| Next revision | Previous revision |
| provenance:programming:crawler:webxray:residue_pass [2026/09/11 19:20] – New: code appendix for the 2026-09-11 second pass over the residue - probe, brief, merge gate, corrections, report, guard, mutation harness and every unedited output including the full 40-domain probe log and the estimator diff. Authored by Claude karel.kubicek.claude | provenance:programming:crawler:webxray:residue_pass [2026/09/11 22:43] (current) – Refresh check_tail_figures.py (derives the hand-changed verdict count from the corrections output and asserts the superseded one absent) and the mutation harness, now 24 mutations, 0 survivors. Authored by Claude karel.kubicek.claude |
|---|
| are on [[literature:corpus]]. | are on [[literature:corpus]]. |
| |
| **Sections are in letter order, A–P, and that is also chronological:** A–H are | **Sections are in letter order, A–R, and that is also chronological:** A–I are |
| the scripts, I–P the outputs. | the scripts, J–R the outputs. |
| |
| **Why this is a third page.** These blocks are 343 kB and the | **Why this is a third page.** These blocks are 392 kB and the |
| random-sample appendix's are already 285 kB; putting them together would | random-sample appendix's are already 286 kB; putting them together would |
| make one 628 kB page carrying two probe logs. Both figures are computed | make one 678 kB page carrying two probe logs. Both figures are computed |
| when this page is generated, from the same file lists the two manifests use, so | when this page is generated, from the same file lists the two manifests use, so |
| neither can go stale while the other moves. That page's own "why this is a | neither can go stale while the other moves. That page's own "why this is a |
| | D | ''scripts/owner_tail_merge.py'' | 7,950 | ''b7fa9c9a573c6e2f'' | no | | | D | ''scripts/owner_tail_merge.py'' | 7,950 | ''b7fa9c9a573c6e2f'' | no | |
| | E | ''scripts/owner_tail_corrections.py'' | 8,654 | ''4f30dc28ccd516bf'' | no | | | E | ''scripts/owner_tail_corrections.py'' | 8,654 | ''4f30dc28ccd516bf'' | no | |
| | F | ''scripts/report_tail_pass.py'' | 15,887 | ''77b0b52f94b5d3cf'' | no | | | F | ''scripts/report_tail_pass.py'' | 20,276 | ''91ef54dfa24bf452'' | no | |
| | G | ''scripts/check_tail_figures.py'' | 8,022 | ''28c545f60b3dcd8f'' | no | | | G | ''scripts/check_tail_figures.py'' | 17,939 | ''401e5956f79c89e9'' | no | |
| | H | ''scripts/check_tail_figures_mutations.sh'' | 2,649 | ''30e527342b74ba57'' | no | | | H | ''scripts/check_tail_merge_mutations.sh'' | 2,704 | ''6a77cccbcb73f91b'' | no | |
| | I | ''out/tail_probe.txt'' | 177,214 | ''5f487708ad8b7ba9'' | **yes** — one lone CR normalised to LF | | | I | ''scripts/check_tail_figures_mutations.sh'' | 4,975 | ''6664ce97cdbfb5a0'' | no | |
| | J | ''out/owner_tail_merge-output.txt'' | 1,042 | ''b1fe634a350c43bc'' | no | | | J | ''out/check_tail_merge_mutations-output.txt'' | 1,142 | ''ebb83f0c555d0527'' | no | |
| | K | ''out/tail_verify-output.txt'' | 2,512 | ''f772acd997a598d8'' | no | | | K | ''out/tail_probe.txt'' | 177,214 | ''5f487708ad8b7ba9'' | **yes** — one lone CR normalised to LF | |
| | L | ''out/owner_tail_corrections-output.txt'' | 6,659 | ''df43d5b6edba64dd'' | no | | | L | ''out/owner_tail_merge-output.txt'' | 1,042 | ''b1fe634a350c43bc'' | no | |
| | M | ''out/owner_random_sample-tail-output.txt'' | 52,594 | ''f46dd7c6f37fb231'' | no | | | M | ''out/tail_verify-output.txt'' | 2,512 | ''f772acd997a598d8'' | no | |
| | N | ''out/owner_random_sample-tail.diff'' | 38,398 | ''e57cc75561a824d8'' | no | | | N | ''out/owner_tail_corrections-output.txt'' | 6,659 | ''df43d5b6edba64dd'' | no | |
| | O | ''out/report_tail_pass-output.txt'' | 9,683 | ''12c6a9465c08b271'' | no | | | O | ''out/owner_random_sample-tail-output.txt'' | 52,594 | ''f46dd7c6f37fb231'' | no | |
| | P | ''out/check_tail_figures_mutations-output.txt'' | 793 | ''788c9e9d841e7c76'' | no | | | P | ''out/owner_random_sample-tail.diff'' | 63,634 | ''07ae04745a451e50'' | no | |
| | | Q | ''out/report_tail_pass-output.txt'' | 13,529 | ''c5e2c644f7b4c35c'' | no | |
| | | R | ''out/check_tail_figures_mutations-output.txt'' | 1,329 | ''9c28a7c4448ac143'' | no | |
| |
| ===== A. scripts/whois43.py — a WHOIS client, because this container has none ===== | ===== A. scripts/whois43.py — a WHOIS client, because this container has none ===== |
| |
| python3 report_tail_pass.py --sample out/owner_sample.json \ | python3 report_tail_pass.py --sample out/owner_sample.json \ |
| --before out/adj_rows_scored.json --after out/adj_rows_tail.json \ | --before out/adj_rows_scored.json --after out/adj_rows_tail_final.json \ |
| --before-output out/owner_random_sample-output.txt \ | --before-output out/owner_random_sample-output.txt \ |
| --after-output out/owner_random_sample-tail-output.txt | --after-output out/owner_random_sample-tail-output.txt |
| |
| A second pass with harder sourcing gives a partial answer, and only a partial | A second pass with harder sourcing gives a partial answer, and only a partial |
| one. The rows it recovers are, by construction, the **easier end of the | one: it shrinks the region nothing can speak for, without putting a bound on |
| | what is left inside it. The rows it recovers are, by construction, the **easier end of the |
| residue** -- the ones a register number or an archived legal page could reach. | residue** -- the ones a register number or an archived legal page could reach. |
| Whatever is still unresolved after it is harder still, so the comparison below | Whatever is still unresolved after it is harder still, so the comparison below |
| print("\n--- B. Are the recovered rows worse than the rows already in the rates? ---") | print("\n--- B. Are the recovered rows worse than the rows already in the rates? ---") |
| print("\nRaw counts, unweighted, because this is a comparison between two groups of") | print("\nRaw counts, unweighted, because this is a comparison between two groups of") |
| print("drawn rows and not an estimate for the list. current+granularity against") | print("drawn rows and not an estimate for the list. Fisher's exact, two-sided. A") |
| print("stale+error; Fisher's exact, two-sided. A significantly worse recovered") | print("significantly worse recovered group would support the worry that the rates") |
| print("group would support the worry that the rates are flattered.") | print("are flattered.") |
| print(f"\n{'List':14} {'already-eligible good/n':>24} {'recovered good/n':>18} {'p (Fisher)':>11}") | print() |
| pooled = [0, 0, 0, 0] | print("**Both metrics, because the answer depends on which one you ask.** The first") |
| for lab in LISTS: | print("version of this script tested current+granularity here and current-only in") |
| old_names = {d for st in eb[lab][1] for d in st} | print("section E2 -- each the choice that makes its own claim mildest -- and did not") |
| og = on = rg = rn = 0 | print("say so. A review pass caught it. Neither metric is privileged: `current` is") |
| for h in range(4): | print("the stricter reading, `current+granularity` is the one the content page says") |
| for (p_, v), d in zip(ea[lab][0][h], ea[lab][1][h]): | print("is right when the unit of analysis is 'which company'.") |
| if d in old_names: | for metric, pick in (("current only", lambda v: v == "current"), |
| on += 1; og += (v in GOOD) | ("current+granularity", lambda v: v in GOOD)): |
| else: | print(f"\n metric: {metric}") |
| rn += 1; rg += (v in GOOD) | print(f" {'List':14} {'already-eligible good/n':>24} {'recovered good/n':>18} {'p (Fisher)':>11}") |
| if rn == 0: | pooled = [0, 0, 0, 0] |
| print(f"{lab:14} {f'{og}/{on}':>24} {'0/0':>18} {'n/a':>11}") | for lab in LISTS: |
| continue | old_names = {d for st in eb[lab][1] for d in st} |
| p = fisher_two_sided(og, on - og, rg, rn - rg) | og = on = rg = rn = 0 |
| pooled[0] += og; pooled[1] += on - og; pooled[2] += rg; pooled[3] += rn - rg | for h in range(4): |
| print(f"{lab:14} {f'{og}/{on}':>24} {f'{rg}/{rn}':>18} {p:>11.3f}") | for (p_, v), d in zip(ea[lab][0][h], ea[lab][1][h]): |
| if pooled[2] + pooled[3]: | if d in old_names: |
| p = fisher_two_sided(*pooled) | on += 1; og += pick(v) |
| print(f"{'pooled':14} {f'{pooled[0]}/{pooled[0]+pooled[1]}':>24} " | else: |
| f"{f'{pooled[2]}/{pooled[2]+pooled[3]}':>18} {p:>11.3f}") | rn += 1; rg += pick(v) |
| | if rn == 0: |
| | print(f" {lab:14} {f'{og}/{on}':>24} {'0/0':>18} {'n/a':>11}") |
| | continue |
| | p = fisher_two_sided(og, on - og, rg, rn - rg) |
| | pooled[0] += og; pooled[1] += on - og; pooled[2] += rg; pooled[3] += rn - rg |
| | print(f" {lab:14} {f'{og}/{on}':>24} {f'{rg}/{rn}':>18} {p:>11.3f}") |
| | if pooled[2] + pooled[3]: |
| | p = fisher_two_sided(*pooled) |
| | print(f" {'pooled':14} {f'{pooled[0]}/{pooled[0]+pooled[1]}':>24} " |
| | f"{f'{pooled[2]}/{pooled[2]+pooled[3]}':>18} {p:>11.3f}") |
| print("\nThe pooled row counts a domain once per list that drew it, so it is not an") | print("\nThe pooled row counts a domain once per list that drew it, so it is not an") |
| print("independent test; it is printed as a summary of direction, not as evidence.") | print("independent test; it is printed as a summary of direction, not as evidence.") |
| # ---- E. rate movement, from the one estimator --------------------------- | # ---- E. rate movement, from the one estimator --------------------------- |
| print("\n--- E. Rate movement, computed with owner_random_sample.py's estimator ---") | print("\n--- E. Rate movement, computed with owner_random_sample.py's estimator ---") |
| print("\nImported, not reimplemented. 95% intervals are stratified bootstrap") | print("\nImported, not reimplemented -- the point estimator is owner_random_sample.py's") |
| print(f"percentiles, {ors.NBOOT:,} resamples, seed {ors.SEED} -- the estimator's own.") | print("own `strat_estimate`. **Point estimates only.** The 95% intervals are in the") |
| | print("estimator's own output and nowhere else: its bootstrap draws for several") |
| | print("verdicts from one generator, so re-running it here from a generator seeded") |
| | print("identically prints a different interval for the same estimate.") |
| for lab in LISTS: | for lab in LISTS: |
| L = S["lists"][lab] | L = S["lists"][lab] |
| print(f" {v:14} {pct(db):>14} {pct(da):>13} {100*(da-db):>+6.1f} " | print(f" {v:14} {pct(db):>14} {pct(da):>13} {100*(da-db):>+6.1f} " |
| f"{pct(pb):>12} {pct(pa):>11} {100*(pa-pb):>+6.1f}") | f"{pct(pb):>12} {pct(pa):>11} {100*(pa-pb):>+6.1f}") |
| rng = random.Random(ors.SEED) | # No interval is printed here. The bootstrap is stochastic and the |
| (dl, dh), (pl, ph) = ors.bootstrap(ea[lab][0], w, lambda x: x == "current", rng) | # estimator draws for several verdicts from ONE generator, so calling it |
| print(f" current, after, 95% CI: domain [{pct(dl)}, {pct(dh)}] " | # again from a fresh generator seeded identically produces a DIFFERENT |
| f"encounter [{pct(pl)}, {pct(ph)}]") | # interval for the same estimate -- 61.7-85.5 here against 61.6-85.3 |
| | # there. Two published intervals for one number is exactly the defect |
| | # this file exists to prevent, and a review pass found it. The intervals |
| | # are in owner_random_sample.py's own output and nowhere else. |
| |
| # ---- E2. the pairwise tests the content page publishes ------------------- | # ---- E2. the pairwise tests the content page publishes ------------------- |
| n = sum(len(x) for x in E[lab][0]) | n = sum(len(x) for x in E[lab][0]) |
| g = sum(1 for h in range(4) for _, v in E[lab][0][h] if v == "current") | g = sum(1 for h in range(4) for _, v in E[lab][0][h] if v == "current") |
| | gg = sum(1 for h in range(4) for _, v in E[lab][0][h] if v in GOOD) |
| raw[(lab, tag)] = (g, n) | raw[(lab, tag)] = (g, n) |
| | raw[(lab, tag, "gran")] = (gg, n) |
| print(f" {'List':14} {'current/eligible before':>24} {'after':>12} {'raw % before':>13} {'after':>7}") | print(f" {'List':14} {'current/eligible before':>24} {'after':>12} {'raw % before':>13} {'after':>7}") |
| for lab in LISTS: | for lab in LISTS: |
| f"{100*gb/nb:>12.1f}% {100*ga/na:>6.1f}%") | f"{100*gb/nb:>12.1f}% {100*ga/na:>6.1f}%") |
| print() | print() |
| print(f" {'Pair':34} {'p before':>9} {'p after':>9} survives 0.0167 after?") | print("**The result depends on the metric, and publishing only one of them was") |
| for a, b in (("Disconnect", "webXray"), ("Disconnect", "Tracker Radar"), | print("this script's most misleading omission.** Counting `granularity` as") |
| ("Tracker Radar", "webXray")): | print("acceptable -- which the content page says is right when the unit of analysis") |
| ga, na = raw[(a, "before")]; gb, nb = raw[(b, "before")] | print("is 'which company' rather than 'which legal entity' -- BOTH Disconnect") |
| pb = fisher_two_sided(ga, na - ga, gb, nb - gb) | print("comparisons clear Bonferroni, before and after, and the second pass overturns") |
| ga, na = raw[(a, "after")]; gb, nb = raw[(b, "after")] | print("nothing. On `current` only, Disconnect-versus-webXray stops clearing it.") |
| pa = fisher_two_sided(ga, na - ga, gb, nb - gb) | print("Report both or report neither.") |
| print(f" {a + ' vs ' + b:34} {pb:>9.3f} {pa:>9.3f} {'yes' if pa < 0.0167 else 'no'}") | for metric, key in (("current only", None), ("current+granularity", "gran")): |
| | print(f"\n metric: {metric}") |
| | print(f" {'Pair':34} {'p before':>9} {'p after':>9} survives 0.0167 before / after?") |
| | for a, b in (("Disconnect", "webXray"), ("Disconnect", "Tracker Radar"), |
| | ("Tracker Radar", "webXray")): |
| | ka = (a, "before") if key is None else (a, "before", key) |
| | kb = (b, "before") if key is None else (b, "before", key) |
| | ga, na = raw[ka]; gb, nb = raw[kb] |
| | pb = fisher_two_sided(ga, na - ga, gb, nb - gb) |
| | ka = (a, "after") if key is None else (a, "after", key) |
| | kb = (b, "after") if key is None else (b, "after", key) |
| | ga2, na2 = raw[ka]; gb2, nb2 = raw[kb] |
| | pa = fisher_two_sided(ga2, na2 - ga2, gb2, nb2 - gb2) |
| | print(f" {a + ' vs ' + b:34} {pb:>9.3f} {pa:>9.3f} " |
| | f"{'yes' if pb < 0.0167 else 'no':>3} / {'yes' if pa < 0.0167 else 'no'}" |
| | f" ({ga2}/{na2} vs {gb2}/{nb2})") |
| |
| # ---- E3. sensitivity: drop the weakest-sourced recovered rows ------------ | # ---- E3. sensitivity: drop the weakest-sourced recovered rows ------------ |
| print() | print() |
| print(" dropped: " + (", ".join(weak) or "<none>")) | print(" dropped: " + (", ".join(weak) or "<none>")) |
| A2 = {d: (dict(r, resolution="unresolved") if d in weak else r) for d, r in A.items()} | |
| e2 = eligible(S, A2) | NEWKINDS = {"domain-register", "archived-legal-doc", "tls-san"} |
| print(f"\n {'List':14} {'eligible':>9} {'current (domain)':>17} {'vs. full pass':>14}") | allnew = sorted(d for d, r in A.items() |
| for lab in LISTS: | if r.get("source_kind") in NEWKINDS and r["resolution"] == "resolved") |
| w = list(zip(S["lists"][lab]["strata_N"], S["lists"][lab]["strata_mass"])) | |
| d2, _ = ors.strat_estimate(e2[lab][0], w, lambda x: x == "current") | def run(dropped, label, note): |
| da, _ = ors.strat_estimate(ea[lab][0], w, lambda x: x == "current") | A2 = {d: (dict(r, resolution="unresolved") if d in dropped else r) |
| print(f" {lab:14} {sum(len(x) for x in e2[lab][0]):>9} {pct(d2):>17} " | for d, r in A.items()} |
| f"{100*(d2-da):>+13.1f}") | e2 = eligible(S, A2) |
| | print(f"\n {label} ({len(dropped)} row(s): {', '.join(dropped) or 'none'})") |
| | print(f" {note}") |
| | print(f" {'List':14} {'eligible':>9} {'current raw':>13} {'current (domain)':>17} " |
| | f"{'vs. full pass':>14}") |
| | rawc = {} |
| | for lab in LISTS: |
| | w = list(zip(S["lists"][lab]["strata_N"], S["lists"][lab]["strata_mass"])) |
| | d2, _ = ors.strat_estimate(e2[lab][0], w, lambda x: x == "current") |
| | da, _ = ors.strat_estimate(ea[lab][0], w, lambda x: x == "current") |
| | n = sum(len(x) for x in e2[lab][0]) |
| | g = sum(1 for h in range(4) for _, v in e2[lab][0][h] if v == "current") |
| | rawc[lab] = (g, n) |
| | print(f" {lab:14} {n:>9} {f'{g}/{n}':>13} {pct(d2):>17} " |
| | f"{100*(d2-da):>+13.1f}") |
| | for a, b in (("Disconnect", "webXray"), ("Disconnect", "Tracker Radar")): |
| | ga, na = rawc[a]; gb, nb = rawc[b] |
| | pv = fisher_two_sided(ga, na - ga, gb, nb - gb) |
| | print(f" {a} vs {b}, current only: p = {pv:.3f} " |
| | f"({'survives' if pv < 0.0167 else 'does not survive'} 0.0167)") |
| | |
| | run(weak, "1. uncorroborated registrant organisation only", |
| | "the narrowest reading: a row whose ONLY evidence is a registrant name nobody validated.") |
| | run(allnew, "2. EVERY row resting on a source kind the first pass did not accept", |
| | "the same-bar comparison: what the rates would be if the second pass had used\n" |
| | " the first pass's bar exactly and settled only what that bar could reach.") |
| | run(["i.ua"], "3. the i.ua rescore, under the reading this run rejected", |
| | "if 'I.UA' is held to name no company at all, the row is an error rather than\n" |
| | " current, which is the reading the hand correction argued against. Dropping the\n" |
