },
+    {
+      "manifest_sha256": "e4d66baf70713764675eb8dba4ef5e2550129302e8216cd4110a50134e620a1d",
+      "output_version": "v0.7.0-multisource-experimental",
+      "posts": 24645,
+      "snapshot": "v0.9.0-github-modern-ai",
+      "threads": 6203
+    },
+    {
+      "manifest_sha256": "48eac9ab6a5ffacec1fd5813c0fbdbceb885a8e0a4c9938773e916087b6cb4d6",
+      "output_version": "v0.7.0-multisource-experimental",
+      "posts": 107090,
+      "snapshot": "v0.9.4-github-modern-ai",
+      "threads": 30450
+    },
+    {
+      "manifest_sha256": "bb5775b7e7cda11c426d9f67ab5582c6904290f9e9e236bbd92d574e559af852",
+      "output_version": "v0.7.0-multisource-experimental",
+      "posts": 63564,
+      "snapshot": "v0.9.5-harmonized-modern-ai",
+      "threads": 16042
+    },
+    {
+      "manifest_sha256": "e09d63b9e5c0c902485e057a2f389faa1f103f8493e86e96c087e122233d10bb",
+      "output_version": "v0.9.9-github-expansion-experimental",
+      "posts": 90206,
+      "snapshot": "v0.9.9-github-expansion",
+      "threads": 21887
+    },
+    {
+      "manifest_sha256": "b8202ebb686d4a461fcd50ea6fdfa410f399bf0c6e55643b1aa3c0bcc29448ef",
+      "output_version": "v0.9.9-gitlab-expansion-experimental",
+      "posts": 1800,
+      "snapshot": "v0.9.9-gitlab-expansion",
+      "threads": 1800
+    },
+    {
+      "manifest_sha256": "ffab2ad04d288a2c95ae672bce60abe6869d27e6a610b0c6cc957dcbe06799c0",
+      "output_version": "v0.4.0-plain-stackexchange-normalized",
+      "posts": 14245,
+      "snapshot": "stackexchange-v0.4.0",
+      "threads": 4039
+    },
+    {
+      "manifest_sha256": "2a74d1e034444a1b69c818ae88d0917b7a950a6eaa67c75d38c51e7033a09a75",
+      "output_version": "v0.5.0-plain-stackexchange-normalized",
+      "posts": 6804,
+      "snapshot": "stackexchange-v0.5.0",
+      "threads": 1880
+    },
+    {
+      "manifest_sha256": "b9451a603c95f310f023da0a067c31d8f4e17a3ed4cfdbb973de96df5722defe",
+      "output_version": "v0.6.0-expanded-plain-stackexchange-normalized",
+      "posts": 26308,
+      "snapshot": "stackexchange-v0.6.0-expanded",
+      "threads": 9894
+    },
+    {
+      "manifest_sha256": "6d1e03a4035b94eb84a6976ba577eb546e9858c74fa5d882eb2c483aef3421cc",
+      "output_version": "v0.7.0-stackexchange-expansion-plain-stackexchange-normalized",
+      "posts": 11155,
+      "snapshot": "stackexchange-v0.7.0-stackexchange-expansion",
+      "threads": 4266
     }
   ],
-  "remote_writes_during_build": false,
+  "published_audit": {
+    "duplicate_post_uids": 0,
+    "duplicate_thread_uids": 0,
+    "empty_posts": 0,
+    "empty_threads": 0,
+    "legacy_stackexchange_v0_2": {
+      "baseline_answer_ids": 8677,
+      "baseline_question_ids": 3409,
+      "excluded_answer_ids": 333,
+      "excluded_by_reason": {
+        "missing_question_attribution": {
+          "answers": 325,
+          "questions": 139
+        },
+        "safety_hold": {
+          "answers": 8,
+          "questions": 3
+        }
+      },
+      "excluded_question_ids": 142,
+      "included_answer_ids": 8344,
+      "included_question_ids": 3267,
+      "source_snapshot": "v0.2.0"
+    },
+    "missing_post_license": 0,
+    "missing_post_license_evidence_url": 0,
+    "missing_post_source_url": 43887,
+    "missing_post_thread_source_url": 0,
+    "missing_thread_license": 0,
+    "missing_thread_license_evidence_url": 0,
+    "missing_thread_source_url": 0,
+    "orphan_posts": 0,
+    "post_licenses": {
+      "AGPL-3.0-only": 643,
+      "Apache-2.0": 941772,
+      "BSD-3-Clause": 88271,
+      "CC BY-SA 3.0": 3584,
+      "CC BY-SA 4.0": 46166,
+      "MIT": 590816,
+      "Matplotlib": 47
+    },
+    "posts_by_platform": {
+      "github": 1620060,
+      "gitlab": 1489,
+      "stackexchange": 49750
+    },
+    "posts_by_source_kind": {
+      "github_discussion": 223678,
+      "github_issue": 1384974,
+      "github_pull_request": 11408,
+      "gitlab_issue": 1081,
+      "gitlab_merge_request": 408,
+      "stackexchange_answer": 32229,
+      "stackexchange_question": 17521
+    },
+    "stackexchange_communities": [
+      "ai.stackexchange.com",
+      "arduino.stackexchange.com",
+      "askubuntu.com",
+      "bioinformatics.stackexchange.com",
+      "blender.stackexchange.com",
+      "codereview.stackexchange.com",
+      "computergraphics.stackexchange.com",
+      "cs.stackexchange.com",
+      "cs50.stackexchange.com",
+      "cstheory.stackexchange.com",
+      "datascience.stackexchange.com",
+      "dba.stackexchange.com",
+      "devops.stackexchange.com",
+      "dsp.stackexchange.com",
