Data layer at release v1.0.0: frozen-snapshot notice replaces the pre-release notice; CITATION.cff version 1.0.0

CITATION.cff

@@ -7,7 +7,7 @@ message: >-
   bundle, and the commit for the tooling that produced a number. See
   the dataset card's Citation section for the citation rules (snapshot not branch; name
   the collection round; cite client version plus service, not the model alone).
-  This is a pre-release (last revised 2026-09-11): cite the snapshot, never a branch.
+  Released 2026-09-11 as version 1.0.0 (methodology v0.1): cite the tag or the DOI, never a branch.
 title: "AI Service QoS: An Independent Outside-In Measurement Methodology, with Case Study 1 (Qwen3.8-Max / Qoder)"
 authors:
   - name: "2makeitwork"
@@ -15,8 +15,8 @@ authors:
     # affiliation is to be added here or in any listing, archive record or citation form
     # Do not attempt to resolve the handle to a person.
 type: dataset
-version: "0.3.0-pre.4"    # matches the tag v0.3.0-pre.4, the newest citable snapshot; scripts/preflight.sh --release <tag> refuses if they differ
-date-released: "2026-09-11"   # last revision of the document set, not a release date
+version: "1.0.0"    # matches the tag v1.0.0, the citable snapshot; scripts/preflight.sh --release <tag> and scripts/analyze.py both refuse if they differ
+date-released: "2026-09-11"   # the release date of version 1.0.0
 # doi: ""                 # dataset DOI, then paper-page DOI
 repository-code: "https://github.com/2makeitwork/paper_ai_Qos"
 repository-artifact: "https://huggingface.co/datasets/2makeitwork/paper_ai_Qos"
@@ -24,7 +24,7 @@ license: >-
   CC-BY-4.0 (text, tables, figures and evidence data; see LICENSE-content.md);
   MIT (executable tooling under scripts/; see LICENSE-code.md)
 # A commit hash cannot be written into the commit it names, so the snapshot to cite is the
-# tag: v0.3.0-pre.4. Zenodo's GitHub integration mints the version DOI from the release page.
+# tag: v1.0.0. Zenodo's GitHub integration mints the version DOI from the release page.
 keywords:
   - AI service quality
   - large language model service measurement
@@ -43,6 +43,14 @@ preferred-citation:
   # url: "https://huggingface.co/papers/..."   # paper page once published
   institution: "Independent"
 versions:
+  - date-released: "2026-09-11"
+    version: "1.0.0"
+    changes: "Formal release of methodology v0.1 with Case Study 1. The contribution is positioned
+      as a novelty of perspective rather than of method: the instruments are shared with prior
+      external measurement, while the vantage — the paying user's side, published as a stratified
+      public aggregate that checks a provider against its own quality-of-service claim without its
+      cooperation — is occupied by none of them. Section 7.1 states the institution this serves and
+      the two design duties not yet discharged."
   - date-released: "2026-09-11"
     version: "0.3.0-pre.4"
     changes: "Pre-release line. Methodology v0.1 with Case Study 1: paper section 1.4 rebuilt as a

README.md

@@ -41,11 +41,14 @@ configs:
         path: evidence/logs/timeout_resume_events.csv
 ---
 
-> **Pre-release, last revised 2026-09-11** (methodology v0.1). Definitions, wording and figures
-> may still change; cite a snapshot — a release tag or a commit, not a branch — so that a later
-> revision cannot move what your citation points at. Every figure quoted in this document set is
-> asserted against the shipped evidence by `scripts/analyze.py`, which fails when a number and the
-> evidence disagree and warns when a file is edited after the date above.
+> **Released 2026-09-11 as v1.0.0** (methodology v0.1). This snapshot is frozen: later
+> corrections and additions ship as a new version with a new tag, never as an edit to the bytes a
+> citation points at — cite the tag or the Digital Object Identifier, not a branch. Two version
+> numbers mean two different things here: the tag names this public snapshot, while *methodology
+> v0.1* names the definitions, which stay unchanged until a second provider has been measured
+> (section 7). Every figure quoted in this document set is asserted against the shipped evidence by
+> `scripts/analyze.py`, which fails when a number and the evidence disagree and warns when a file
+> is edited after the date above.
 
 **AI Service QoS — Independent Measurement of What Users Actually Receive** — methodology v0.1, and the first data product of an independent, outside-in measurement framework. The system measured here is **Case Study 1**, not the subject: the framework claims provider neutrality and one provider is not proof of it — a second provider measured with *these same definitions and no new ones* is the test (paper section 7, Stage 3).
 