| | " row entirely is the conservative stand-in, since scoring it `error` would lower\n" |
| | " Disconnect further than dropping it does.") |
| |
| # ---- F. what is still unresolved ---------------------------------------- | # ---- F. what is still unresolved ---------------------------------------- |
| ap.add_argument("--page", required=True) | ap.add_argument("--page", required=True) |
| ap.add_argument("--output", required=True) | ap.add_argument("--output", required=True) |
| | ap.add_argument("--corrections", default=None, |
| | help="out/owner_tail_corrections-output.txt, for the " |
| | "hand-changed verdict count the report cannot produce") |
| args = ap.parse_args() | args = ap.parse_args() |
| page = open(args.page, encoding="utf8").read() | page = open(args.page, encoding="utf8").read() |
| |
| # ---- B. the recovered-versus-existing test ------------------------------- | # ---- B. the recovered-versus-existing test ------------------------------- |
| secB = re.findall(r"^(webXray|Tracker Radar|Disconnect)\s+(\d+)/(\d+)\s+(\d+)/(\d+)\s+([\d.]+)\s*$", | # Section B prints ONE BLOCK PER METRIC since 2026-09-11 (a review pass found |
| out, re.M) | # that the script tested current+granularity here and current-only in E2 -- |
| if len(secB) != 3: | # each the choice that made its own claim mildest -- and said neither). |
| sys.exit("FATAL: section B of the report did not parse") | # Parse per metric and check both, or a metric switch could go unnoticed |
| for lab, og, on, rg, rn, p in secB: | # exactly as it did the first time. |
| need(f"{lab} recovered good/n", f"{rg} of {rn}") | blocks = {} |
| need(f"{lab} already-eligible good/n", f"{og} of {on}") | for mb in re.finditer(r"^ metric: (.+)$((?:\n(?! metric:).*)+)", out, re.M): |
| # the page rounds p to two decimals | rows = re.findall(r"^ *(webXray|Tracker Radar|Disconnect)\s+(\d+)/(\d+)\s+" |
| need(f"{lab} Fisher p (recovered vs existing)", f"{float(p):.2f}") | r"(\d+)/(\d+)\s+([\d.]+)\s*$", mb.group(2), re.M) |
| | if len(rows) == 3 and mb.group(1) not in blocks: |
| | blocks[mb.group(1).strip()] = rows |
| | if len(blocks) != 2: |
| | sys.exit(f"FATAL: section B of the report did not parse: " |
| | f"{len(blocks)} metric block(s), expected 2 ({sorted(blocks)})") |
| | |
| | # The page publishes the current+granularity counts verbatim ("11 of 12 |
| | # against 41 of 42"), so those are checked as literal substrings. |
| | gran = [k for k in blocks if "gran" in k] |
| | if len(gran) != 1: |
| | sys.exit(f"FATAL: no current+granularity block in section B: {sorted(blocks)}") |
| | for lab, og, on, rg, rn, pv in blocks[gran[0]]: |
| | need(f"{lab} recovered good/n (current+gran)", f"{rg} of {rn}") |
| | need(f"{lab} already-eligible good/n (current+gran)", f"{og} of {on}") |
| | need(f"{lab} Fisher p (recovered vs existing, current+gran)", f"{float(pv):.2f}") |
| | |
| | # The current-only Disconnect figures are quoted in a possessive phrasing |
| | # ("9 of Disconnect's 12 ... against 39 of its previous 42"), so they are |
| | # checked by regex rather than as a literal. Both must be on the page: the |
| | # whole point of the fix is that the page names which metric it is using. |
| | conly = [k for k in blocks if "gran" not in k] |
| | if len(conly) != 1: |
| | sys.exit(f"FATAL: no current-only block in section B: {sorted(blocks)}") |
| | for lab, og, on, rg, rn, pv in blocks[conly[0]]: |
| | if lab != "Disconnect": |
| | continue |
| | checked += 2 |
| | if not re.search(rf"\b{rg}\b[^.]{{0,40}}\b{rn}\b", page): |
| | fails.append(f"{lab} recovered good/n (current only): " |
| | f"{rg} .. {rn} is not on the page") |
| | if not re.search(rf"\b{og}\b[^.]{{0,40}}\b{on}\b", page): |
| | fails.append(f"{lab} already-eligible good/n (current only): " |
| | f"{og} .. {on} is not on the page") |
| |
| # ---- D. the worst case, both conventions --------------------------------- | # ---- D. the worst case, both conventions --------------------------------- |
| |
| # ---- E2. the pairwise comparisons ---------------------------------------- | # ---- E2. the pairwise comparisons ---------------------------------------- |
| | # Sliced out of the report by heading rather than matched across the whole |
| | # file: section B prints rows in the same shape, and a regex over the whole |
| | # output silently took whichever came last. |
| | mE2 = re.search(r"^--- E2\..*?$(.*?)^--- E3\.", out, re.M | re.S) |
| | if not mE2: |
| | sys.exit("FATAL: section E2 of the report did not parse: no E2..E3 slice") |
| | e2 = mE2.group(1) |
| secE2 = re.findall(r"^ (Disconnect vs webXray|Disconnect vs Tracker Radar|" | secE2 = re.findall(r"^ (Disconnect vs webXray|Disconnect vs Tracker Radar|" |
| r"Tracker Radar vs webXray)\s+([\d.]+)\s+([\d.]+)\s+(yes|no)\s*$", | r"Tracker Radar vs webXray)\s+([\d.]+)\s+([\d.]+)\s+" |
| out, re.M) | r"(yes|no) / (yes|no)\s", e2, re.M) |
| if len(secE2) != 3: | # Two metric blocks since 2026-09-11, three pairs each. |
| sys.exit("FATAL: section E2 of the report did not parse") | if len(secE2) != 6: |
| | sys.exit(f"FATAL: section E2 of the report did not parse: " |
| | f"{len(secE2)} pair row(s), expected 6") |
| def need_p(what, value): | def need_p(what, value): |
| """A p-value may appear at three or two decimals: the page writes 0.031 | """A p-value may appear at three or two decimals: the page writes 0.031 |
| fails.append(f"{what}: p {value} is not on the page in any of {sorted(forms)}") | fails.append(f"{what}: p {value} is not on the page in any of {sorted(forms)}") |
| |
| for pair, pb, pa, surv in secE2: | for pair, pb, pa, sb, sa in secE2: |
| need_p(f"{pair}: p after", pa) | need_p(f"{pair}: p after", pa) |
| need_p(f"{pair}: p before", pb) | need_p(f"{pair}: p before", pb) |
| raw = dict((lab, (g, n)) for lab, g, n in re.findall( | raw = dict((lab, (g, n)) for lab, g, n in re.findall( |
| r"^ (webXray|Tracker Radar|Disconnect)\s+\d+/\d+\s+(\d+)/(\d+)\s", out, re.M)) | r"^ (webXray|Tracker Radar|Disconnect)\s+\d+/\d+\s+(\d+)/(\d+)\s", e2, re.M)) |
| if len(raw) != 3: | if len(raw) != 3: |
| sys.exit("FATAL: section E2's raw-count table did not parse") | sys.exit("FATAL: section E2's raw-count table did not parse") |
| need(f"{lab} current/eligible after", f"{g} of {n}") | need(f"{lab} current/eligible after", f"{g} of {n}") |
| |
| # ---- E3. the sensitivity run --------------------------------------------- | # ---- E3. the sensitivity runs -------------------------------------------- |
| m = re.search(r"^ dropped: (.+)$", out, re.M) | # THREE runs since 2026-09-11, not one. A review pass found that publishing |
| | # only the narrowest of them (2 rows of the 11 that rest on the new bar) and |
| | # calling it "moves nothing" left the same-bar comparison -- the one that |
| | # answers the objection -- unpublished, and left the i.ua alternative |
| | # reading promised in the correction log but never computed. All three are |
| | # checked here so dropping one back out of the page fails the guard. |
| | mE3 = re.search(r"^--- E3\..*?$(.*?)^--- F\.", out, re.M | re.S) |
| | if not mE3: |
| | sys.exit("FATAL: section E3 of the report did not parse: no E3..F slice") |
| | e3 = mE3.group(1) |
| | m = re.search(r"^ dropped: (.+)$", e3, re.M) |
| if not m: | if not m: |
| sys.exit("FATAL: section E3 of the report did not parse") | sys.exit("FATAL: section E3 of the report did not parse: no 'dropped:' line") |
| for d in m.group(1).split(", "): | for d in m.group(1).split(", "): |
| need(f"sensitivity dropped {d}", f"''{d}''") | need(f"sensitivity dropped {d}", f"''{d}''") |
| secE3 = re.findall(r"^ (webXray|Tracker Radar|Disconnect)\s+\d+\s+[\d.]+%\s+([+-][\d.]+)\s*$", | |
| out, re.M) | runs = re.split(r"^ (\d)\. ", e3, flags=re.M)[1:] |
| if len(secE3) != 3: | runs = [(runs[i], runs[i + 1]) for i in range(0, len(runs), 2)] |
| sys.exit("FATAL: section E3's delta table did not parse") | if len(runs) != 3: |
| need("sensitivity deltas", | sys.exit(f"FATAL: section E3 has {len(runs)} sensitivity run(s), expected 3") |
| ", ".join(f"{lab} {d.replace('-', '−')}" for lab, d in secE3)) | rates = {} |
| | for num, body in runs: |
| | rows = re.findall(r"^ (webXray|Tracker Radar|Disconnect)\s+\d+\s+\d+/\d+\s+" |
| | r"([\d.]+)%\s+([+-][\d.]+)\s*$", body, re.M) |
| | if len(rows) != 3: |
| | sys.exit(f"FATAL: E3 run {num}'s delta table did not parse " |
| | f"({len(rows)} rows)") |
| | rates[num] = rows |
| | |
| | # run 1: the page quotes the deltas, not the rates. |
| | MINUS = "\u2212" # the page uses U+2212, the script's output ASCII '-' |
| | need("sensitivity run 1 deltas", |
| | ", ".join(lab + " " + d.replace("-", MINUS) for lab, _, d in rates["1"])) |
| | # run 2: the same-bar comparison. The page quotes rate AND delta per list. |
| | for lab, pct, d in rates["2"]: |
| | need(f"same-bar run: {lab} rate", f"{pct}%") |
| | need(f"same-bar run: {lab} delta", "(" + d.replace("-", MINUS) + ")") |
| | # run 3: the i.ua alternative reading. Only Disconnect moves. |
| | d3 = dict((lab, pct) for lab, pct, _ in rates["3"]) |
| | need("i.ua alternative reading: Disconnect rate", f"{d3['Disconnect']}%") |
| |
| # ---- F. the remaining residue, every domain named ------------------------- | # ---- F. the remaining residue, every domain named ------------------------- |
| for d in residue: | for d in residue: |
| need(f"residue domain {d}", f"''{d}''") | need(f"residue domain {d}", f"''{d}''") |
| | |
| | # ---- G. SUPERSEDED figures must be ABSENT -------------------------------- |
| | # Every check above asserts a correct figure is PRESENT. That is only half |
| | # of what a page can get wrong, and on 2026-09-11 it was the wrong half: |
| | # the page's headline recommendation box carried 94.4% and 73.7% -- the |
| | # pre-second-pass rates -- while both guards exited 0, because the correct |
| | # 89.2% and 73.9% were present two sections away. Three review passes each |
| | # found it independently; neither guard could. |
| | # |
| | # A superseded value may legitimately appear in a //before// column, a |
| | # historical reviewer log, a published diff or a mutation harness's own |
| | # output, so absence is asserted **per line**: the line must either name the |
| | # new value too, or be marked as historical. That rule is narrow enough to |
| | # bite and wide enough not to forbid an honest before/after table. |
| | # ``^[<>] `` is a unified-diff line and nothing else. An earlier version |
| | # wrote ``^\s*[<>|]``, which exempted **every table row on the page** -- and |
| | # the before/after table is a table, so two of the three mutations that |
| | # reintroduce a superseded rate walked straight through the check that |
| | # exists to catch exactly them. Caught by the mutation harness, which is the |
| | # only reason this comment is here rather than a second silent guard. |
| | HIST = re.compile(r"//was |before the second pass|\bwas \d|2026-09-05 pass|" |
| | r"^[<>] |mutate |^caught ", re.I) |
| | sup = {} |
| | for lab, (g, n) in raw.items(): |
| | m2 = re.search(r"^ " + re.escape(lab) + r"\s+(\d+)/(\d+)\s", out, re.M) |
| | sup[f"{m2.group(1)} of {m2.group(2)}"] = f"{g} of {n}" |
| | for lab, b_, a_, _, _, _ in secD: |
| | sup[f"{b_}%"] = f"{a_}%" |
| | # Section E's per-list rate table is where the HEADLINE rates move, and the |
| | # first version of this check did not read it -- so it did not bite on the |
| | # very page state that motivated it (94.4% in the recommendation box). The |
| | # control for this check is therefore: run it against the pre-fix export and |
| | # watch it fail. If it passes there, it is asserting nothing. |
| | # Only the ``current`` and ``current+gran.`` rows. The granularity, stale and |
| | # error sub-rates are single-digit percentages -- 0.0%, 2.8%, 4.4%, 4.5% -- |
| | # and they collide with unrelated figures all over a 66 kB page: sweeping |
| | # them produced five false positives on a correct page and would have |
| | # trained the next run to ignore this check. The headline rate is what a |
| | # recommendation box quotes, and it is what went stale. |
| | for verdict, db, da, eb_, ea_ in re.findall( |
| | r"^ (current|current\+gran\.)\s+([\d.]+)%\s+([\d.]+)%\s+[+-][\d.]+\s+" |
| | r"([\d.]+)%\s+([\d.]+)%\s+[+-][\d.]+\s*$", out, re.M): |
| | for b_, a_ in ((db, da), (eb_, ea_)): |
| | if b_ != a_: |
| | sup[f"{b_}%"] = f"{a_}%" |
| | # Figures the REPORT does not produce, but the page carries and this sitting |
| | # moved. Section G can only sweep what the report prints, and the total |
| | # number of hand-changed verdicts is not one of those: it is the first |
| | # sitting's 29 plus whatever this one re-scored. It went stale in exactly |
| | # the way this check exists to stop -- the page carried it TWICE, the round |
| | # four patch fixed one copy, and the site-wide sweep found the other. So it |
| | # is derived here from the corrections script's own output rather than |
| | # typed, and both directions are asserted. |
| | if args.corrections: |
| | ctext = open(args.corrections, encoding="utf8").read() |
| | mres = re.search(r"^actions: .*rescore=(\d+)", ctext, re.M) |
| | if not mres: |
| | sys.exit("FATAL: could not read the rescore count from " |
| | f"{args.corrections}") |
| | prior = 29 # the 2026-09-05 sitting's own hand changes |
| | total = prior + int(mres.group(1)) |
| | need("hand-changed verdicts, total", f"{total} of the 525") |
| | sup[f"{prior} of the 525"] = f"{total} of the 525" |
| | |
| | # **p-values are deliberately NOT swept for absence.** They are written at |
| | # two or three decimals interchangeably, and a two-decimal form is a prefix |
| | # of other numbers on the page: "0.025" rounds to "0.03", which matches |
| | # inside "0.031", and "0.014" rounds to "0.01", which matches inside the |
| | # Bonferroni threshold "0.0167". Every one of those is a false positive on a |
| | # correct page, and a guard that cries wolf is a guard the next run turns |
| | # off. The p-values are covered by the presence checks above and by the |
| | # page's own explicit before/after column. |
| | # Values collected above are `str`; the p-value entries are `set`. Normalise |
| | # so the loop below has one shape. |
| | sup = {k: (v if isinstance(v, set) else {v}) for k, v in sup.items()} |
| | for old_s, new_forms in sup.items(): |
| | # A digit or a dot immediately before the match means this is the tail of |
| | # a longer number -- "2.4%" inside "92.4%" -- and not the figure at all. |
| | pat = re.compile(r"(?<![\d.])" + re.escape(old_s)) |
| | for line in page.splitlines(): |
| | if pat.search(line) and not any(f in line for f in new_forms) \ |
| | and not HIST.search(line): |
| | checked += 1 |
| | fails.append(f"superseded figure {old_s!r} on a line that does not also " |
| | f"carry any of {sorted(new_forms)} and is not marked " |
| | f"historical: {line.strip()[:120]!r}") |
| | checked += len(sup) |
| |
| print(f"checked {checked} figures from {args.output} against {args.page}") | print(f"checked {checked} figures from {args.output} against {args.page}") |
| </file> | </file> |
| |
| ===== H. scripts/check_tail_figures_mutations.sh — proving that guard bites ===== | ===== H. scripts/check_tail_merge_mutations.sh — proving the merge gate bites ===== |
| | |
| | Eleven corruptions of one adjudication batch, one gate each: a verdict outside the vocabulary, the retired ''self-named'' verdict returning, an ''absent'' revised in either direction, a ''resolved'' row with no source or no quote, a bogus ''source_kind'', an ''unresolved'' row still carrying a verdict, a dropped row, a duplicate, and a row for a domain outside the residue. The gate exiting 0 on the real run means nothing until it has been seen to exit non-zero. |
| | |
| | <file bash check_tail_merge_mutations.sh> |
| | #!/bin/bash |
| | # Prove that owner_tail_merge.py's gates actually reject a corrupt second pass. |
| | # |
| | # The gate is the only thing standing between a sub-agent's JSON and the |
| | # published rates, and it exited 0 on the real run -- which tells you nothing |
| | # unless it can be seen to exit non-zero. Each mutation below corrupts one |
| | # adjudication file in one way the gate claims to catch, and asserts it fails. |
| | # |
| | # bash scripts/check_tail_merge_mutations.sh |
| | set -u |
| | TMP=$(mktemp -d) |
| | pass=0; fail=0 |
| | |
| | run() { # run the gate over a mutated copy of out/tail_adj |
| | python3 scripts/owner_tail_merge.py --sample out/owner_sample.json \ |
| | --base out/adj_rows_scored.json --tail out/tail40.json \ |
| | --dir "$TMP/adj" --out "$TMP/rows.json" >"$TMP/log" 2>&1 |
| | } |
| | |
| | mutate() { |
| | desc="$1"; jqprog="$2" |
| | rm -rf "$TMP/adj"; cp -r out/tail_adj "$TMP/adj" |
| | jq "$jqprog" out/tail_adj/result1.json > "$TMP/adj/result1.json" || { echo "BROKEN $desc (jq failed)"; fail=$((fail+1)); return; } |
| | if cmp -s out/tail_adj/result1.json "$TMP/adj/result1.json"; then |
| | echo "BROKEN $desc -- the jq changed nothing, so this asserts nothing" |
| | fail=$((fail+1)); return |