+      "economics.stackexchange.com",
+      "electronics.stackexchange.com",
+      "engineering.stackexchange.com",
+      "gamedev.stackexchange.com",
+      "genai.stackexchange.com",
+      "gis.stackexchange.com",
+      "linguistics.stackexchange.com",
+      "math.stackexchange.com",
+      "mathematica.stackexchange.com",
+      "mattermodeling.stackexchange.com",
+      "medicalsciences.stackexchange.com",
+      "networkengineering.stackexchange.com",
+      "opendata.stackexchange.com",
+      "opensource.stackexchange.com",
+      "or.stackexchange.com",
+      "physics.stackexchange.com",
+      "pm.stackexchange.com",
+      "proofassistants.stackexchange.com",
+      "quantumcomputing.stackexchange.com",
+      "raspberrypi.stackexchange.com",
+      "reverseengineering.stackexchange.com",
+      "robotics.stackexchange.com",
+      "scicomp.stackexchange.com",
+      "security.stackexchange.com",
+      "serverfault.com",
+      "softwareengineering.stackexchange.com",
+      "sqa.stackexchange.com",
+      "stackoverflow.com",
+      "stats.stackexchange.com",
+      "superuser.com",
+      "sustainability.stackexchange.com",
+      "tex.stackexchange.com",
+      "unix.stackexchange.com",
+      "webmasters.stackexchange.com"
+    ],
+    "thread_licenses": {
+      "AGPL-3.0-only": 62,
+      "Apache-2.0": 271318,
+      "BSD-3-Clause": 20321,
+      "CC BY-SA 3.0": 1010,
+      "CC BY-SA 4.0": 16511,
+      "MIT": 179974,
+      "Matplotlib": 12
+    },
+    "threads_by_platform": {
+      "github": 469944,
+      "gitlab": 1743,
+      "stackexchange": 17521
+    },
+    "threads_by_source_kind": {
+      "github_discussion": 66766,
+      "github_issue": 398602,
+      "github_pull_request": 4576,
+      "gitlab_issue": 1149,
+      "gitlab_merge_request": 594,
+      "stackexchange_question": 17521
+    }
+  },
+  "remote_writes": false,
   "rows": {
-    "posts": 1464177,
-    "threads": 425362
+    "posts": 1671299,
+    "threads": 489208
   },
   "safety_and_provenance": {
     "included": "safety_status=pass and provenance_status=redistributable_with_attribution",
     "posts_require_included_parent_thread": true
   },
   "schema_version": 1,
-  "source_build_dataset_version": "v0.11.1-plain-multisource-local",
+  "scope": {
+    "data_shape": "ordinary threads and posts",
+    "embeddings": false,
+    "near_duplicate_removal": false,
+    "retrieval_chunks": false,
+    "task_specific_views": false
+  },
   "source_counts": {
-    "posts:github:github_discussion": 216251,
-    "posts:github:github_issue": 1275392,
-    "posts:github:github_pull_request": 11101,
-    "posts:gitlab:gitlab_issue": 1200,
-    "posts:gitlab:gitlab_merge_request": 600,
-    "threads:github:github_discussion": 66308,
-    "threads:github:github_issue": 365303,
-    "threads:github:github_pull_request": 4140,
-    "threads:gitlab:gitlab_issue": 1200,
-    "threads:gitlab:gitlab_merge_request": 600
+    "posts:github:github_discussion": 241759,
+    "posts:github:github_issue": 1428710,
+    "posts:github:github_pull_request": 14993,
+    "posts:gitlab:gitlab_issue": 1440,
+    "posts:gitlab:gitlab_merge_request": 750,
+    "posts:stackexchange:stackexchange_answer": 38433,
+    "posts:stackexchange:stackexchange_question": 20079,
+    "threads:github:github_discussion": 72319,
+    "threads:github:github_issue": 407910,
+    "threads:github:github_pull_request": 5483,
+    "threads:gitlab:gitlab_issue": 1440,
+    "threads:gitlab:gitlab_merge_request": 750,
+    "threads:stackexchange:stackexchange_question": 20079
+  },
+  "source_license_policy": {
+    "collection_license": "mixed; no single license applies to the complete dataset",
+    "redistribution_gate": "source license/attribution must be present and provenance_status must be redistributable_with_attribution",
+    "row_level_fields": [
+      "license",
+      "license_evidence_url",
+      "source_url",
+      "author_name",
+      "author_url"
+    ],
+    "stackexchange_license_source": "API per-post content_license field; post timeline evidence URL retained"
+  },
+  "source_reconstruction_exclusions": {
+    "stackexchange-v0.4.0": {
+      "missing_question_attribution": 175
+    },
+    "stackexchange-v0.5.0": {
+      "missing_question_attribution": 50
+    },
+    "stackexchange-v0.6.0-expanded": {
+      "missing_question_attribution": 523
+    },
+    "stackexchange-v0.7.0-stackexchange-expansion": {
+      "missing_question_attribution": 161
+    }
   },
   "splits": {
     "posts": "train only",
     "threads": "train only"
   },
-  "status": "experimental"
+  "status": "ready_to_publish"
 }