@@ -81,7 +84,7 @@ recomputed from these files rather than taken on trust.
 The whole published repository was downloaded back into an empty directory and compared file by
 file with the tag it was built from — which is exactly what `scripts/check_release_sync.py` does,
 so this is re-checkable by a reader rather than autobiographical. Measured 2026-09-11 against tag
-`v0.3.0-pre.4`: **every data-layer file agrees byte for byte**, `README.md` included — that file
+`v1.0.0`: **every data-layer file agrees byte for byte**, `README.md` included — that file
 is this card with its three internal-instruction lines removed, and the source repository's own
 `README.md` is a different document, its homepage. The hub holds one file no tag contains: the
 `.gitattributes` it created with the repository. The generated table ends with a line carrying a

anonymization.md

@@ -1,10 +1,13 @@
 # Anonymization policy
 
-> **Pre-release, last revised 2026-09-11** (methodology v0.1). Definitions, wording and figures
-> may still change; cite a snapshot — a release tag or a commit, not a branch — so that a later
-> revision cannot move what your citation points at. Every figure quoted in this document set is
-> asserted against the shipped evidence by `scripts/analyze.py`, which fails when a number and the
-> evidence disagree and warns when a file is edited after the date above.
+> **Released 2026-09-11 as v1.0.0** (methodology v0.1). This snapshot is frozen: later
+> corrections and additions ship as a new version with a new tag, never as an edit to the bytes a
+> citation points at — cite the tag or the Digital Object Identifier, not a branch. Two version
+> numbers mean two different things here: the tag names this public snapshot, while *methodology
+> v0.1* names the definitions, which stay unchanged until a second provider has been measured
+> (section 7). Every figure quoted in this document set is asserted against the shipped evidence by
+> `scripts/analyze.py`, which fails when a number and the evidence disagree and warns when a file
+> is edited after the date above.
 
 Everything under `evidence/` is produced by `scripts/anonymize.py`, which
 builds a deterministic pseudonym map from the raw dump and refuses to finish

methodology.md

@@ -1,10 +1,13 @@
 # Methodology
 
-> **Pre-release, last revised 2026-09-11** (methodology v0.1). Definitions, wording and figures
-> may still change; cite a snapshot — a release tag or a commit, not a branch — so that a later
-> revision cannot move what your citation points at. Every figure quoted in this document set is
-> asserted against the shipped evidence by `scripts/analyze.py`, which fails when a number and the
-> evidence disagree and warns when a file is edited after the date above.
+> **Released 2026-09-11 as v1.0.0** (methodology v0.1). This snapshot is frozen: later
+> corrections and additions ship as a new version with a new tag, never as an edit to the bytes a
+> citation points at — cite the tag or the Digital Object Identifier, not a branch. Two version
+> numbers mean two different things here: the tag names this public snapshot, while *methodology
+> v0.1* names the definitions, which stay unchanged until a second provider has been measured
+> (section 7). Every figure quoted in this document set is asserted against the shipped evidence by
+> `scripts/analyze.py`, which fails when a number and the evidence disagree and warns when a file
+> is edited after the date above.
 
 > **Methodology v0.1** — the concepts below are frozen. The stage ladder to v0.5, a second provider and a continuous observatory is `paper_ai_QoS.md` section 7.
 
@@ -28,7 +31,7 @@ and real conversation identifiers):
 
 | Step | Command | Result of the attempt |
 |---|---|---|
-| Derived statistics from the published evidence | `python3 scripts/analyze.py` | **reproduces exactly.** 57 assertions pass. The output file ends with a content hash plus a second-precision timestamp, and when the hash is unchanged the file is not rewritten at all — so running the checks cannot dirty the tree, and two copies with the same hash hold the same statistics whatever their stamps say |
+| Derived statistics from the published evidence | `python3 scripts/analyze.py` | **reproduces exactly.** 58 assertions pass. The output file ends with a content hash plus a second-precision timestamp, and when the hash is unchanged the file is not rewritten at all — so running the checks cannot dirty the tree, and two copies with the same hash hold the same statistics whatever their stamps say |
 | Card against files | `python3 scripts/verify_dataset_card.py` | **reproduces exactly**, eight checks, in either repository |
 | Log extracts from the pinned snapshot | `LOGSRC=<snapshot>/logs scripts/collect_evidence.sh` then `scripts/anonymize.py` | **does not reproduce the frozen bundle.** Of 24 published items, 4 matched byte for byte, 11 differed, and 9 were not produced at all (`evidence/screenshots/`, `evidence/user_statements.md`, four extracts, two statistics files, the round-2 directory) |
 

scripts/analyze.py

@@ -302,24 +302,33 @@ else:
     print("  SKIP no drift check: the paper and the case report are not in this "
           "checkout (data-layer repository)")
 