| | fi |
| | if run; then |
| | echo "SURVIVED $desc"; fail=$((fail+1)) |
| | else |
| | echo "caught $desc [$(grep -m1 -oE ' [a-z0-9.-]+: .*' "$TMP/log" | head -c 110)]" |
| | pass=$((pass+1)) |
| | fi |
| | } |
| | |
| | mutate "a verdict outside the vocabulary" '[.[] | if .domain=="wishabi.com" then .webXray="probably" else . end]' |
| | mutate "the retired self-named verdict returns" '[.[] | if .domain=="wishabi.com" then .disconnect="self-named" else . end]' |
| | mutate "an absent verdict revised to current" '[.[] | if .domain=="htplayground.com" then .webXray="current" else . end]' |
| | mutate "a real verdict downgraded to absent" '[.[] | if .domain=="wishabi.com" then .webXray="absent" else . end]' |
| | mutate "resolved with no source" '[.[] | if .domain=="wishabi.com" then .source=null else . end]' |
| | mutate "resolved with no quote" '[.[] | if .domain=="wishabi.com" then .quote="" else . end]' |
| | mutate "a source_kind outside the vocabulary" '[.[] | if .domain=="wishabi.com" then .source_kind="vibes" else . end]' |
| | mutate "unresolved but still carrying a verdict" '[.[] | if .domain=="1rx.io" then .tracker_radar="current" else . end]' |
| | mutate "a row dropped from the batch" '[.[] | select(.domain!="wishabi.com")]' |
| | mutate "a duplicate row" '. + [.[0]]' |
| | mutate "a row for a domain not in the residue" '[.[] | if .domain=="wishabi.com" then .domain="gstatic.com" else . end]' |
| | |
| | echo |
| | echo "mutations caught: $pass; survived or broken: $fail" |
| | rm -rf "$TMP" |
| | [ "$fail" -eq 0 ] |
| | </file> |
| | |
| | ===== I. scripts/check_tail_figures_mutations.sh — proving that guard bites ===== |
| |
| Fifteen corruptions, one figure each, twelve changes and three deletions. One survived the first run and that is why the residue check above is a contiguous run. | Fifteen corruptions, one figure each, twelve changes and three deletions. One survived the first run and that is why the residue check above is a contiguous run. |
| # are DELETIONS rather than changes, because the failure mode that matters most | # are DELETIONS rather than changes, because the failure mode that matters most |
| # is a guard passing because the string it was looking for is no longer there. | # is a guard passing because the string it was looking for is no longer there. |
| | # |
| | # LIMIT, stated because it was observed here: check_tail_figures.py asks whether |
| | # a figure is PRESENT. A page that carries a figure twice and gets one copy |
| | # wrong still passes it, so two mutations below need the sed `g` flag to assert |
| | # anything at all -- without it they corrupt the first copy and the guard reads |
| | # the second and is content. That hole is real and is covered by a different |
| | # tool: scripts/sweep_moved_figures.py scans every page in the wiki for every |
| | # superseded value, which is how the four surviving copies of the pre-second- |
| | # pass rates were found on 2026-09-11. Do not expect this harness to find them. |
| # | # |
| # bash scripts/check_tail_figures_mutations.sh | # bash scripts/check_tail_figures_mutations.sh |
| set -u | set -u |
| PAGE=pages/own_tail.txt | PAGE=${1:-pages/own_tail7.txt} |
| OUT=out/report_tail_pass-output.txt | OUT=out/report_tail_pass-output.txt |
| | CORR=out/owner_tail_corrections-output.txt |
| TMP=$(mktemp -d) | TMP=$(mktemp -d) |
| pass=0; fail=0 | pass=0; fail=0 |
| fail=$((fail+1)); return | fail=$((fail+1)); return |
| fi | fi |
| if python3 scripts/check_tail_figures.py --page "$TMP/p.txt" --output "$OUT" >/dev/null 2>&1; then | if python3 scripts/check_tail_figures.py --page "$TMP/p.txt" --output "$OUT" \ |
| | --corrections "$CORR" >/dev/null 2>&1; then |
| echo "SURVIVED $desc" | echo "SURVIVED $desc" |
| fail=$((fail+1)) | fail=$((fail+1)) |
| # --- changed figures --------------------------------------------------------- | # --- changed figures --------------------------------------------------------- |
| mutate "eligible after 56 / 56 / 54 -> 56 / 56 / 53" 's|\*\*56 / 56 / 54\*\*|**56 / 56 / 53**|' | mutate "eligible after 56 / 56 / 54 -> 56 / 56 / 53" 's|\*\*56 / 56 / 54\*\*|**56 / 56 / 53**|' |
| mutate "Disconnect recovered 11 of 12 -> 12 of 12" 's|11 of 12|12 of 12|' | mutate "Disconnect recovered 11 of 12 -> 12 of 12" 's|11 of 12|12 of 12|g' |
| mutate "webXray recovered-vs-existing p 0.64 -> 0.04" 's|0\.64|0.04|' | mutate "webXray recovered-vs-existing p 0.64 -> 0.04" 's|0\.64|0.04|' |
| mutate "Disconnect vs webXray p 0.031 -> 0.013" 's|0\.031|0.013|' | mutate "Disconnect vs webXray p 0.031 -> 0.013" 's|0\.031|0.013|' |
| mutate "delete the residue domain ''mmstat.com''" "s|''mmstat.com''||" | mutate "delete the residue domain ''mmstat.com''" "s|''mmstat.com''||" |
| mutate "delete the sensitivity-dropped ''sa-as.com''" "s|''sa-as.com''||" | mutate "delete the sensitivity-dropped ''sa-as.com''" "s|''sa-as.com''||" |
| mutate "delete the error domain ''km0trk.com''" "s|''km0trk.com''||" | mutate "delete the error domain ''km0trk.com''" "s|''km0trk.com''||g" |
| | |
| | # --- the three sensitivity runs (E3), added 2026-09-11 after the generic pass - |
| | # The page used to publish only the narrowest of the three and call it "moves |
| | # nothing". These assert that the same-bar run and the i.ua alternative reading |
| | # cannot quietly leave the page again. |
| | mutate "same-bar Tracker Radar 72.1% -> 72.6%" 's|72\.1%|72.6%|' |
| | mutate "same-bar Disconnect 89.0% -> 89.5%" 's|89\.0%|89.5%|' |
| | mutate "same-bar Tracker Radar delta (−1.8) -> (−1.9)" 's|(−1\.8)|(−1.9)|' |
| | mutate "delete the same-bar Disconnect delta (−0.2)" 's|(−0\.2)||' |
| | mutate "i.ua alternative reading 88.6% -> 87.6%" 's|88\.6%|87.6%|' |
| | |
| | # --- superseded figures reintroduced (section G of the guard) ----------------- |
| | # These are the mutations the guard could NOT catch before 2026-09-11: three |
| | # review passes each reported, independently, that the page's recommendation box |
| | # carried the pre-second-pass rates while both guards exited 0. |
| | mutate "reintroduce the superseded Disconnect rate 89.2% -> 94.4%" 's|\*\*89\.2%\*\*|94.4%|' |
| | mutate "reintroduce the superseded Tracker Radar rate 73.9% -> 73.7%" 's|73\.9% | 73.7% |' |
| | mutate "reintroduce the superseded webXray encounter rate 93.8% -> 93.9%" 's|\*\*93\.8%\*\*|93.9%|' |
| | # This one is a single copy of a figure the page carries TWICE. It is the drift |
| | # the site-wide sweep found after round four had fixed the other copy, and the |
| | # guard could not see it because the report script does not print the quantity. |
| | mutate "revert ONE copy of the hand-changed count 30 -> 29 of the 525" 's|30 of the 525|29 of the 525|' |
| |
| echo | echo |
| </file> | </file> |
| |
| ===== I. Unedited output: scripts/owner_tail_probe.sh, all 40 residue domains ===== | ===== J. Unedited output: scripts/check_tail_merge_mutations.sh ===== |
| | |
| | Run as ''bash scripts/check_tail_merge_mutations.sh''. |
| | |
| | <code> |
| | caught a verdict outside the vocabulary [ wishabi.com: verdict webXray='probably' outside the vocabulary] |
| | caught the retired self-named verdict returns [ wishabi.com: disconnect='self-named' -- the verdict was retired on 2026-09-05] |
| | caught an absent verdict revised to current [ htplayground.com: webXray was absent in the base rows and this pass set 'current'; absent is a fact about th] |
| | caught a real verdict downgraded to absent [ wishabi.com: webXray set to absent but the base rows say 'unknown'] |
| | caught resolved with no source [ wishabi.com: resolution=resolved but source/quote empty] |
| | caught resolved with no quote [ wishabi.com: resolution=resolved but source/quote empty] |
| | caught a source_kind outside the vocabulary [ wishabi.com: source_kind='vibes' outside the vocabulary] |
| | caught unresolved but still carrying a verdict [ 1rx.io: resolution=unresolved but tracker_radar='current'; an unresolved row may only carry absent or unknow] |
| | caught a row dropped from the batch [] |
| | caught a duplicate row [] |
| | caught a row for a domain not in the residue [] |
| | |
| | mutations caught: 11; survived or broken: 0 |
| | </code> |
| | |
| | ===== K. Unedited output: scripts/owner_tail_probe.sh, all 40 residue domains ===== |
| |
| Run as ''bash scripts/owner_tail_probe.sh out/tail40.json out/tail_probe''. | Run as ''bash scripts/owner_tail_probe.sh out/tail40.json out/tail_probe''. |
| </code> | </code> |
| |
| ===== J. Unedited output: scripts/owner_tail_merge.py ===== | ===== L. Unedited output: scripts/owner_tail_merge.py ===== |
| |
| Run as ''python3 scripts/owner_tail_merge.py --sample out/owner_sample.json --base out/adj_rows_scored.json --tail out/tail40.json --dir out/tail_adj --out out/adj_rows_tail.json''. | Run as ''python3 scripts/owner_tail_merge.py --sample out/owner_sample.json --base out/adj_rows_scored.json --tail out/tail40.json --dir out/tail_adj --out out/adj_rows_tail.json''. |
| </code> | </code> |
| |
| ===== K. Unedited output: scripts/owner_verify_sources.py on the second pass's rows ===== | ===== M. Unedited output: scripts/owner_verify_sources.py on the second pass's rows ===== |
| |
| Run as ''python3 scripts/owner_verify_sources.py --rows out/tail_rows_only.json --cache out/tail_verify_cache --out out/tail_verify.json''. | Run as ''python3 scripts/owner_verify_sources.py --rows out/tail_rows_only.json --cache out/tail_verify_cache --out out/tail_verify.json''. |
| </code> | </code> |
| |
| ===== L. Unedited output: scripts/owner_tail_corrections.py ===== | ===== N. Unedited output: scripts/owner_tail_corrections.py ===== |
| |
| Run as ''python3 scripts/owner_tail_corrections.py --rows out/adj_rows_tail.json --out out/adj_rows_tail_final.json''. | Run as ''python3 scripts/owner_tail_corrections.py --rows out/adj_rows_tail.json --out out/adj_rows_tail_final.json''. |
| </code> | </code> |
| |
| ===== M. Unedited output: scripts/owner_random_sample.py re-run after the second pass ===== | ===== O. Unedited output: scripts/owner_random_sample.py re-run after the second pass ===== |
| |
| Run as ''python3 scripts/owner_random_sample.py --sample out/owner_sample.json --rows out/adj_rows_tail_final.json --table --wiki''. | Run as ''python3 scripts/owner_random_sample.py --sample out/owner_sample.json --rows out/adj_rows_tail_final.json --table --wiki''. |
| </code> | </code> |
| |
| ===== N. The diff: the 2026-09-05 estimate against the 2026-09-11 one ===== | ===== P. The diff: the 2026-09-05 estimate against the 2026-09-11 one ===== |
| |
| Run as ''diff out/owner_random_sample-output.txt out/owner_random_sample-tail-output.txt''. | Run as ''diff out/owner_random_sample-output.txt out/owner_random_sample-tail-output.txt''. |
| |
| <code> | <code> |
| 17,19c17,19 | --- out/owner_random_sample-output.txt 2026-09-05 18:50:30.299249612 +0000 |
| < webXray 60 11 49 Q1 20/24 Q2 10/12 Q3 7/12 Q4 12/12 | +++ out/owner_random_sample-tail-output.txt 2026-09-11 19:07:23.934060685 +0000 |
| < Tracker Radar 60 13 47 Q1 17/24 Q2 9/12 Q3 11/12 Q4 10/12 | @@ -14,9 +14,9 @@ |
| < Disconnect 60 18 42 Q1 20/24 Q2 7/12 Q3 7/12 Q4 8/12 | the tail, so it is reported first and per quartile rather than buried. |
| --- | |
| > webXray 60 4 56 Q1 22/24 Q2 12/12 Q3 10/12 Q4 12/12 | List drawn unresolved eligible per quartile (eligible/drawn) |
| > Tracker Radar 60 4 56 Q1 22/24 Q2 10/12 Q3 12/12 Q4 12/12 | -webXray 60 11 49 Q1 20/24 Q2 10/12 Q3 7/12 Q4 12/12 |
| > Disconnect 60 6 54 Q1 22/24 Q2 11/12 Q3 9/12 Q4 12/12 | -Tracker Radar 60 13 47 Q1 17/24 Q2 9/12 Q3 11/12 Q4 10/12 |
| 29c29 | -Disconnect 60 18 42 Q1 20/24 Q2 7/12 Q3 7/12 Q4 8/12 |
| < webXray (covers 664 of 32,337 frame domains = 2.1%; n eligible = 49) | +webXray 60 4 56 Q1 22/24 Q2 12/12 Q3 10/12 Q4 12/12 |
| --- | +Tracker Radar 60 4 56 Q1 22/24 Q2 10/12 Q3 12/12 Q4 12/12 |
| > webXray (covers 664 of 32,337 frame domains = 2.1%; n eligible = 56) | +Disconnect 60 6 54 Q1 22/24 Q2 11/12 Q3 9/12 Q4 12/12 |
| 31,33c31,33 | |
| < current 69.9% [ 55.5%, 83.1%] 93.9% [ 80.9%, 98.8%] | --- B. Per-list verdict rates, with 95% bootstrap intervals ----------- |
| < granularity 10.2% [ 2.1%, 20.9%] 0.2% [ 0.0%, 0.8%] | |
| < stale 19.8% [ 9.2%, 32.3%] 5.9% [ 1.0%, 19.1%] | @@ -26,36 +26,34 @@ |
| --- | list covers equally; 'encounter-weighted' weights by Tracker Radar prevalence, |
| > current 69.2% [ 56.1%, 81.4%] 93.8% [ 80.9%, 98.6%] | i.e. how often a crawl actually meets the domain. |
| > granularity 8.8% [ 2.1%, 17.5%] 0.2% [ 0.0%, 0.7%] | |
| > stale 22.0% [ 11.4%, 33.8%] 5.9% [ 1.1%, 18.7%] | -webXray (covers 664 of 32,337 frame domains = 2.1%; n eligible = 49) |
| 36,38c36,38 | +webXray (covers 664 of 32,337 frame domains = 2.1%; n eligible = 56) |
| < The usable statement is the rule-of-three upper bound: with 49 eligible rows and 0 seen, the true rate is below 6.1% at 95% confidence. | verdict domain-level 95% CI encounter-wtd 95% CI |
| < current+gran. 80.2% [ 67.7%, 90.8%] 94.1% [ 81.4%, 99.0%] | - current 69.9% [ 55.5%, 83.1%] 93.9% [ 80.9%, 98.8%] |
| < stale+error 19.8% [ 9.0%, 32.6%] 5.9% [ 1.1%, 19.4%] | - granularity 10.2% [ 2.1%, 20.9%] 0.2% [ 0.0%, 0.8%] |
| --- | - stale 19.8% [ 9.2%, 32.3%] 5.9% [ 1.0%, 19.1%] |
| > The usable statement is the rule-of-three upper bound: with 56 eligible rows and 0 seen, the true rate is below 5.4% at 95% confidence. | + current 69.2% [ 56.1%, 81.4%] 93.8% [ 80.9%, 98.6%] |
| > current+gran. 78.0% [ 65.8%, 88.8%] 94.1% [ 81.4%, 98.8%] | + granularity 8.8% [ 2.1%, 17.5%] 0.2% [ 0.0%, 0.7%] |
| > stale+error 22.0% [ 11.0%, 33.8%] 5.9% [ 1.1%, 18.9%] | + stale 22.0% [ 11.4%, 33.8%] 5.9% [ 1.1%, 18.7%] |
| 40c40 | error 0.0% [ 0.0%, 0.0%] 0.0% [ 0.0%, 0.0%] |
| < Tracker Radar (covers 5,566 of 32,337 frame domains = 17.2%; n eligible = 47) | zero cells (error): the bootstrap interval on a cell with no observations is [0,0] and asserts nothing. |
| --- | - The usable statement is the rule-of-three upper bound: with 49 eligible rows and 0 seen, the true rate is below 6.1% at 95% confidence. |
| > Tracker Radar (covers 5,566 of 32,337 frame domains = 17.2%; n eligible = 56) | - current+gran. 80.2% [ 67.7%, 90.8%] 94.1% [ 81.4%, 99.0%] |
| 42,47c42,47 | - stale+error 19.8% [ 9.0%, 32.6%] 5.9% [ 1.1%, 19.4%] |
| < current 73.7% [ 60.8%, 85.8%] 68.6% [ 29.3%, 97.0%] | + The usable statement is the rule-of-three upper bound: with 56 eligible rows and 0 seen, the true rate is below 5.4% at 95% confidence. |
| < granularity 2.8% [ 0.0%, 8.3%] 0.1% [ 0.0%, 0.4%] | + current+gran. 78.0% [ 65.8%, 88.8%] 94.1% [ 81.4%, 98.8%] |
| < stale 22.1% [ 10.5%, 35.1%] 26.8% [ 0.4%, 64.6%] | + stale+error 22.0% [ 11.0%, 33.8%] 5.9% [ 1.1%, 18.9%] |
| < error 1.5% [ 0.0%, 4.4%] 4.5% [ 0.0%, 19.3%] | |
| < current+gran. 76.5% [ 63.5%, 88.0%] 68.7% [ 29.2%, 97.7%] | -Tracker Radar (covers 5,566 of 32,337 frame domains = 17.2%; n eligible = 47) |
| < stale+error 23.5% [ 11.7%, 36.1%] 31.3% [ 2.9%, 71.2%] | +Tracker Radar (covers 5,566 of 32,337 frame domains = 17.2%; n eligible = 56) |
| --- | verdict domain-level 95% CI encounter-wtd 95% CI |
| > current 73.9% [ 61.6%, 85.3%] 74.3% [ 41.1%, 97.8%] | - current 73.7% [ 60.8%, 85.8%] 68.6% [ 29.3%, 97.0%] |
| > granularity 2.5% [ 0.0%, 7.5%] 0.1% [ 0.0%, 0.3%] | - granularity 2.8% [ 0.0%, 8.3%] 0.1% [ 0.0%, 0.4%] |
| > stale 19.2% [ 9.1%, 30.5%] 21.6% [ 0.3%, 53.1%] | - stale 22.1% [ 10.5%, 35.1%] 26.8% [ 0.4%, 64.6%] |
| > error 4.4% [ 0.0%, 9.8%] 4.0% [ 0.0%, 15.7%] | - error 1.5% [ 0.0%, 4.4%] 4.5% [ 0.0%, 19.3%] |
| > current+gran. 76.4% [ 64.6%, 87.5%] 74.4% [ 41.6%, 97.8%] | - current+gran. 76.5% [ 63.5%, 88.0%] 68.7% [ 29.2%, 97.7%] |
| > stale+error 23.6% [ 12.5%, 35.6%] 25.6% [ 2.2%, 58.5%] | - stale+error 23.5% [ 11.7%, 36.1%] 31.3% [ 2.9%, 71.2%] |
| 49c49 | + current 73.9% [ 61.6%, 85.3%] 74.3% [ 41.1%, 97.8%] |
| < Disconnect (covers 2,264 of 32,337 frame domains = 7.0%; n eligible = 42) | + granularity 2.5% [ 0.0%, 7.5%] 0.1% [ 0.0%, 0.3%] |
| --- | + stale 19.2% [ 9.1%, 30.5%] 21.6% [ 0.3%, 53.1%] |
| > Disconnect (covers 2,264 of 32,337 frame domains = 7.0%; n eligible = 54) | + error 4.4% [ 0.0%, 9.8%] 4.0% [ 0.0%, 15.7%] |
| 51,58c51,56 | + current+gran. 76.4% [ 64.6%, 87.5%] 74.4% [ 41.6%, 97.8%] |
| < current 94.4% [ 86.9%,100.0%] 83.1% [ 53.7%,100.0%] | + stale+error 23.6% [ 12.5%, 35.6%] 25.6% [ 2.2%, 58.5%] |
| < granularity 4.4% [ 0.0%, 11.2%] 16.5% [ 0.0%, 45.8%] | |
| < stale 1.2% [ 0.0%, 3.8%] 0.4% [ 0.0%, 1.5%] | -Disconnect (covers 2,264 of 32,337 frame domains = 7.0%; n eligible = 42) |
| < error 0.0% [ 0.0%, 0.0%] 0.0% [ 0.0%, 0.0%] | +Disconnect (covers 2,264 of 32,337 frame domains = 7.0%; n eligible = 54) |
| < zero cells (error): the bootstrap interval on a cell with no observations is [0,0] and asserts nothing. | verdict domain-level 95% CI encounter-wtd 95% CI |
| < The usable statement is the rule-of-three upper bound: with 42 eligible rows and 0 seen, the true rate is below 7.1% at 95% confidence. | - current 94.4% [ 86.9%,100.0%] 83.1% [ 53.7%,100.0%] |
| < current+gran. 98.8% [ 96.2%,100.0%] 99.6% [ 98.5%,100.0%] | - granularity 4.4% [ 0.0%, 11.2%] 16.5% [ 0.0%, 45.8%] |
| < stale+error 1.2% [ 0.0%, 3.8%] 0.4% [ 0.0%, 1.5%] | - stale 1.2% [ 0.0%, 3.8%] 0.4% [ 0.0%, 1.5%] |
| --- | - error 0.0% [ 0.0%, 0.0%] 0.0% [ 0.0%, 0.0%] |
| > current 89.2% [ 80.5%, 96.8%] 82.8% [ 53.8%,100.0%] | - zero cells (error): the bootstrap interval on a cell with no observations is [0,0] and asserts nothing. |
| > granularity 7.6% [ 1.1%, 15.2%] 16.8% [ 0.0%, 45.6%] | - The usable statement is the rule-of-three upper bound: with 42 eligible rows and 0 seen, the true rate is below 7.1% at 95% confidence. |
| > stale 1.1% [ 0.0%, 3.4%] 0.4% [ 0.0%, 1.5%] | - current+gran. 98.8% [ 96.2%,100.0%] 99.6% [ 98.5%,100.0%] |
| > error 2.1% [ 0.0%, 6.2%] 0.0% [ 0.0%, 0.0%] | - stale+error 1.2% [ 0.0%, 3.8%] 0.4% [ 0.0%, 1.5%] |
| > current+gran. 96.8% [ 91.5%,100.0%] 99.6% [ 98.5%,100.0%] | + current 89.2% [ 80.5%, 96.8%] 82.8% [ 53.8%,100.0%] |
| > stale+error 3.2% [ 0.0%, 8.5%] 0.4% [ 0.0%, 1.5%] | + granularity 7.6% [ 1.1%, 15.2%] 16.8% [ 0.0%, 45.6%] |
| 76,84c74,82 | + stale 1.1% [ 0.0%, 3.4%] 0.4% [ 0.0%, 1.5%] |
| < webXray heaviest row gstatic.com verdict current carries 39.4% of the estimate | + error 2.1% [ 0.0%, 6.2%] 0.0% [ 0.0%, 0.0%] |
| < encounter-weighted 'current' 93.9% -> 54.4% if that one row were stale | + current+gran. 96.8% [ 91.5%,100.0%] 99.6% [ 98.5%,100.0%] |
| < top three rows together carry 53.1% | + stale+error 3.2% [ 0.0%, 8.5%] 0.4% [ 0.0%, 1.5%] |
| < Tracker Radar heaviest row omtrdc.net verdict current carries 24.9% of the estimate | |
| < encounter-weighted 'current' 68.6% -> 43.7% if that one row were stale | --- C. Coverage is not estimated: it is counted -------------------- |
| < top three rows together carry 64.9% | |
| < Disconnect heaviest row id5-sync.com verdict current carries 16.7% of the estimate | @@ -73,69 +71,41 @@ |
| < encounter-weighted 'current' 83.1% -> 66.4% if that one row were stale | verdict is flipped to stale. This is a sensitivity check, not a result: the |
| < top three rows together carry 46.0% | flipped figure is what the estimate WOULD be, not a claim that it is. |
| --- | |
| > webXray heaviest row gstatic.com verdict current carries 38.8% of the estimate | - webXray heaviest row gstatic.com verdict current carries 39.4% of the estimate |
| > encounter-weighted 'current' 93.8% -> 55.0% if that one row were stale | - encounter-weighted 'current' 93.9% -> 54.4% if that one row were stale |
| > top three rows together carry 52.2% | - top three rows together carry 53.1% |
| > Tracker Radar heaviest row omtrdc.net verdict current carries 20.0% of the estimate | - Tracker Radar heaviest row omtrdc.net verdict current carries 24.9% of the estimate |
| > encounter-weighted 'current' 74.3% -> 54.3% if that one row were stale | - encounter-weighted 'current' 68.6% -> 43.7% if that one row were stale |
| > top three rows together carry 52.4% | - top three rows together carry 64.9% |
| > Disconnect heaviest row id5-sync.com verdict current carries 16.6% of the estimate | - Disconnect heaviest row id5-sync.com verdict current carries 16.7% of the estimate |
| > encounter-weighted 'current' 82.8% -> 66.2% if that one row were stale | - encounter-weighted 'current' 83.1% -> 66.4% if that one row were stale |
| > top three rows together carry 45.6% | - top three rows together carry 46.0% |
| 91,93c89,91 | + webXray heaviest row gstatic.com verdict current carries 38.8% of the estimate |
| < webXray absent=96 current=12 granularity=1 stale=5 unknown=1 | + encounter-weighted 'current' 93.8% -> 55.0% if that one row were stale |
| < Tracker Radar absent=18 current=58 error=1 granularity=6 stale=16 unknown=16 | + top three rows together carry 52.2% |
| < Disconnect absent=49 current=44 granularity=8 unknown=14 | + Tracker Radar heaviest row omtrdc.net verdict current carries 20.0% of the estimate |
| --- | + encounter-weighted 'current' 74.3% -> 54.3% if that one row were stale |
| > webXray absent=96 current=13 granularity=1 stale=5 | + top three rows together carry 52.4% |
| > Tracker Radar absent=18 current=68 error=1 granularity=6 stale=18 unknown=4 | + Disconnect heaviest row id5-sync.com verdict current carries 16.6% of the estimate |
| > Disconnect absent=49 current=50 granularity=12 unknown=4 | + encounter-weighted 'current' 82.8% -> 66.2% if that one row were stale |
| 97,138c95,108 | + top three rows together carry 45.6% |
| < webXray Q1 adgrx.com unknown unresolved Strong secondary-source consensus (Netify, press coverage) that adgrx.com is S | |
| < webXray Q1 company-target.com unknown unresolved Domain has live DNS (CloudFront IP range) but TLS handshake fails and no page | --- D. Verdicts on domains drawn for another list (NOT in any estimate) --- |
| < webXray Q1 1rx.io unknown unresolved Nexxen's own privacy-policy pages (nexxen.com/services-privacy-policy/, nexxen | |
| < webXray Q1 agkn.com unknown unresolved Secondary reporting (not fetched as primary) traces a chain Aggregate Knowledg | Each drawn domain was adjudicated once and the verdict recorded for all three |
| < webXray Q2 yceml.net unknown unresolved Domain itself had no TLS handshake at probe time. Conversant LLC (webXray/Trac | lists. Only a list's own sample enters its estimate above; these are the rest, |
| < webXray Q2 blogblog.com unknown unresolved Domain 404s and only serves a DV *.blogger.com wildcard cert (Google Trust Ser | printed so the discarded observations are visible rather than silently dropped. |
| < webXray Q3 awltovhc.com unknown unresolved Long-recognized Conversant/ValueClick-style cookie-sync domain. RDAP shows an | - webXray absent=96 current=12 granularity=1 stale=5 unknown=1 |
| < webXray Q3 mmstat.com unknown unresolved Registrar is Alibaba Cloud Computing (Beijing) Co., Ltd. with ns3/4/5.aliyun.c | - Tracker Radar absent=18 current=58 error=1 granularity=6 stale=16 unknown=16 |
| < webXray Q3 lduhtrp.net unknown unresolved Same Conversant/ValueClick-style cookie-sync domain family as awltovhc.com, wi | - Disconnect absent=49 current=44 granularity=8 unknown=14 |
| < webXray Q3 collective-media.net unknown unresolved 'Collective, Inc.' (the ad network) split its ad-network division into 'Compas | + webXray absent=96 current=13 granularity=1 stale=5 |
| < webXray Q3 wishabi.com unknown unresolved wishabi.com's TLS handshake serves Flipp's certificate (CN=flipp.com) and Flip | + Tracker Radar absent=18 current=68 error=1 granularity=6 stale=18 unknown=4 |
| < Tracker Radar Q1 app-us1.com unknown unresolved app-us1.com serves no browsable page (CODE 000); ActiveCampaign's own fetched | + Disconnect absent=49 current=50 granularity=12 unknown=4 |
| < Tracker Radar Q1 travelpayouts.com unknown unresolved Homepage confirms the operating brand 'Travelpayouts' but not the legal entity | |
| < Tracker Radar Q1 acint.net unknown unresolved Site self-describes as 'Acint (Artificial Computation Intelligence)' with only | --- E. Residue: every row excluded from an estimate -------------------- |
| < Tracker Radar Q1 cdnbasket.net unknown unresolved bounceexchange.com 302-redirects to wunderkind.co today, and Wunderkind's fetc | |
| < Tracker Radar Q1 spot.im unknown unresolved OpenWeb's own privacy policy confirms 'Open Web Technologies Ltd.' is a real, | - webXray Q1 adgrx.com unknown unresolved Strong secondary-source consensus (Netify, press coverage) that adgrx.com is S |
| < Tracker Radar Q1 cnevids.com unknown unresolved cnevids.com itself no longer resolves, but its own player.cnevids.com subdomai | - webXray Q1 company-target.com unknown unresolved Domain has live DNS (CloudFront IP range) but TLS handshake fails and no page |
| < Tracker Radar Q1 marphezis.com unknown unresolved Tracker Radar's claim (Online Media Solutions Ltd. dba Brightcom) is plausible | - webXray Q1 1rx.io unknown unresolved Nexxen's own privacy-policy pages (nexxen.com/services-privacy-policy/, nexxen |
| < Tracker Radar Q2 hqseek.com unknown unresolved hqseek.com redirects to the adult site hqporn.pics; WebFetch of both was refus | - webXray Q1 agkn.com unknown unresolved Secondary reporting (not fetched as primary) traces a chain Aggregate Knowledg |
| < Tracker Radar Q2 contentabc.com unknown unresolved Disconnect's entity 'ContentABC' is a trivial capitalisation of the domain lab | - webXray Q2 yceml.net unknown unresolved Domain itself had no TLS handshake at probe time. Conversant LLC (webXray/Trac |
| < Tracker Radar Q2 trustpilot.net unknown unresolved Separately confirmed via https://corporate.trustpilot.com/legal/for-reviewers/ | - webXray Q2 blogblog.com unknown unresolved Domain 404s and only serves a DV *.blogger.com wildcard cert (Google Trust Ser |
| < Tracker Radar Q3 cratecamera.com unknown unresolved pgl.yoyo.org's blocklist tags cratecamera.com's category as 'admiral', consist | - webXray Q3 awltovhc.com unknown unresolved Long-recognized Conversant/ValueClick-style cookie-sync domain. RDAP shows an |
| < Tracker Radar Q4 cjponyparts.com unknown unresolved 'CJ Pony Parts, Inc.' is a real, independent Mustang-parts retailer (not a sel | - webXray Q3 mmstat.com unknown unresolved Registrar is Alibaba Cloud Computing (Beijing) Co., Ltd. with ns3/4/5.aliyun.c |
| < Tracker Radar Q4 cedscdn.it unknown unresolved Third-party WHOIS aggregators attribute registration to CED Digital & Servizi | - webXray Q3 lduhtrp.net unknown unresolved Same Conversant/ValueClick-style cookie-sync domain family as awltovhc.com, wi |
| < Disconnect Q1 govx.com unknown unresolved GOVX is a real San Diego company per secondary sources (BBB listing, LinkedIn) | - webXray Q3 collective-media.net unknown unresolved 'Collective, Inc.' (the ad network) split its ad-network division into 'Compas |
| < Disconnect Q1 cdnbasket.net unknown unresolved bounceexchange.com 302-redirects to wunderkind.co today, and Wunderkind's fetc | - webXray Q3 wishabi.com unknown unresolved wishabi.com's TLS handshake serves Flipp's certificate (CN=flipp.com) and Flip |
| < Disconnect Q1 agkn.com unknown unresolved Secondary reporting (not fetched as primary) traces a chain Aggregate Knowledg | - Tracker Radar Q1 app-us1.com unknown unresolved app-us1.com serves no browsable page (CODE 000); ActiveCampaign's own fetched |
| < Disconnect Q1 opti-digital.com unknown unresolved opti-digital.com fails to resolve at all (curl: could not resolve host); a liv | - Tracker Radar Q1 travelpayouts.com unknown unresolved Homepage confirms the operating brand 'Travelpayouts' but not the legal entity |
| < Disconnect Q2 at-o.net unknown unresolved at-o.net has no DNS/TLS response; AT Internet's (now Piano's) own data-protect | - Tracker Radar Q1 acint.net unknown unresolved Site self-describes as 'Acint (Artificial Computation Intelligence)' with only |
| < Disconnect Q2 mapixl.com unknown unresolved Searches for a link between Disconnect's claimed 'MarketingArchitects' (a Minn | - Tracker Radar Q1 cdnbasket.net unknown unresolved bounceexchange.com 302-redirects to wunderkind.co today, and Wunderkind's fetc |
| < Disconnect Q2 ksearchnet.com unknown unresolved Klevu's own SDK README (klevultd GitHub org, now branded Athos Commerce per kl | - Tracker Radar Q1 spot.im unknown unresolved OpenWeb's own privacy policy confirms 'Open Web Technologies Ltd.' is a real, |
| < Disconnect Q2 gssprt.jp unknown unresolved webXray, Tracker Radar and Disconnect all agree on Geniee, and third-party WHO | - Tracker Radar Q1 cnevids.com unknown unresolved cnevids.com itself no longer resolves, but its own player.cnevids.com subdomai |
| < Disconnect Q2 kameleoon.io unknown unresolved Kameleoon SAS's own legal notice confirms it operates kameleoon.com/app.kamele | - Tracker Radar Q1 marphezis.com unknown unresolved Tracker Radar's claim (Online Media Solutions Ltd. dba Brightcom) is plausible |
| < Disconnect Q3 stat-track.com unknown unresolved Disconnect's 'StackTrack' is a near-verbatim mangling of the domain label 'sta | - Tracker Radar Q2 hqseek.com unknown unresolved hqseek.com redirects to the adult site hqporn.pics; WebFetch of both was refus |
| < Disconnect Q3 sa-as.com unknown unresolved Domain does not resolve over TLS and no primary source (newsroom/filing/legal- | - Tracker Radar Q2 contentabc.com unknown unresolved Disconnect's entity 'ContentABC' is a trivial capitalisation of the domain lab |
| < Disconnect Q3 stripst.com unknown unresolved stripst.com returned HTTP 522 on every fetch attempt (both root and a guessed | - Tracker Radar Q2 trustpilot.net unknown unresolved Separately confirmed via https://corporate.trustpilot.com/legal/for-reviewers/ |
| < Disconnect Q3 htplayground.com unknown unresolved Disconnect's entry is literally the string 'htplayground.com', identical to th | - Tracker Radar Q3 cratecamera.com unknown unresolved pgl.yoyo.org's blocklist tags cratecamera.com's category as 'admiral', consist |
| < Disconnect Q3 pages02.net unknown unresolved Leads (non-primary) indicate pages0X.net domains are Silverpop/IBM Watson Mark | - Tracker Radar Q4 cjponyparts.com unknown unresolved 'CJ Pony Parts, Inc.' is a real, independent Mustang-parts retailer (not a sel |
| < Disconnect Q4 globo.com unknown unresolved Public secondary sources describe Globo Comunicação e Participações S.A. (form | - Tracker Radar Q4 cedscdn.it unknown unresolved Third-party WHOIS aggregators attribute registration to CED Digital & Servizi |
| < Disconnect Q4 offshoregeology.com unknown unresolved Disconnect's 'Admiral' claim could not be verified against a primary source; a | - Disconnect Q1 govx.com unknown unresolved GOVX is a real San Diego company per secondary sources (BBB listing, LinkedIn) |
| < Disconnect Q4 i.ua unknown unresolved RDAP lists registrant 'Digital Ventures LLC' and registrar 'Internet Invest LL | - Disconnect Q1 cdnbasket.net unknown unresolved bounceexchange.com 302-redirects to wunderkind.co today, and Wunderkind's fetc |
| < Disconnect Q4 km0trk.com unknown unresolved Disconnect's entry name is simply the domain itself (km0trk.com), which is sel | - Disconnect Q1 agkn.com unknown unresolved Secondary reporting (not fetched as primary) traces a chain Aggregate Knowledg |
| --- | - Disconnect Q1 opti-digital.com unknown unresolved opti-digital.com fails to resolve at all (curl: could not resolve host); a liv |
| > webXray Q1 1rx.io unknown unresolved register-by-number: no company/VAT number available anywhere for 'Blinkx'/'Rhy | - Disconnect Q2 at-o.net unknown unresolved at-o.net has no DNS/TLS response; AT Internet's (now Piano's) own data-protect |
| > webXray Q1 agkn.com unknown unresolved Register-by-number: SEC EDGAR full-text search for the exact phrase "agkn.com" | - Disconnect Q2 mapixl.com unknown unresolved Searches for a link between Disconnect's claimed 'MarketingArchitects' (a Minn |
| > webXray Q3 mmstat.com unknown unresolved Register-by-number: tried China's official ICP/beian registry at beian.miit.go | - Disconnect Q2 ksearchnet.com unknown unresolved Klevu's own SDK README (klevultd GitHub org, now branded Athos Commerce per kl |
| > webXray Q3 collective-media.net unknown unresolved Tried all three routes. Register-by-number: SEC EDGAR full-text search for '"C | - Disconnect Q2 gssprt.jp unknown unresolved webXray, Tracker Radar and Disconnect all agree on Geniee, and third-party WHO |
| > Tracker Radar Q1 cdnbasket.net unknown unresolved All three routes came back empty. Register-by-number: no identifying company/V | - Disconnect Q2 kameleoon.io unknown unresolved Kameleoon SAS's own legal notice confirms it operates kameleoon.com/app.kamele |
| > Tracker Radar Q1 marphezis.com unknown unresolved register-by-number: no identifier available for 'Online Media Solutions Ltd. d | - Disconnect Q3 stat-track.com unknown unresolved Disconnect's 'StackTrack' is a near-verbatim mangling of the domain label 'sta |
| > Tracker Radar Q2 hqseek.com unknown unresolved All three routes tried. register-by-number: no imprint or identifiable number | - Disconnect Q3 sa-as.com unknown unresolved Domain does not resolve over TLS and no primary source (newsroom/filing/legal- |
| > Tracker Radar Q2 contentabc.com unknown unresolved Register-by-number: no imprint, VAT, or company number for either candidate (' | - Disconnect Q3 stripst.com unknown unresolved stripst.com returned HTTP 522 on every fetch attempt (both root and a guessed |
| > Disconnect Q1 cdnbasket.net unknown unresolved All three routes came back empty. Register-by-number: no identifying company/V | - Disconnect Q3 htplayground.com unknown unresolved Disconnect's entry is literally the string 'htplayground.com', identical to th |
| > Disconnect Q1 agkn.com unknown unresolved Register-by-number: SEC EDGAR full-text search for the exact phrase "agkn.com" | - Disconnect Q3 pages02.net unknown unresolved Leads (non-primary) indicate pages0X.net domains are Silverpop/IBM Watson Mark |
| > Disconnect Q2 mapixl.com unknown unresolved All three routes tried and none produced an identifier for the actual operator | - Disconnect Q4 globo.com unknown unresolved Public secondary sources describe Globo Comunicação e Participações S.A. (form |