-# ---- pre-release notice: present, and carrying the same revision date, everywhere ----
+# ---- release notice: one text, one version, present everywhere ----
 NOTICE_DATE = "2026-09-11"
+RELEASED = "1.0.0"        # the released version; CITATION.cff must carry this same string
+NOTICE_MARK = f"**Released {NOTICE_DATE} as v{RELEASED}**"
 PUB_DOCS = [doc for doc in (EV.parent / n for n in
             ("paper_ai_QoS.md", "report_qwenAliServiceQuality.md", "README.md", "abstract.md",
              "methodology.md", "anonymization.md", "ask_vendor.md", "DATA_REQUEST.md",
              "DATASET_CARD.md", "zenodo/README.md")) if doc.exists()]
 notice_problems = [doc.name for doc in PUB_DOCS
-                   if f"**Pre-release, last revised {NOTICE_DATE}**" not in doc.read_text(errors="replace")]
+                   if NOTICE_MARK not in doc.read_text(errors="replace")]
 if PUB_DOCS:
-    check("pre-release notice present and dated " + NOTICE_DATE + " in every published document",
+    check(f"release notice {NOTICE_MARK} present in every published document",
           notice_problems, [])
+    # The version lives in two places, so one is checked against the other: a notice that names a
+    # version CITATION.cff does not would send a citation to a snapshot that is not this one.
+    cff = EV.parent / "CITATION.cff"
+    cff_found = re.search(r'^version: "([^"]+)"', cff.read_text(errors="replace"), re.M) if cff.exists() else None
+    if cff.exists():
+        check("CITATION.cff and the release notices carry the same version",
+              cff_found.group(1) if cff_found else "unreadable", RELEASED)
     late = [doc.name for doc in PUB_DOCS
             if datetime.fromtimestamp(doc.stat().st_mtime).date() > datetime.strptime(NOTICE_DATE, "%Y-%m-%d").date()]
     if late:
-        print("  ADVISORY: edited after the stated revision date — bump NOTICE_DATE and the "
-              "notices together: " + ", ".join(sorted(late)))
+        print("  ADVISORY: edited after the release date — a released snapshot is frozen, so "
+              "these edits belong in the next version: " + ", ".join(sorted(late)))
 else:
-    print("  SKIP pre-release notice check: no published documents in this checkout (data layer)")
+    print("  SKIP release notice check: no published documents in this checkout (data layer)")
 
 # ---- authorship: one handle, no institution, anywhere in the public set ----
 BYLINE = "independent, unaffiliated"
@@ -342,7 +351,7 @@ if byline_docs:
         check("archive metadata carries the pseudonym with an empty affiliation", bad, [])
 
 # ---- published documents must not send a reader to a file that is not published ----
-PRIVATE = ["PUBLICATION.md", "prior-works.md", "reposition.md", "methodology_guidance"]
+PRIVATE = ["PUBLICATION.md", "prior-works.md", "methodology_guidance"]
 DISCLOSED = ["not published", "not part of", "deliberately", "private", "ignored",
              "working notes", "kept local", "no reading obligation"]
 private_problems = []

scripts/check_release_sync.py

@@ -64,7 +64,7 @@ STAMP_RE = re.compile(r"sha256:([0-9a-f]{64})")
 ALWAYS_BAD = ("Paste this file", "How to publish", "how to publish")
 # Names of files that stay on the author's workstation. Naming one is acceptable only on a
 # line that says so, which is what the source repository's own document table does.
-PRIVATE_NAMES = ("PUBLICATION.md", "prior-works.md", "reposition.md", "methodology_guidance")
+PRIVATE_NAMES = ("PUBLICATION.md", "prior-works.md", "methodology_guidance")
 DISCLOSED_WORDS = ("not published", "not part of", "deliberately", "private", "ignored",
                    "working notes", "kept local", "no reading obligation", "local-only")
 

scripts/verify_dataset_card.py

@@ -129,7 +129,7 @@ def check_licenses(yaml_text: str) -> None:
 
 
 INTERNAL_MARKERS = ("Paste this file", "How to publish", "how to publish",
-                    "PUBLICATION.md", "prior-works.md", "reposition.md", "methodology_guidance")
+                    "PUBLICATION.md", "prior-works.md", "methodology_guidance")
 # A fenced command is an instruction the reader will try. Every script it names must be
 # present in whichever repository this card is published into, or the line has to say who it
 # belongs to - which is how the tools/ gate and the paste-in note escaped into the open.