| > Disconnect Q3 stat-track.com unknown unresolved All three routes were tried. Register-by-number: Disconnect's 'StackTrack' is | - Disconnect Q4 offshoregeology.com unknown unresolved Disconnect's 'Admiral' claim could not be verified against a primary source; a |
| > Disconnect Q3 stripst.com unknown unresolved Register-by-number: no imprint or company number is available anywhere for str | - Disconnect Q4 i.ua unknown unresolved RDAP lists registrant 'Digital Ventures LLC' and registrar 'Internet Invest LL |
| > Disconnect Q3 htplayground.com unknown unresolved register-by-number: Disconnect's own entry is just the domain string, so there | - Disconnect Q4 km0trk.com unknown unresolved Disconnect's entry name is simply the domain itself (km0trk.com), which is sel |
| 150c120 | + webXray Q1 1rx.io unknown unresolved register-by-number: no company/VAT number available anywhere for 'Blinkx'/'Rhy |
| < | ''1rx.io'' | 0.05313 | //unresolved// | UNRESOLVED -- domain is dead (no TLS handshake, no HTTP response) and no primary source fetched this session ties 1rx.io itself to RhythmOne or Nexxen | unknown | unknown | unknown | //none// | | + webXray Q1 agkn.com unknown unresolved Register-by-number: SEC EDGAR full-text search for the exact phrase "agkn.com" |
| --- | + webXray Q3 mmstat.com unknown unresolved Register-by-number: tried China's official ICP/beian registry at beian.miit.go |
| > | ''1rx.io'' | 0.05313 | UNKNOWN | UNRESOLVED -- no primary source located | unknown | unknown | unknown | [[NONE -- no primary source reached|none]] | | + webXray Q3 collective-media.net unknown unresolved Tried all three routes. Register-by-number: SEC EDGAR full-text search for '"C |
| 158c128 | + Tracker Radar Q1 cdnbasket.net unknown unresolved All three routes came back empty. Register-by-number: no identifying company/V |
| < | ''agkn.com'' | 0.03580 | //unresolved// | UNRESOLVED -- domain is dead today (NXDOMAIN/no TLS handshake) though RDAP shows it remains registered (not expired, locked, registrar GoDaddy Corporate Domains) with no disclosed registrant org; could not reach any current primary source naming an operator | unknown | unknown | unknown | //none// | | + Tracker Radar Q1 marphezis.com unknown unresolved register-by-number: no identifier available for 'Online Media Solutions Ltd. d |
| --- | + Tracker Radar Q2 hqseek.com unknown unresolved All three routes tried. register-by-number: no imprint or identifiable number |
| > | ''agkn.com'' | 0.03580 | UNRESOLVED | UNRESOLVED -- registry WHOIS registrant is privacy-proxied (Brandsight/GoDaddy Corporate Domains) with no organization disclosed, and TransUnion's own pages that would plausibly name AdAdvisor/agkn.com are bot-walled | unknown | unknown | unknown | //none// | | + Tracker Radar Q2 contentabc.com unknown unresolved Register-by-number: no imprint, VAT, or company number for either candidate (' |
| 172c142 | + Disconnect Q1 cdnbasket.net unknown unresolved All three routes came back empty. Register-by-number: no identifying company/V |
| < | ''spot.im'' | 0.01101 | //unresolved// | UNRESOLVED -- spot.im is dead (TLS handshake failure on both spot.im and www.spot.im, which also 502s); no primary source found tying the domain itself to a current operator | absent | unknown | unknown | [[https://www.openweb.com/legal-and-privacy/privacy/|legal-doc]] | | + Disconnect Q1 agkn.com unknown unresolved Register-by-number: SEC EDGAR full-text search for the exact phrase "agkn.com" |
| --- | + Disconnect Q2 mapixl.com unknown unresolved All three routes tried and none produced an identifier for the actual operator |
| > | ''spot.im'' | 0.01101 | Open Web Technologies Ltd. (OpenWeb) | renamed 2020 (Spot.IM -> OpenWeb) | absent | current | current | [[https://www.openweb.com/newsroom|parent-site]] | | + Disconnect Q3 stat-track.com unknown unresolved All three routes were tried. Register-by-number: Disconnect's 'StackTrack' is |
| 177c147 | + Disconnect Q3 stripst.com unknown unresolved Register-by-number: no imprint or company number is available anywhere for str |
| < | ''company-target.com'' | 0.00897 | //unresolved// | UNRESOLVED -- domain resolves to CloudFront-range IPs but TLS handshake fails (alert 552) and HTTP returns 403/no response; no content or primary source reachable | unknown | unknown | unknown | //none// | | + Disconnect Q3 htplayground.com unknown unresolved register-by-number: Disconnect's own entry is just the domain string, so there |
| --- | |
| > | ''company-target.com'' | 0.00897 | Demandbase, Inc. | n/a -- registered 2012-08-06, no evidence of any transfer; Demandbase's own documentation still calls it their domain as of 2025 | current | current | current | [[https://web.archive.org/web/20250908011403id_/https://support.demandbase.com/hc/en-us/articles/360019872411-Data-Collection-Overview|parent-site]] | | --- F. The adjudication table --------------------------------------- |
| 179,180c149,150 | |
| < | ''adgrx.com'' | 0.00681 | //unresolved// | UNRESOLVED -- news coverage says AdGear Technologies (acquired by Samsung Electronics Canada, 2016) operates adgrx.com as part of Samsung Ads, but no fetch reached a primary source naming adgrx.com: samsungads.ca/en/privacy/ has an expired TLS certificate, help.samsungads.com/docs/samsung-ad-manager-privacy-notice failed TLS handshake, and samsung.com's general ads privacy page names no subsidiary or domain. adgrx.com itself has no TLS handshake and no RDAP. | unknown | unknown | unknown | //none// | | @@ -147,7 +117,7 @@ |
| < | ''marphezis.com'' | 0.00666 | //unresolved// | UNRESOLVED -- domain does not resolve (no TLS handshake, no RDAP, HTTP code 000); Brightcom's own homepage and privacy policy (brightcom.com) make no mention of marphezis.com or any operated-domains list. | absent | unknown | absent | //none// | | | ''smartadserver.com'' | 0.07333 | Equativ SAS (RCS Paris 487 613 481) | renamed 2023 -- Smart Adserver / Smartadserver S.A.S. became Equativ after its merger with DAX | stale | stale | current | [[https://www.equativ.com/legal-mentions|legal-doc]] | |
| --- | | ''lijit.com'' | 0.06419 | Sovrn Holdings, Inc. | n/a -- Sovrn Holdings, Inc. operates lijit.com today | current | current | current | [[https://lijit.com/|legal-doc]] | |
| > | ''adgrx.com'' | 0.00681 | AdGear Technologies, Inc. ("AdGear", trading as Samsung Ads; wholly-owned subsidiary of Samsung Electronics Canada, Inc.) | n/a -- current since Samsung's 2018 acquisition of AdGear | current | current | granularity | [[https://samsungads.ca/en/cookie-policy/|legal-doc]] | | | ''bidr.io'' | 0.05861 | Beeswax Inc. (a FreeWheel / Comcast company) | closed ~2021-01 -- FreeWheel (Comcast) agreed 2020-12-17 to acquire Beeswax | current | current | granularity | [[https://www.beeswax.com/privacy/|legal-doc]] | |
| > | ''marphezis.com'' | 0.00666 | UNKNOWN | UNRESOLVED -- no primary source located | absent | unknown | absent | [[NONE -- no primary source reached|none]] | | -| ''1rx.io'' | 0.05313 | //unresolved// | UNRESOLVED -- domain is dead (no TLS handshake, no HTTP response) and no primary source fetched this session ties 1rx.io itself to RhythmOne or Nexxen | unknown | unknown | unknown | //none// | |
| 190,191c160,161 | +| ''1rx.io'' | 0.05313 | UNKNOWN | UNRESOLVED -- no primary source located | unknown | unknown | unknown | [[NONE -- no primary source reached|none]] | |
| < | ''app-us1.com'' | 0.00351 | //unresolved// | UNRESOLVED -- could not fetch a primary source naming the operator of this specific domain | absent | unknown | unknown | //none// | | | ''simpli.fi'' | 0.05248 | Simplifi Holdings, LLC (GTCR-backed, privately held) | n/a -- privately held, longstanding GTCR portfolio company | current | granularity | current | [[https://simpli.fi/website-privacy-policy|legal-doc]] | |
| < | ''govx.com'' | 0.00330 | //unresolved// | UNRESOLVED -- every fetch of govx.com returned only the homepage title, no legal-entity document reached | absent | unknown | unknown | //none// | | | ''gumgum.com'' | 0.05151 | GumGum, Inc. | n/a -- independent | current | current | current | [[https://gumgum.com/terms-and-policies/privacy-policy|legal-doc]] | |
| --- | | ''quantserve.com'' | 0.04428 | Quantcast Corporation (Quantcast) | n/a -- independent | current | current | current | [[https://legal.quantcast.com/|legal-doc]] | |
| > | ''app-us1.com'' | 0.00351 | ActiveCampaign, LLC (ActiveCampaign) | n/a -- independent, registered 2013, no ownership change found | absent | current | current | [[whois://whois.markmonitor.com/app-us1.com|domain-register]] | | @@ -155,7 +125,7 @@ |
| > | ''govx.com'' | 0.00330 | GovX, Inc. | n/a -- independent, no ownership change found | absent | current | current | [[https://www.sec.gov/Archives/edgar/data/1623323/000162332316000003/xslFormDX01/primary_doc.xml|filing]] | | | ''klaviyo.com'' | 0.04172 | Klaviyo, Inc. | n/a -- independent | absent | current | current | [[https://www.klaviyo.com/de/|legal-doc]] | |
| 193c163 | | ''loopme.me'' | 0.04039 | LoopMe Ltd (LoopMe) | n/a -- independent | current | error | current | [[https://legal.loopme.com/privacy-center|legal-doc]] | |
| < | ''cdnbasket.net'' | 0.00305 | //unresolved// | UNRESOLVED -- domain is dead (no TLS handshake, CODE 000); only a third-party WHOIS aggregator ties it historically to Bounce Exchange, not a primary source | absent | unknown | unknown | //none// | | | ''fontawesome.com'' | 0.03873 | Fonticons, Inc. | n/a -- independent | current | current | absent | [[https://fontawesome.com/tos|legal-doc]] | |
| --- | -| ''agkn.com'' | 0.03580 | //unresolved// | UNRESOLVED -- domain is dead today (NXDOMAIN/no TLS handshake) though RDAP shows it remains registered (not expired, locked, registrar GoDaddy Corporate Domains) with no disclosed registrant org; could not reach any current primary source naming an operator | unknown | unknown | unknown | //none// | |
| > | ''cdnbasket.net'' | 0.00305 | UNRESOLVED | UNRESOLVED -- no primary source ties cdnbasket.net to any named entity | absent | unknown | unknown | [[https://www.wunderkind.co/privacy-policy|legal-doc]] | | +| ''agkn.com'' | 0.03580 | UNRESOLVED | UNRESOLVED -- registry WHOIS registrant is privacy-proxied (Brandsight/GoDaddy Corporate Domains) with no organization disclosed, and TransUnion's own pages that would plausibly name AdAdvisor/agkn.com are bot-walled | unknown | unknown | unknown | //none// | |
| 197c167 | | ''wp.com'' | 0.02694 | Automattic, Inc. (WordPress.com) | n/a -- independent, operated by Automattic since founding (2005) | current | current | current | [[https://automattic.com/|parent-site]] | |
| < | ''opti-digital.com'' | 0.00211 | //unresolved// | UNRESOLVED -- domain does not resolve (DNS failure, no TLS handshake) as of 2026-09-05 | absent | unknown | unknown | [[https://optidigital.com/legal-notice/|legal-doc]] | | | ''eyeota.net'' | 0.02519 | Eyeota, a Dun & Bradstreet company | acquired by Dun & Bradstreet, announced 2021-11-04 (Eyeota retains its own brand as a D&B subsidiary) | current | current | granularity | [[https://www.eyeota.com/blog/eyeota-is-now-a-dun-bradstreet-company|newsroom]] | |
| --- | | ''newrelic.com'' | 0.02217 | New Relic, Inc. | n/a -- independent, unchanged | current | current | current | [[https://newrelic.com/termsandconditions/terms|legal-doc]] | |
| > | ''opti-digital.com'' | 0.00211 | Opti Digital SAS (2 Rue des Cortalets, 66400 Céret, France) | n/a -- independent, live company | absent | current | current | [[https://optidigital.com/legal-notice/|legal-doc]] | | @@ -169,15 +139,15 @@ |
| 206c176 | | ''ctnsnet.com'' | 0.01392 | Crimtan Holdings Limited | n/a -- independent, unchanged | current | current | current | [[https://find-and-update.company-information.service.gov.uk/company/07810698|register]] | |
| < | ''ksearchnet.com'' | 0.00129 | //unresolved// | UNRESOLVED -- no qualifying primary source found | absent | absent | unknown | [[https://github.com/klevultd/frontend-sdk/blob/master/packages/klevu-core/README.md|other (non-qualifying: SDK docs on GitHub, not a newsroom/filing/legal-doc/register/parent-site)]] | | | ''sharethis.com'' | 0.01349 | Predactiv, Inc. (ShareThis) | n/a -- ShareThis currently operates as a division of Predactiv, Inc.; the date this structure began was not sourced | granularity | granularity | current | [[https://sharethis.com/privacy/|legal-doc]] | |
| --- | | ''mountain.com'' | 0.01288 | MNTN, Inc. | renamed 2022-06-14 (MNTN Digital, Inc. -> MNTN, Inc.), per SEC EDGAR | absent | stale | current | [[https://data.sec.gov/submissions/CIK0001891027.json|filing]] | |
| > | ''ksearchnet.com'' | 0.00129 | Klevu Oy (operating subsidiary of Athos Commerce) | n/a -- Klevu (and Klevu Oy, its Finnish operating entity) merged with Searchspring in 2024/2025 to form Athos Commerce; ksearchnet.com is still run under the Klevu Oy legal name today | absent | absent | granularity | [[tls://eucs23v2.ksearchnet.com:443|register]] | | -| ''spot.im'' | 0.01101 | //unresolved// | UNRESOLVED -- spot.im is dead (TLS handshake failure on both spot.im and www.spot.im, which also 502s); no primary source found tying the domain itself to a current operator | absent | unknown | unknown | [[https://www.openweb.com/legal-and-privacy/privacy/|legal-doc]] | |
| 208c178 | +| ''spot.im'' | 0.01101 | Open Web Technologies Ltd. (OpenWeb) | renamed 2020 (Spot.IM -> OpenWeb) | absent | current | current | [[https://www.openweb.com/newsroom|parent-site]] | |
| < | ''kameleoon.io'' | 0.00108 | //unresolved// | UNRESOLVED -- domain is dead (no TLS handshake, no HTTP response) and no fetched source names kameleoon.io specifically | absent | absent | unknown | [[https://www.kameleoon.com/legal-notice|legal-doc]] | | | ''cloudfront.net'' | 0.01081 | Amazon Web Services, Inc. (Amazon) | n/a -- Amazon since inception | current | granularity | current | [[https://aws.amazon.com/privacy/|legal-doc]] | |
| --- | | ''hotjar.io'' | 0.01015 | Contentsquare | UNRESOLVED -- exact Hotjar acquisition date not sourced this session; current state confirmed via redirect + copyright | stale | stale | current | [[https://contentsquare.com/hotjar/?utm_campaign=exp__hj_hp_redirection|parent-site]] | |
| > | ''kameleoon.io'' | 0.00108 | Kameleoon SAS (Kameleoon) | n/a -- independent, no ownership change found | absent | absent | current | [[https://www.kameleoon.com/legal-notice|legal-doc]] | | | ''stripe.com'' | 0.00967 | Stripe, LLC | UNRESOLVED -- exact date of the Stripe, Inc. -> Stripe, LLC change not sourced; current name confirmed via site footer + EV TLS cert | absent | stale | current | [[https://stripe.com/de-ch/legal/consumer|legal-doc]] | |
| 212c182 | | ''azure.com'' | 0.00960 | Microsoft Corporation | n/a -- Microsoft since inception | absent | current | current | [[tls://azure.microsoft.com:443|register]] | |
| < | ''gssprt.jp'' | 0.00101 | //unresolved// | UNRESOLVED -- domain unreachable in probe (no TLS handshake, no RDAP) and no fetched Geniee source names gssprt.jp | unknown | unknown | unknown | [[https://en.geniee.co.jp/privacy/|legal-doc]] | | -| ''company-target.com'' | 0.00897 | //unresolved// | UNRESOLVED -- domain resolves to CloudFront-range IPs but TLS handshake fails (alert 552) and HTTP returns 403/no response; no content or primary source reachable | unknown | unknown | unknown | //none// | |
| --- | +| ''company-target.com'' | 0.00897 | Demandbase, Inc. | n/a -- registered 2012-08-06, no evidence of any transfer; Demandbase's own documentation still calls it their domain as of 2025 | current | current | current | [[https://web.archive.org/web/20250908011403id_/https://support.demandbase.com/hc/en-us/articles/360019872411-Data-Collection-Overview|parent-site]] | |
| > | ''gssprt.jp'' | 0.00101 | Geniee, Inc. | n/a -- independent since domain creation (2013) | current | current | current | [[whois://whois.jprs.jp/gssprt.jp|domain-register]] | | | ''paypal.com'' | 0.00751 | PayPal, Inc. | n/a -- independent, no ownership change found | absent | current | current | [[https://www.paypal.com/us/legalhub/paypal/privacy-full|legal-doc]] | |
| 218c188 | -| ''adgrx.com'' | 0.00681 | //unresolved// | UNRESOLVED -- news coverage says AdGear Technologies (acquired by Samsung Electronics Canada, 2016) operates adgrx.com as part of Samsung Ads, but no fetch reached a primary source naming adgrx.com: samsungads.ca/en/privacy/ has an expired TLS certificate, help.samsungads.com/docs/samsung-ad-manager-privacy-notice failed TLS handshake, and samsung.com's general ads privacy page names no subsidiary or domain. adgrx.com itself has no TLS handshake and no RDAP. | unknown | unknown | unknown | //none// | |
| < | ''acint.net'' | 0.00065 | //unresolved// | UNRESOLVED -- no primary source (company filing, legal doc with a real legal name, or parent site) identifies an operator; the only registrant lead is an individual's WHOIS record, which is not an admissible primary source | absent | unknown | unknown | [[https://acint.net/|legal-doc]] | | -| ''marphezis.com'' | 0.00666 | //unresolved// | UNRESOLVED -- domain does not resolve (no TLS handshake, no RDAP, HTTP code 000); Brightcom's own homepage and privacy policy (brightcom.com) make no mention of marphezis.com or any operated-domains list. | absent | unknown | absent | //none// | |
| --- | +| ''adgrx.com'' | 0.00681 | AdGear Technologies, Inc. ("AdGear", trading as Samsung Ads; wholly-owned subsidiary of Samsung Electronics Canada, Inc.) | n/a -- current since Samsung's 2018 acquisition of AdGear | current | current | granularity | [[https://samsungads.ca/en/cookie-policy/|legal-doc]] | |
| > | ''acint.net'' | 0.00065 | Poshibalov Evgeny Vasilyevich, operating the self-titled project 'Acint (Artificial Computation Intelligence)' | n/a -- same registrant since domain creation 2014-02-06 per registrar WHOIS | absent | current | current | [[whois://whois.publicdomainregistry.com/acint.net|domain-register]] | | +| ''marphezis.com'' | 0.00666 | UNKNOWN | UNRESOLVED -- no primary source located | absent | unknown | absent | [[NONE -- no primary source reached|none]] | |
| 220c190 | | ''iqzone.com'' | 0.00663 | IQzone Inc. | n/a -- independent, live and self-operated; no ownership change found | absent | current | current | [[https://iqzone.com/|legal-doc]] | |
| < | ''yceml.net'' | 0.00062 | //unresolved// | UNRESOLVED -- structural evidence (conversantmedia.com now 301-redirects to www.epsilon.com, and its /legal/privacy path redirects to legal.epsilon.com) points to Conversant now sitting under Epsilon/Publicis Groupe, but no fetched page carried on-page text naming that relationship, so it does not clear the quote bar | unknown | unknown | unknown | [[https://www.conversantmedia.com/|parent-site]] | | | ''histats.com'' | 0.00650 | Wisecode S.r.l. Unipersonale (Histats) | n/a -- Wisecode operates Histats.com directly; no ownership change found | current | current | current | [[https://www.histats.com/|legal-doc]] | |
| --- | | ''adition.com'' | 0.00612 | Virtual Minds GmbH (Adition) | ADITION technologies AG merged into the Virtual Minds group (~2013); Virtual Minds AG converted its legal form to Virtual Minds GmbH (name/legal-form change, ~2021); Adition is now marketed as a Virtual Minds GmbH product | stale | stale | current | [[https://virtualminds.com/adition/|parent-site]] | |
| > | ''yceml.net'' | 0.00062 | Conversant, Inc. (operating brand Epsilon; ultimate parent Publicis Groupe) | granularity: Conversant folded into Epsilon after Publicis Groupe's 2019 acquisition of Epsilon; 'Here, There & Everywhere' (webXray's parent field, an Alliance Data-era name) is superseded/stale | stale | current | granularity | [[whois://whois.markmonitor.com/yceml.net|domain-register]] | | @@ -187,14 +157,14 @@ |
| 222c192 | | ''forter.com'' | 0.00419 | Forter Ltd. (Israel; affiliates include Forter, Inc. — US, Forter Solutions UK Ltd., Forter Pte Ltd.) | n/a -- independent | absent | current | current | [[https://www.forter.com/privacy-policy/|legal-doc]] | |
| < | ''travelpayouts.com'' | 0.00049 | //unresolved// | UNRESOLVED -- privacy-policy/terms pages 403'd or 404'd on every URL tried this session; only the homepage was fetchable | absent | unknown | absent | [[https://www.travelpayouts.com/|legal-doc]] | | | ''zoominfo.com'' | 0.00405 | ZoomInfo Technologies LLC (ZoomInfo) | n/a -- current operating entity per live privacy policy; 'Zoom Information, Inc.' was a company ZoomInfo acquired in Feb 2019, since folded into the ZoomInfo corporate structure | absent | stale | current | [[https://www.zoominfo.com/privacy-policy|legal-doc]] | |
| --- | | ''braze.com'' | 0.00353 | Braze, Inc. | n/a -- independent, publicly traded (NASDAQ: BRZE) | current | current | current | [[https://www.braze.com/privacy|legal-doc]] | |
| > | ''travelpayouts.com'' | 0.00049 | Go Travel Un Limited (Hong Kong; trading as Travelpayouts) | n/a -- current per the domain's own Terms of Service, archived 2026-02-17 (~7 months old); Hong Kong entity registered 18.08.2011, Registration number 1658681 | absent | current | absent | [[https://web.archive.org/web/20260217212541id_/https://support.travelpayouts.com/hc/en-us/articles/360004162111-Terms-of-the-Travelpayouts-Travel-Affiliate-Network?tp_place=footer_wp_wa|archived-legal-doc]] | | -| ''app-us1.com'' | 0.00351 | //unresolved// | UNRESOLVED -- could not fetch a primary source naming the operator of this specific domain | absent | unknown | unknown | //none// | |
| 225c195 | -| ''govx.com'' | 0.00330 | //unresolved// | UNRESOLVED -- every fetch of govx.com returned only the homepage title, no legal-entity document reached | absent | unknown | unknown | //none// | |
| < | ''blogblog.com'' | 0.00041 | //unresolved// | UNRESOLVED -- no primary source names this specific domain | unknown | unknown | absent | //none// | | +| ''app-us1.com'' | 0.00351 | ActiveCampaign, LLC (ActiveCampaign) | n/a -- independent, registered 2013, no ownership change found | absent | current | current | [[whois://whois.markmonitor.com/app-us1.com|domain-register]] | |
| --- | +| ''govx.com'' | 0.00330 | GovX, Inc. | n/a -- independent, no ownership change found | absent | current | current | [[https://www.sec.gov/Archives/edgar/data/1623323/000162332316000003/xslFormDX01/primary_doc.xml|filing]] | |
| > | ''blogblog.com'' | 0.00041 | Google LLC (Blogger) | n/a -- registered 2000-09-15, currently Google-operated (Blogger), no change evidenced | current | current | absent | [[whois://whois.markmonitor.com/blogblog.com|domain-register]] | | | ''visualstudio.com'' | 0.00325 | Microsoft Corporation | n/a -- Microsoft has always operated this domain (Visual Studio product site) | absent | current | current | [[tls://visualstudio.com:443|register]] | |
| 230,231c200,201 | -| ''cdnbasket.net'' | 0.00305 | //unresolved// | UNRESOLVED -- domain is dead (no TLS handshake, CODE 000); only a third-party WHOIS aggregator ties it historically to Bounce Exchange, not a primary source | absent | unknown | unknown | //none// | |
| < | ''at-o.net'' | 0.00027 | //unresolved// | UNRESOLVED -- domain does not resolve; no operator identifiable from a primary source | absent | absent | unknown | //none// | | +| ''cdnbasket.net'' | 0.00305 | UNRESOLVED | UNRESOLVED -- no primary source ties cdnbasket.net to any named entity | absent | unknown | unknown | [[https://www.wunderkind.co/privacy-policy|legal-doc]] | |
| < | ''cnevids.com'' | 0.00026 | //unresolved// | UNRESOLVED -- no primary source names the owner | absent | unknown | absent | //none// | | | ''elfsight.com'' | 0.00303 | Elfsight, SL | n/a -- current operator per live Terms of Service and About page | absent | error | current | [[https://elfsight.com/terms-of-service/|legal-doc]] | |
| --- | | ''ispot.tv'' | 0.00293 | iSpot.tv, Inc. | n/a -- independent | absent | current | current | [[https://www.ispot.tv/privacy|legal-doc]] | |
| > | ''at-o.net'' | 0.00027 | Applied Technologies Internet SAS (AT Internet), controlled by Piano Software, Inc. | acquired by Piano -- exact date not confirmed by a primary source fetched this session | absent | absent | current | [[https://recherche-entreprises.api.gouv.fr/search?q=Applied%20Technologies%20Internet|register]] | | | ''reson8.com'' | 0.00239 | Resonate Networks, Inc. (Resonate) | n/a -- current, per live privacy policy naming reson8.com URLs as its own opt-out/tracking infrastructure | current | current | current | [[https://www.resonate.com/privacy-policy/|legal-doc]] | |
| > | ''cnevids.com'' | 0.00026 | Condé Nast Entertainment (Advance Publications) | n/a -- Sabin, Bermant & Gould LLP was never the operator, only the registrant's outside-counsel contact | absent | error | absent | [[https://player.cnevids.com/|legal-doc]] | | -| ''opti-digital.com'' | 0.00211 | //unresolved// | UNRESOLVED -- domain does not resolve (DNS failure, no TLS handshake) as of 2026-09-05 | absent | unknown | unknown | [[https://optidigital.com/legal-notice/|legal-doc]] | |
| 233c203 | +| ''opti-digital.com'' | 0.00211 | Opti Digital SAS (2 Rue des Cortalets, 66400 Céret, France) | n/a -- independent, live company | absent | current | current | [[https://optidigital.com/legal-notice/|legal-doc]] | |
| < | ''mapixl.com'' | 0.00022 | //unresolved// | UNRESOLVED -- site returns Cloudflare 403, no TLS org identity, no RDAP/registrant data, and no primary source ties any company to this domain | absent | absent | unknown | //none// | | | ''nofraud.com'' | 0.00205 | Wyllo LLC (rebranded from NoFraud) | rebranded NoFraud -> Wyllo; exact rebrand date UNRESOLVED (current privacy policy dated 2026-03-19) | absent | stale | stale | [[https://wyllo.ai/privacy/|legal-doc]] | |
| --- | | ''zopim.com'' | 0.00200 | Zendesk, Inc. | acquired by Zendesk 2014-04-10 | current | current | current | [[https://www.zendesk.com/company/press/zendesk-acquires-live-chat-leader-zopim/|newsroom]] | |
| > | ''mapixl.com'' | 0.00022 | UNRESOLVED -- Cloudflare-walled site, no legal page ever archived, registrant is a privacy proxy | UNRESOLVED -- no dated evidence either way | absent | absent | unknown | [[whois://whois.godaddy.com/mapixl.com|domain-register]] | | | ''mparticle.com'' | 0.00192 | mParticle, Inc. (a Rokt company, post-merger) | merged into Rokt, announced 2025-01-16 (~US$300M deal) | absent | current | granularity | [[https://www.mparticle.com/news/rokt-and-mparticle-merge/|newsroom]] | |
| 237c207 | @@ -203,70 +173,70 @@ |
| < | ''htplayground.com'' | 0.00016 | //unresolved// | UNRESOLVED -- domain unreachable (no TLS handshake, no HTTP response), WHOIS/RDAP registrant is a privacy-protection proxy, and the only archived content is a 2015 Wayback snapshot with no operator identification | absent | absent | unknown | //none// | | | ''squarecdn.com'' | 0.00153 | Block, Inc. | Square, Inc. renamed to Block, Inc. 2021-12-10 | absent | current | absent | [[https://squareup.com/us/en/legal/general/privacy|legal-doc]] | |
| --- | | ''sendtonews.com'' | 0.00144 | Minute Media (via its STN Video subsidiary/brand, formerly SendtoNews) | acquired by Minute Media, announced 2024-01 (~US$150M, per STN Video's own archived press release) | absent | absent | current | [[https://web.archive.org/web/20240718175821/https://www.stnvideo.com/press/minute-media-acquires-stn-video/|newsroom]] | |
| > | ''htplayground.com'' | 0.00016 | UNKNOWN | UNRESOLVED -- no primary source located | absent | absent | unknown | [[NONE -- no primary source reached|none]] | | | ''amung.us'' | 0.00143 | whos.amung.us Inc. | n/a -- independent | current | current | current | [[https://whos.amung.us/legal/terms/|legal-doc]] | |
| 242c212 | -| ''ksearchnet.com'' | 0.00129 | //unresolved// | UNRESOLVED -- no qualifying primary source found | absent | absent | unknown | [[https://github.com/klevultd/frontend-sdk/blob/master/packages/klevu-core/README.md|other (non-qualifying: SDK docs on GitHub, not a newsroom/filing/legal-doc/register/parent-site)]] | |
| < | ''contentabc.com'' | 0.00012 | //unresolved// | UNRESOLVED -- domain does not resolve (NXDOMAIN on direct fetch, probe CODE:000, no TLS handshake); WebSearch leads (urlscan, whois, Aylo's own brands page) point to Aylo but Aylo's brands page (https://www.aylo.com/brands/) does NOT name contentabc.com, and no privacy policy/legal page naming the domain could be fetched | absent | unknown | unknown | //none// | | +| ''ksearchnet.com'' | 0.00129 | Klevu Oy (operating subsidiary of Athos Commerce) | n/a -- Klevu (and Klevu Oy, its Finnish operating entity) merged with Searchspring in 2024/2025 to form Athos Commerce; ksearchnet.com is still run under the Klevu Oy legal name today | absent | absent | granularity | [[tls://eucs23v2.ksearchnet.com:443|register]] | |
| --- | | ''everestjs.net'' | 0.00121 | Adobe Inc. | n/a -- long-standing Adobe Advertising infrastructure domain | current | current | current | [[https://experienceleague.adobe.com/en/docs/id-service/using/reference/csp|parent-site]] | |
| > | ''contentabc.com'' | 0.00012 | UNRESOLVED | UNRESOLVED -- no live page, fully privacy-proxied registrant, zero Wayback captures, no primary source ties either TR's 'Aylo' or Disconnect's 'ContentABC' to the domain | absent | unknown | unknown | //none// | | -| ''kameleoon.io'' | 0.00108 | //unresolved// | UNRESOLVED -- domain is dead (no TLS handshake, no HTTP response) and no fetched source names kameleoon.io specifically | absent | absent | unknown | [[https://www.kameleoon.com/legal-notice|legal-doc]] | |
| 244c214 | +| ''kameleoon.io'' | 0.00108 | Kameleoon SAS (Kameleoon) | n/a -- independent, no ownership change found | absent | absent | current | [[https://www.kameleoon.com/legal-notice|legal-doc]] | |
| < | ''pages02.net'' | 0.00012 | //unresolved// | UNRESOLVED -- could not fetch a primary source tying pages02.net specifically to a current operator | absent | unknown | unknown | //none// | | | ''aralego.com'' | 0.00105 | ucfunnel | n/a -- ucfunnel operates aralego.com as its publisher/buyer dashboard domain | absent | current | current | [[https://dashboard.aralego.com/|legal-doc]] | |
| --- | | ''govdelivery.com'' | 0.00105 | Granicus, LLC | merged 2016-10-25 (GovDelivery merged with Granicus, majority-backed by Vista Equity Partners) | granularity | current | absent | [[https://granicus.com/blog/granicus-govdelivery-announce-merger/|newsroom]] | |
| > | ''pages02.net'' | 0.00012 | Acoustic, L.P. | sold 2019 (IBM divested Watson Marketing/Commerce to Centerbridge Partners, relaunched as Acoustic) | absent | stale | current | [[whois://whois.markmonitor.com/pages02.net|domain-register]] | | | ''cookiefirst.com'' | 0.00104 | Digital Data Solutions B.V. | n/a -- long-standing operator | absent | current | absent | [[https://cookiefirst.com/privacy-policy/|legal-doc]] | |
| 247,248c217,218 | -| ''gssprt.jp'' | 0.00101 | //unresolved// | UNRESOLVED -- domain unreachable in probe (no TLS handshake, no RDAP) and no fetched Geniee source names gssprt.jp | unknown | unknown | unknown | [[https://en.geniee.co.jp/privacy/|legal-doc]] | |
| < | ''stat-track.com'' | 0.00011 | //unresolved// | UNRESOLVED -- domain does not resolve (probe CODE:000, no TLS); no primary source found | absent | absent | unknown | //none// | | +| ''gssprt.jp'' | 0.00101 | Geniee, Inc. | n/a -- independent since domain creation (2013) | current | current | current | [[whois://whois.jprs.jp/gssprt.jp|domain-register]] | |
| < | ''trustpilot.net'' | 0.00010 | //unresolved// | UNRESOLVED -- domain does not serve content under its own hostname; TLS cert covers only *.trustpilot.com (mismatch on direct fetch) and carries no validated-organisation (O=) field, so it does not meet the register-grade bar | absent | unknown | absent | //none// | | | ''researchnow.com'' | 0.00097 | Dynata, LLC | renamed 2019-01-15 (Research Now SSI rebranded as Dynata); ownership itself changed again 2024-07-02 when lienholders took 100% equity in a court-approved restructuring | current | stale | current | [[https://www.dynata.com/why-dynata/about-dynata/press/research-now-and-ssi-merger-successfully-completed/|newsroom]] | |
| --- | | ''vidazoo.com'' | 0.00078 | Perion Network Ltd. | acquired 2021-10-04 | absent | stale | current | [[https://www.sec.gov/Archives/edgar/data/1338940/000117891321003077/exhibit_99-1.htm|filing]] | |
| > | ''stat-track.com'' | 0.00011 | UNRESOLVED | UNRESOLVED -- domain shows no activity since a 2002-2011 self-hosted install that predates the current 2016 registration | absent | absent | unknown | [[https://rdap.verisign.com/com/v1/domain/stat-track.com|domain-register]] | | | ''tmdb.org'' | 0.00075 | TiVo Platform Technologies LLC (a subsidiary of Xperi Inc.) | changed -- Fanhattan LLC acquired TMDB circa 2016; exact date of the subsequent transition to TiVo Platform Technologies LLC / Xperi Inc. not verified this session | absent | stale | absent | [[https://www.themoviedb.org/terms-of-use|legal-doc]] | |
| > | ''trustpilot.net'' | 0.00010 | Trustpilot A/S (Pilestraede 58, 5th floor, DK-1112 Copenhagen K, Denmark) | n/a -- defensive/redirect domain of the same live company, not a separate acquisition | absent | current | absent | [[https://corporate.trustpilot.com/legal/for-reviewers/privacy-policy-end-user|legal-doc]] | | | ''tns-counter.ru'' | 0.00071 | AO ADFACT / JSC ADFACT (Cyrillic АО «ЭДФАКТ»), a subsidiary of AO Mediascope | n/a -- ADFACT has run under the Mediascope brand since Mediascope's 2017 rebrand from TNS Russia | current | current | granularity | [[https://www.mediascope.net/about/|parent-site]] | |
| 254,256c224,226 | | ''owneriq.net'' | 0.00067 | Inmar, Inc. | acquired 2019-10-22 | stale | current | current | [[https://www.globenewswire.com/news-release/2019/10/22/1933083/0/en/Inmar-Acquires-ownerIQ-Adding-A-Critical-Data-and-Media-Solution-to-Inmar-s-Newly-Launched-Innovator-Ecosystem.html|newsroom]] | |
| < | ''awltovhc.com'' | 0.00007 | //unresolved// | UNRESOLVED -- domain dead (no TLS handshake, HTTP 000) and no primary source names this exact domain | unknown | unknown | unknown | [[n/a -- no primary source located|n/a]] | | -| ''acint.net'' | 0.00065 | //unresolved// | UNRESOLVED -- no primary source (company filing, legal doc with a real legal name, or parent site) identifies an operator; the only registrant lead is an individual's WHOIS record, which is not an admissible primary source | absent | unknown | unknown | [[https://acint.net/|legal-doc]] | |
| < | ''collective-media.net'' | 0.00007 | //unresolved// | UNRESOLVED -- no confirmed current owner found | unknown | unknown | unknown | [[n/a -- no primary source located|n/a]] | | +| ''acint.net'' | 0.00065 | Poshibalov Evgeny Vasilyevich, operating the self-titled project 'Acint (Artificial Computation Intelligence)' | n/a -- same registrant since domain creation 2014-02-06 per registrar WHOIS | absent | current | current | [[whois://whois.publicdomainregistry.com/acint.net|domain-register]] | |
| < | ''mmstat.com'' | 0.00007 | //unresolved// | UNRESOLVED -- no primary source located | unknown | unknown | unknown | [[n/a -- no primary source located|n/a]] | | | ''force.com'' | 0.00064 | Salesforce, Inc. | n/a -- current; salesforce.com, inc. formally renamed to Salesforce, Inc. effective 2022-04-04 | absent | stale | current | [[https://investor.salesforce.com/news/news-details/2022/Salesforce-Changes-Legal-Name-to-Salesforce-Inc/default.aspx|newsroom]] | |
| --- | -| ''yceml.net'' | 0.00062 | //unresolved// | UNRESOLVED -- structural evidence (conversantmedia.com now 301-redirects to www.epsilon.com, and its /legal/privacy path redirects to legal.epsilon.com) points to Conversant now sitting under Epsilon/Publicis Groupe, but no fetched page carried on-page text naming that relationship, so it does not clear the quote bar | unknown | unknown | unknown | [[https://www.conversantmedia.com/|parent-site]] | |
| > | ''awltovhc.com'' | 0.00007 | Epsilon (d/b/a 'Epsilon PeopleCloud Digital Media Solutions', formerly Conversant); ultimate parent Publicis Groupe SA | Conversant retired as the public-facing brand by 2021 per Epsilon's own press release; Epsilon itself was acquired by Publicis Groupe in 2019 | stale | stale | granularity | [[https://www.epsilon.com/us/about-us/pressroom/epsilon-peoplecloud-digital-media-solutions-receives-accreditation-from-the-media-rating-council-for-correlated-outcomes|newsroom]] | | +| ''yceml.net'' | 0.00062 | Conversant, Inc. (operating brand Epsilon; ultimate parent Publicis Groupe) | granularity: Conversant folded into Epsilon after Publicis Groupe's 2019 acquisition of Epsilon; 'Here, There & Everywhere' (webXray's parent field, an Alliance Data-era name) is superseded/stale | stale | current | granularity | [[whois://whois.markmonitor.com/yceml.net|domain-register]] | |
| > | ''collective-media.net'' | 0.00007 | UNKNOWN | UNRESOLVED -- dead domain, redacted WHOIS, no usable archived legal page | unknown | unknown | unknown | [[n/a|n/a]] | | | ''conviva.com'' | 0.00049 | Conviva | n/a -- independent, current | current | current | current | [[https://www.conviva.ai/|legal-doc]] | |
| > | ''mmstat.com'' | 0.00007 | UNRESOLVED | UNRESOLVED -- registrar is Alibaba Cloud Computing (Beijing) but no Registrant Organization is disclosed in WHOIS, the official Chinese ICP registry is behind a JS challenge, and Alibaba's own live privacy pages do not mention this domain | unknown | unknown | unknown | //none// | | -| ''travelpayouts.com'' | 0.00049 | //unresolved// | UNRESOLVED -- privacy-policy/terms pages 403'd or 404'd on every URL tried this session; only the homepage was fetchable | absent | unknown | absent | [[https://www.travelpayouts.com/|legal-doc]] | |
| 258c228 | +| ''travelpayouts.com'' | 0.00049 | Go Travel Un Limited (Hong Kong; trading as Travelpayouts) | n/a -- current per the domain's own Terms of Service, archived 2026-02-17 (~7 months old); Hong Kong entity registered 18.08.2011, Registration number 1658681 | absent | current | absent | [[https://web.archive.org/web/20260217212541id_/https://support.travelpayouts.com/hc/en-us/articles/360004162111-Terms-of-the-Travelpayouts-Travel-Affiliate-Network?tp_place=footer_wp_wa|archived-legal-doc]] | |
| < | ''lduhtrp.net'' | 0.00006 | //unresolved// | UNRESOLVED -- domain dead (no TLS handshake, HTTP 000) and no primary source names this exact domain | unknown | unknown | unknown | [[n/a -- no primary source located|n/a]] | | | ''newsmemory.com'' | 0.00044 | Tecnavia | n/a -- independent, current | absent | current | absent | [[https://www.tecnavia.com/about-us|newsroom]] | |
| --- | | ''solarwinds.com'' | 0.00043 | SolarWinds Worldwide, LLC (operating entity); ultimate parent Turn/River Capital | 2025-04-16 -- Turn/River Capital completed take-private acquisition | stale | current | current | [[https://www.solarwinds.com/company/newsroom/press-releases/turnriver-completes-acquisition-of-solarwinds|newsroom]] | |
| > | ''lduhtrp.net'' | 0.00006 | Conversant, Inc. (Epsilon / Publicis Groupe) | n/a -- registrant unchanged; Conversant's parent Epsilon was acquired by Publicis Groupe, closed 2019-07-02 | current | current | granularity | [[whois://whois.markmonitor.com/lduhtrp.net|domain-register]] | | -| ''blogblog.com'' | 0.00041 | //unresolved// | UNRESOLVED -- no primary source names this specific domain | unknown | unknown | absent | //none// | |
| 260c230 | +| ''blogblog.com'' | 0.00041 | Google LLC (Blogger) | n/a -- registered 2000-09-15, currently Google-operated (Blogger), no change evidenced | current | current | absent | [[whois://whois.markmonitor.com/blogblog.com|domain-register]] | |
| < | ''hqseek.com'' | 0.00005 | //unresolved// | UNRESOLVED -- could not fetch a primary source | absent | unknown | absent | //none// | | | ''blogspot.com'' | 0.00034 | Google LLC (Blogger; EEA/Switzerland: Google Ireland Limited) | n/a -- long-standing Google/Blogger property | current | current | current | [[https://policies.google.com/privacy|legal-doc]] | |
| --- | | ''snapengage.com'' | 0.00034 | SnapEngage, LLC (subsidiary of TeamSupport LLC) | 2021-05-20 -- TeamSupport acquired SnapEngage (per public reporting); ownership confirmed current via SnapEngage's own privacy policy | absent | stale | current | [[https://snapengage.com/privacy-policy|legal-doc]] | |
| > | ''hqseek.com'' | 0.00005 | UNRESOLVED -- privacy-proxied registrant, and the one named operator found predates the current registration state by two decades | UNRESOLVED -- no continuity evidence across the gap | absent | unknown | absent | [[https://web.archive.org/web/20040404075437id_/http://www.hqseek.com/|archived-legal-doc]] | | | ''yieldlove-ad-serving.net'' | 0.00034 | Yieldlove GmbH (majority-owned by Ströer SE & Co. KGaA) | 2017-10-04 -- Ströer acquired a majority shareholding in Yieldlove GmbH | absent | current | current | [[https://www.stroeer.de/en/newsroom/press/expansion-of-technological-platform-marketing-stroeer-acquires-programmatic-platform-and-header-bidding-specialists-yieldlove/|newsroom]] | |
| 262,263c232,233 | | ''ibb.co'' | 0.00029 | ImgBB | n/a -- independent, brand-operated service | absent | current | absent | [[https://imgbb.com/tos|legal-doc]] | |
| < | ''stripst.com'' | 0.00005 | //unresolved// | UNRESOLVED -- site unreachable in this session | absent | absent | unknown | //none// | | -| ''at-o.net'' | 0.00027 | //unresolved// | UNRESOLVED -- domain does not resolve; no operator identifiable from a primary source | absent | absent | unknown | //none// | |
| < | ''wishabi.com'' | 0.00005 | //unresolved// | UNRESOLVED -- no fetchable primary source naming wishabi.com | unknown | unknown | unknown | //none// | | -| ''cnevids.com'' | 0.00026 | //unresolved// | UNRESOLVED -- no primary source names the owner | absent | unknown | absent | //none// | |
| --- | +| ''at-o.net'' | 0.00027 | Applied Technologies Internet SAS (AT Internet), controlled by Piano Software, Inc. | acquired by Piano -- exact date not confirmed by a primary source fetched this session | absent | absent | current | [[https://recherche-entreprises.api.gouv.fr/search?q=Applied%20Technologies%20Internet|register]] | |
| > | ''stripst.com'' | 0.00005 | UNRESOLVED | UNRESOLVED -- fully privacy-proxied registrant, DV-only TLS cert, zero Wayback captures, and Stripchat's own live legal pages are bot-blocked | absent | absent | unknown | //none// | | +| ''cnevids.com'' | 0.00026 | Condé Nast Entertainment (Advance Publications) | n/a -- Sabin, Bermant & Gould LLP was never the operator, only the registrant's outside-counsel contact | absent | error | absent | [[https://player.cnevids.com/|legal-doc]] | |
| > | ''wishabi.com'' | 0.00005 | Flipp Corp. | n/a -- current | current | current | current | [[https://flipp.com/en-us/privacy-policy|legal-doc]] | | | ''pushnami.com'' | 0.00024 | Pushnami, LLC | n/a -- independent, no ownership change found | absent | current | current | [[https://pushnami.com/privacy-policy/|legal-doc]] | |
| 269c239 | -| ''mapixl.com'' | 0.00022 | //unresolved// | UNRESOLVED -- site returns Cloudflare 403, no TLS org identity, no RDAP/registrant data, and no primary source ties any company to this domain | absent | absent | unknown | //none// | |
| < | ''sa-as.com'' | 0.00003 | //unresolved// | UNRESOLVED -- no primary source located; site has no TLS handshake and no title/copyright to check | absent | absent | unknown | //none// | | +| ''mapixl.com'' | 0.00022 | UNRESOLVED -- Cloudflare-walled site, no legal page ever archived, registrant is a privacy proxy | UNRESOLVED -- no dated evidence either way | absent | absent | unknown | [[whois://whois.godaddy.com/mapixl.com|domain-register]] | |
| --- | | ''responsiveads.com'' | 0.00018 | ResponsiveAds, Inc. | n/a -- independent, no ownership change known | absent | current | absent | [[https://responsiveads.com/|legal-doc]] | |
| > | ''sa-as.com'' | 0.00003 | FoundryCo, Inc. | n/a -- independent per current WHOIS, no ownership-change evidence found | absent | absent | current | [[whois://whois.markmonitor.com/sa-as.com|domain-register]] | | | ''crsspxl.com'' | 0.00016 | Cross Pixel Media, Inc. | n/a -- independent, no ownership change known | current | current | current | [[https://crosspixel.net/privacy-policy/|legal-doc]] | |
| 288c258 | | ''cudasvc.com'' | 0.00016 | Barracuda Networks, Inc. | n/a -- independent, no ownership change known | absent | current | current | [[https://campus.barracuda.com/product/campus/doc/89096320/required-outbound-connections-for-barracuda-networks-appliances/|parent-site]] | |
| < | ''cratecamera.com'' | 0.00001 | //unresolved// | UNRESOLVED -- domain 404s and carries only a DV Let's Encrypt cert (no validated organisation); no reachable newsroom/legal-doc/register source | absent | unknown | unknown | [[n/a|n/a]] | | -| ''htplayground.com'' | 0.00016 | //unresolved// | UNRESOLVED -- domain unreachable (no TLS handshake, no HTTP response), WHOIS/RDAP registrant is a privacy-protection proxy, and the only archived content is a 2015 Wayback snapshot with no operator identification | absent | absent | unknown | //none// | |
| --- | +| ''htplayground.com'' | 0.00016 | UNKNOWN | UNRESOLVED -- no primary source located | absent | absent | unknown | [[NONE -- no primary source reached|none]] | |
| > | ''cratecamera.com'' | 0.00001 | Leven Labs, Inc. (DBA Admiral) | n/a -- registrant unchanged since domain creation 2020-08-31 per registrar WHOIS | absent | current | current | [[whois://whois.namecheap.com/cratecamera.com|domain-register]] | | | ''userzoom.com'' | 0.00016 | UserTesting, Inc. (brand: UserTesting; formerly independent UserZoom, Inc.) | merged 2023-04-03 (Thoma Bravo portfolio companies UserTesting and UserZoom merged and now operate under the UserTesting name) | absent | stale | absent | [[https://www.thomabravo.com/press-releases/usertesting-and-userzoom-merge-to-help-organizations-build-human-centered-experiences-for-all|newsroom]] | |
| 291c261 | | ''cnzz.com'' | 0.00014 | Alibaba (China) Technology Co., Ltd. (Alibaba Group; operates as part of the Umeng+ / CNZZ analytics brand) | n/a -- CNZZ merged with Umeng and Diyuanxin into Umeng+ in 2016 under Alibaba; ownership unchanged since | current | granularity | granularity | [[tls://cnzz.com:443|register]] | |
| < | ''globo.com'' | 0.00001 | //unresolved// | UNRESOLVED -- WebFetch could not reach any globo.com/politicas.globo.com/privacidade.globo.com page to quote a legal-entity statement | absent | unknown | unknown | [[n/a|n/a]] | | | ''yahooapis.com'' | 0.00014 | Yahoo Inc. (majority owned by Apollo Funds/Apollo Global Management; Verizon retains a 10% stake) | changed 2021-09-01 (Apollo Funds completed acquisition of Yahoo, formerly Verizon Media) | stale | current | current | [[https://www.apollo.com/insights-news/pressreleases/2021/09/apollo-funds-complete-acquisition-of-yahoo-161530593|newsroom]] | |
| --- | | ''cnn.com'' | 0.00012 | Warner Bros. Discovery, Inc. | changed 2022-04-08 (WarnerMedia-Discovery merger formed Warner Bros. Discovery, Inc.) | absent | current | current | [[tls://cnn.com:443|register]] | |
| > | ''globo.com'' | 0.00001 | Globo Comunicação e Participações S.A. (CNPJ 27.865.757/0001-02) | n/a -- current, live operating entity | absent | current | granularity | [[https://privacidade.globo.com/privacy-policy/|legal-doc]] | | -| ''contentabc.com'' | 0.00012 | //unresolved// | UNRESOLVED -- domain does not resolve (NXDOMAIN on direct fetch, probe CODE:000, no TLS handshake); WebSearch leads (urlscan, whois, Aylo's own brands page) point to Aylo but Aylo's brands page (https://www.aylo.com/brands/) does NOT name contentabc.com, and no privacy policy/legal page naming the domain could be fetched | absent | unknown | unknown | //none// | |
| 293c263 | +| ''contentabc.com'' | 0.00012 | UNRESOLVED | UNRESOLVED -- no live page, fully privacy-proxied registrant, zero Wayback captures, no primary source ties either TR's 'Aylo' or Disconnect's 'ContentABC' to the domain | absent | unknown | unknown | //none// | |
| < | ''km0trk.com'' | 0.00001 | //unresolved// | UNRESOLVED -- domain gives no TLS handshake and no reachable content; operator could not be identified from any primary source | absent | absent | unknown | [[n/a|n/a]] | | | ''glomex.com'' | 0.00012 | glomex GmbH | n/a -- current, confirmed via own Impressum fetched 2026-09-05 | absent | absent | current | [[https://www.glomex.com/de/impressum|legal-doc]] | |
| --- | -| ''pages02.net'' | 0.00012 | //unresolved// | UNRESOLVED -- could not fetch a primary source tying pages02.net specifically to a current operator | absent | unknown | unknown | //none// | |
| > | ''km0trk.com'' | 0.00001 | Good On You Pty Ltd (Good On You) | n/a -- independent, ABN 75 608 419 085 active | absent | absent | error | [[https://goodonyou.eco/privacy-policy/|parent-site]] | | +| ''pages02.net'' | 0.00012 | Acoustic, L.P. | sold 2019 (IBM divested Watson Marketing/Commerce to Centerbridge Partners, relaunched as Acoustic) | absent | stale | current | [[whois://whois.markmonitor.com/pages02.net|domain-register]] | |
| 296c266 | | ''tqlkg.com'' | 0.00012 | Conversant, LLC (Epsilon Data Management, LLC; ultimate parent Publicis Groupe S.A.) | 2019-07 -- Publicis Groupe completed acquisition of Epsilon (including its Conversant business) from Alliance Data Systems | stale | current | granularity | [[https://www.publicisgroupe.com/en/news/press-releases/publicis-groupe-to-acquire-epsilon|newsroom]] | |
| < | ''offshoregeology.com'' | 0.00001 | //unresolved// | UNRESOLVED -- domain is dead (404), no primary source ties it to any current operator | absent | absent | unknown | [[NONE -- no primary source reached|n/a]] | | | ''appspot.com'' | 0.00011 | Google LLC (Alphabet Inc.) | n/a -- Google since appspot.com's 2008 launch as Google App Engine | current | current | absent | [[https://policies.google.com/terms|legal-doc]] | |
| --- | -| ''stat-track.com'' | 0.00011 | //unresolved// | UNRESOLVED -- domain does not resolve (probe CODE:000, no TLS); no primary source found | absent | absent | unknown | //none// | |
| > | ''offshoregeology.com'' | 0.00001 | Admiral (Leven Labs, Inc.) | n/a -- domain-register shows no ownership change since creation (2015-12-01, last WHOIS update 2025-11-02, no transfer flag) | absent | absent | current | [[https://levenlabs.com/|parent-site]] | | -| ''trustpilot.net'' | 0.00010 | //unresolved// | UNRESOLVED -- domain does not serve content under its own hostname; TLS cert covers only *.trustpilot.com (mismatch on direct fetch) and carries no validated-organisation (O=) field, so it does not meet the register-grade bar | absent | unknown | absent | //none// | |
| 303,304c273,274 | +| ''stat-track.com'' | 0.00011 | UNRESOLVED | UNRESOLVED -- domain shows no activity since a 2002-2011 self-hosted install that predates the current 2016 registration | absent | absent | unknown | [[https://rdap.verisign.com/com/v1/domain/stat-track.com|domain-register]] | |
| < | ''cedscdn.it'' | 0.00001 | //unresolved// | UNRESOLVED -- domain does not resolve (no TLS handshake); no primary source ties it to any current operator | absent | unknown | absent | [[NONE -- no primary source reached|n/a]] | | +| ''trustpilot.net'' | 0.00010 | Trustpilot A/S (Pilestraede 58, 5th floor, DK-1112 Copenhagen K, Denmark) | n/a -- defensive/redirect domain of the same live company, not a separate acquisition | absent | current | absent | [[https://corporate.trustpilot.com/legal/for-reviewers/privacy-policy-end-user|legal-doc]] | |
| < | ''cjponyparts.com'' | 0.00001 | //unresolved// | UNRESOLVED -- site returns HTTP 403 (Cloudflare challenge) on every page fetched this session | absent | unknown | absent | [[NONE -- no primary source reached|n/a]] | | | ''atgsvcs.com'' | 0.00008 | Oracle Corporation (successor to Art Technology Group, Inc.) | 2011-01-05 -- Oracle completed its acquisition; Art Technology Group, Inc. became a wholly owned subsidiary of Oracle | absent | current | current | [[https://www.sec.gov/Archives/edgar/data/1086195/000095012311000543/b83864e8vk.htm|filing]] | |
| --- | | ''juicer.cc'' | 0.00008 | Logly, Inc. | n/a -- current; product rebranded from 'Juicer' to 'LOGLY Audience Analytics' but still Logly-operated | absent | current | absent | [[https://juicer.cc/|legal-doc]] | |
| > | ''cedscdn.it'' | 0.00001 | CED Digital & Servizi S.r.l. (Caltagirone Editore group) | n/a -- registrant unchanged since domain creation 2014-09-12 per .it registry WHOIS | absent | error | absent | [[whois://whois.nic.it/cedscdn.it|register]] | | | ''richrelevance.com'' | 0.00008 | RichRelevance, Inc. (brand of Algonomy, which is now part of ADA) | changed 2021-01-19 (RichRelevance merged with Manthan Software to form Algonomy); Algonomy itself now shown as part of ADA per algonomy.com (exact date unconfirmed) | current | current | granularity | [[tls://richrelevance.com:443|register]] | |
| > | ''cjponyparts.com'' | 0.00001 | CJ Pony Parts, Inc. | n/a -- independent, continuously CJ Pony Parts, Inc. since at least 2015 | absent | current | absent | [[https://web.archive.org/web/20150405015158id_/http://www.cjponyparts.com/terms-of-use/a/110/?SID=|archived-legal-doc]] | | | ''medallia.com.au'' | 0.00008 | Medallia, Inc. | n/a -- independent | absent | absent | current | [[https://medallia.com.au/|legal-doc]] | |
| 309c279 | | ''medchatapp.com'' | 0.00008 | Medchat, LLC | n/a -- independent | current | absent | absent | [[https://medchatapp.com/site/|legal-doc]] | |
| < | ''i.ua'' | 0.00001 | //unresolved// | UNRESOLVED -- no primary legal document reachable; site is a JS-rendered shell with no static footer/legal text, and /legal returned 404 | absent | absent | unknown | //none// | | -| ''awltovhc.com'' | 0.00007 | //unresolved// | UNRESOLVED -- domain dead (no TLS handshake, HTTP 000) and no primary source names this exact domain | unknown | unknown | unknown | [[n/a -- no primary source located|n/a]] | |
| --- | -| ''collective-media.net'' | 0.00007 | //unresolved// | UNRESOLVED -- no confirmed current owner found | unknown | unknown | unknown | [[n/a -- no primary source located|n/a]] | |
| > | ''i.ua'' | 0.00001 | ТОВ «КЕПРЕЙТ ПАРТНЕРС» (LLC Keprait Partners), Ukrainian register code 33500955 | n/a -- registrant since 2019-07-21 per the registry record, no evidence of a later change | absent | absent | current | [[https://help.i.ua/agreement/|legal-doc]] | | -| ''mmstat.com'' | 0.00007 | //unresolved// | UNRESOLVED -- no primary source located | unknown | unknown | unknown | [[n/a -- no primary source located|n/a]] | |
| 323,325c293,295 | +| ''awltovhc.com'' | 0.00007 | Epsilon (d/b/a 'Epsilon PeopleCloud Digital Media Solutions', formerly Conversant); ultimate parent Publicis Groupe SA | Conversant retired as the public-facing brand by 2021 per Epsilon's own press release; Epsilon itself was acquired by Publicis Groupe in 2019 | stale | stale | granularity | [[https://www.epsilon.com/us/about-us/pressroom/epsilon-peoplecloud-digital-media-solutions-receives-accreditation-from-the-media-rating-council-for-correlated-outcomes|newsroom]] | |
| < | current | 69.9% | 55.5%–83.1% | 93.9% | 80.9%–98.8% | | +| ''collective-media.net'' | 0.00007 | UNKNOWN | UNRESOLVED -- dead domain, redacted WHOIS, no usable archived legal page | unknown | unknown | unknown | [[n/a|n/a]] | |
| < | granularity | 10.2% | 2.1%–20.9% | 0.2% | 0.0%–0.8% | | +| ''mmstat.com'' | 0.00007 | UNRESOLVED | UNRESOLVED -- registrar is Alibaba Cloud Computing (Beijing) but no Registrant Organization is disclosed in WHOIS, the official Chinese ICP registry is behind a JS challenge, and Alibaba's own live privacy pages do not mention this domain | unknown | unknown | unknown | //none// | |
| < | stale | 19.8% | 9.2%–32.3% | 5.9% | 1.0%–19.1% | | | ''compass-fit.jp'' | 0.00006 | MicroAd, Inc. | n/a -- no ownership change found; product appears discontinued (domain unreachable, product page now 404) but no other owner was ever named | absent | current | current | [[https://www.microad.co.jp/news/detail/1185/|newsroom]] | |
| --- | -| ''lduhtrp.net'' | 0.00006 | //unresolved// | UNRESOLVED -- domain dead (no TLS handshake, HTTP 000) and no primary source names this exact domain | unknown | unknown | unknown | [[n/a -- no primary source located|n/a]] | |
| > | current | 69.2% | 56.1%–81.4% | 93.8% | 80.9%–98.6% | | +| ''lduhtrp.net'' | 0.00006 | Conversant, Inc. (Epsilon / Publicis Groupe) | n/a -- registrant unchanged; Conversant's parent Epsilon was acquired by Publicis Groupe, closed 2019-07-02 | current | current | granularity | [[whois://whois.markmonitor.com/lduhtrp.net|domain-register]] | |
| > | granularity | 8.8% | 2.1%–17.5% | 0.2% | 0.0%–0.7% | | | ''cedexis.com'' | 0.00005 | Cloud Software Group, Inc. | changed 2018-02-12 (Citrix Systems acquired Cedexis); changed again 2022 (Citrix combined with TIBCO into Cloud Software Group) | stale | stale | current | [[tls://cedexis.com:443|register]] | |
| > | stale | 22.0% | 11.4%–33.8% | 5.9% | 1.1%–18.7% | | -| ''hqseek.com'' | 0.00005 | //unresolved// | UNRESOLVED -- could not fetch a primary source | absent | unknown | absent | //none// | |
| 327,328c297,298 | +| ''hqseek.com'' | 0.00005 | UNRESOLVED -- privacy-proxied registrant, and the one named operator found predates the current registration state by two decades | UNRESOLVED -- no continuity evidence across the gap | absent | unknown | absent | [[https://web.archive.org/web/20040404075437id_/http://www.hqseek.com/|archived-legal-doc]] | |
| < | current+gran. | 80.2% | 67.7%–90.8% | 94.1% | 81.4%–99.0% | | | ''sparkasse.de'' | 0.00005 | S-Communication Services GmbH (S-Com) | renamed 2022-09, effective 2023-01-01 -- formerly Sparkassen-Finanzportal GmbH | absent | stale | absent | [[https://www.sparkasse.de/impressum.html|legal-doc]] | |
| < | stale+error | 19.8% | 9.0%–32.6% | 5.9% | 1.1%–19.4% | | -| ''stripst.com'' | 0.00005 | //unresolved// | UNRESOLVED -- site unreachable in this session | absent | absent | unknown | //none// | |
| --- | -| ''wishabi.com'' | 0.00005 | //unresolved// | UNRESOLVED -- no fetchable primary source naming wishabi.com | unknown | unknown | unknown | //none// | |
| > | current+gran. | 78.0% | 65.8%–88.8% | 94.1% | 81.4%–98.8% | | +| ''stripst.com'' | 0.00005 | UNRESOLVED | UNRESOLVED -- fully privacy-proxied registrant, DV-only TLS cert, zero Wayback captures, and Stripchat's own live legal pages are bot-blocked | absent | absent | unknown | //none// | |
| > | stale+error | 22.0% | 11.0%–33.8% | 5.9% | 1.1%–18.9% | | +| ''wishabi.com'' | 0.00005 | Flipp Corp. | n/a -- current | current | current | current | [[https://flipp.com/en-us/privacy-policy|legal-doc]] | |
| 332,337c302,307 | | ''twiago.com'' | 0.00004 | twiago GmbH | n/a -- independent, no ownership change found | absent | current | absent | [[https://www.twiago.com/impressum|legal-doc]] | |
| < | current | 73.7% | 60.8%–85.8% | 68.6% | 29.3%–97.0% | | | ''wrbm.com'' | 0.00004 | William Reed Ltd | renamed 2022-02-09 -- formerly William Reed Business Media Ltd | absent | stale | absent | [[https://find-and-update.company-information.service.gov.uk/company/02883992|register]] | |
| < | granularity | 2.8% | 0.0%–8.3% | 0.1% | 0.0%–0.4% | | | ''hitslink.com'' | 0.00003 | Net Applications, Inc. | n/a -- no ownership change found | absent | absent | current | [[https://hitslink.com/|legal-doc]] | |
| < | stale | 22.1% | 10.5%–35.1% | 26.8% | 0.4%–64.6% | | | ''klarna.app'' | 0.00003 | Klarna Bank AB (publ) | n/a -- no ownership change found | absent | current | absent | [[https://www.klarna.com/international/|legal-doc]] | |
| < | error | 1.5% | 0.0%–4.4% | 4.5% | 0.0%–19.3% | | | ''onecount.net'' | 0.00003 | GCN Publishing, Inc. (d/b/a GCN Media Services; product: ONEcount) | n/a -- domain defunct (TLS cert expired 2022, no longer resolving); product now served from one-count.com by the same operator | granularity | current | current | [[https://www.one-count.com/privacy-policy/|parent-site]] | |
| < | current+gran. | 76.5% | 63.5%–88.0% | 68.7% | 29.2%–97.7% | | -| ''sa-as.com'' | 0.00003 | //unresolved// | UNRESOLVED -- no primary source located; site has no TLS handshake and no title/copyright to check | absent | absent | unknown | //none// | |
| < | stale+error | 23.5% | 11.7%–36.1% | 31.3% | 2.9%–71.2% | | +| ''sa-as.com'' | 0.00003 | FoundryCo, Inc. | n/a -- independent per current WHOIS, no ownership-change evidence found | absent | absent | current | [[whois://whois.markmonitor.com/sa-as.com|domain-register]] | |
| --- | | ''chango.com'' | 0.00003 | Chango & Co. LLC | UNRESOLVED exact date -- domain has been resold/repurposed to an unrelated interior design business; the ad-tech company 'Chango' (acquired by Rubicon Project, later renamed Magnite) no longer operates here | stale | stale | absent | [[https://www.chango.com/|legal-doc]] | |
| > | current | 73.9% | 61.6%–85.3% | 74.3% | 41.1%–97.8% | | | ''google.nl'' | 0.00003 | Google LLC (subsidiary of Alphabet Inc.) | n/a -- longstanding | granularity | current | current | [[https://policies.google.com/privacy|legal-doc]] | |
| > | granularity | 2.5% | 0.0%–7.5% | 0.1% | 0.0%–0.3% | | | ''americanexpress.com'' | 0.00002 | American Express Company | n/a -- longstanding | absent | current | current | [[tls://americanexpress.com:443|register]] | |
| > | stale | 19.2% | 9.1%–30.5% | 21.6% | 0.3%–53.1% | | @@ -285,28 +255,28 @@ |
| > | error | 4.4% | 0.0%–9.8% | 4.0% | 0.0%–15.7% | | | ''bestbuy.com'' | 0.00001 | Best Buy Co., Inc. | n/a -- independent | absent | current | absent | [[tls://bestbuy.com:443|register]] | |
| > | current+gran. | 76.4% | 64.6%–87.5% | 74.4% | 41.6%–97.8% | | | ''bilibili.com'' | 0.00001 | 上海幻电信息科技有限公司 (Shanghai Huandian Information Technology Co., Ltd.), operating subsidiary of Bilibili Inc. (NASDAQ: BILI) | n/a -- longstanding | absent | granularity | granularity | [[tls://bilibili.com:443|register]] | |
| > | stale+error | 23.6% | 12.5%–35.6% | 25.6% | 2.2%–58.5% | | | ''britishairways.com'' | 0.00001 | British Airways Plc | n/a -- independent operating subsidiary (part of International Airlines Group since 2011 merger; no change to the domain's operating entity) | absent | current | absent | [[tls://britishairways.com:443|register]] | |
| 341,346c311,316 | -| ''cratecamera.com'' | 0.00001 | //unresolved// | UNRESOLVED -- domain 404s and carries only a DV Let's Encrypt cert (no validated organisation); no reachable newsroom/legal-doc/register source | absent | unknown | unknown | [[n/a|n/a]] | |
| < | current | 94.4% | 86.9%–100.0% | 83.1% | 53.7%–100.0% | | +| ''cratecamera.com'' | 0.00001 | Leven Labs, Inc. (DBA Admiral) | n/a -- registrant unchanged since domain creation 2020-08-31 per registrar WHOIS | absent | current | current | [[whois://whois.namecheap.com/cratecamera.com|domain-register]] | |
| < | granularity | 4.4% | 0.0%–11.2% | 16.5% | 0.0%–45.8% | | | ''cxt.ms'' | 0.00001 | Taboola, Inc. (operating Connexity) | acquired 2021-09-01 -- Taboola completed its acquisition of Connexity | stale | stale | absent | [[https://www.taboola.com/press-releases/taboola-closes-connexity-acquisition/|newsroom]] | |
| < | stale | 1.2% | 0.0%–3.8% | 0.4% | 0.0%–1.5% | | | ''domdex.com'' | 0.00001 | Deloitte Digital (Deloitte Consulting LLP) | acquired 2018-09-10 -- Deloitte acquired Magnetic Media Online, Inc.'s AI/ad-tech platform business | current | stale | current | [[https://www.prnewswire.com/news-releases/deloitte-acquires-magnetics-artificial-intelligence-platform-business-300709565.html|newsroom]] | |
| < | error | 0.0% | 0.0%–0.0% | 0.0% | 0.0%–0.0% | | -| ''globo.com'' | 0.00001 | //unresolved// | UNRESOLVED -- WebFetch could not reach any globo.com/politicas.globo.com/privacidade.globo.com page to quote a legal-entity statement | absent | unknown | unknown | [[n/a|n/a]] | |
| < | current+gran. | 98.8% | 96.2%–100.0% | 99.6% | 98.5%–100.0% | | +| ''globo.com'' | 0.00001 | Globo Comunicação e Participações S.A. (CNPJ 27.865.757/0001-02) | n/a -- current, live operating entity | absent | current | granularity | [[https://privacidade.globo.com/privacy-policy/|legal-doc]] | |
| < | stale+error | 1.2% | 0.0%–3.8% | 0.4% | 0.0%–1.5% | | | ''hitc.com'' | 0.00001 | GRV Media Ltd | n/a -- independent, GRV Media Ltd has operated HITC since rebranding from Here Is The City in 2015 | absent | current | absent | [[https://grv.media/privacy/|legal-doc]] | |
| --- | -| ''km0trk.com'' | 0.00001 | //unresolved// | UNRESOLVED -- domain gives no TLS handshake and no reachable content; operator could not be identified from any primary source | absent | absent | unknown | [[n/a|n/a]] | |
| > | current | 89.2% | 80.5%–96.8% | 82.8% | 53.8%–100.0% | | +| ''km0trk.com'' | 0.00001 | Good On You Pty Ltd (Good On You) | n/a -- independent, ABN 75 608 419 085 active | absent | absent | error | [[https://goodonyou.eco/privacy-policy/|parent-site]] | |
| > | granularity | 7.6% | 1.1%–15.2% | 16.8% | 0.0%–45.6% | | | ''mediaset.es'' | 0.00001 | Grupo Audiovisual Mediaset España Comunicación, S.A.U. (subsidiary of MFE-MediaForEurope) | renamed/reorganized 2023-03-15 -- Mediaset España Comunicación, S.A. segregated its entire business to Grupo Audiovisual Mediaset España Comunicación, S.A.U. (BORME-C-2023-1010) | absent | stale | absent | [[tls://mediaset.es:443|register]] | |
| > | stale | 1.1% | 0.0%–3.4% | 0.4% | 0.0%–1.5% | | | ''ml.com'' | 0.00001 | Bank of America Corporation (Merrill Lynch) | n/a -- Merrill Lynch has been a Bank of America subsidiary since the 2009 acquisition; no recent change | absent | current | absent | [[tls://ml.com:443|register]] | |
| > | error | 2.1% | 0.0%–6.2% | 0.0% | 0.0%–0.0% | | -| ''offshoregeology.com'' | 0.00001 | //unresolved// | UNRESOLVED -- domain is dead (404), no primary source ties it to any current operator | absent | absent | unknown | [[NONE -- no primary source reached|n/a]] | |
| > | current+gran. | 96.8% | 91.5%–100.0% | 99.6% | 98.5%–100.0% | | +| ''offshoregeology.com'' | 0.00001 | Admiral (Leven Labs, Inc.) | n/a -- domain-register shows no ownership change since creation (2015-12-01, last WHOIS update 2025-11-02, no transfer flag) | absent | absent | current | [[https://levenlabs.com/|parent-site]] | |
| > | stale+error | 3.2% | 0.0%–8.5% | 0.4% | 0.0%–1.5% | | | ''qbox.me'' | 0.00001 | Shanghai Qiniu Information Technology Co., Ltd. (Qiniu Cloud) | n/a -- independent, no ownership change found | current | current | absent | [[https://www-static.qbox.me/en/company|legal-doc]] | |
| | | ''report-uri.io'' | 0.00001 | Report-URI Ltd. | n/a -- independent, no ownership change found | absent | current | absent | [[https://find-and-update.company-information.service.gov.uk/company/10943557|register]] | |
| | | ''vg.hu'' | 0.00001 | Mediaworks Hungary Zrt. | n/a -- independent, no ownership change found | absent | absent | current | [[https://www.vg.hu/impresszum|legal-doc]] | |
| | | ''webmd.com'' | 0.00001 | WebMD LLC (an Internet Brands company; Internet Brands is a KKR portfolio company) | closed 2017-09 -- Internet Brands (KKR) completed its tender offer acquisition of WebMD Health Corp | current | current | absent | [[https://www.webmd.com/|legal-doc]] | |
| | | ''99static.com'' | 0.00001 | 99designs Pty Ltd (99designs by Vista; a Cimpress plc subsidiary) | n/a -- currently a Cimpress plc subsidiary; historical acquisition date not verified against a primary source this session | absent | current | absent | [[https://www.sec.gov/Archives/edgar/data/0001262976/000126297626000027/ex211subsidiariesofcimpres.htm|filing]] | |
| | | ''adrta.com'' | 0.00001 | Pixalate, Inc. | n/a -- independent, no ownership change found | current | current | current | [[tls://adrta.com:443|register]] | |
| | -| ''cedscdn.it'' | 0.00001 | //unresolved// | UNRESOLVED -- domain does not resolve (no TLS handshake); no primary source ties it to any current operator | absent | unknown | absent | [[NONE -- no primary source reached|n/a]] | |
| | -| ''cjponyparts.com'' | 0.00001 | //unresolved// | UNRESOLVED -- site returns HTTP 403 (Cloudflare challenge) on every page fetched this session | absent | unknown | absent | [[NONE -- no primary source reached|n/a]] | |
| | +| ''cedscdn.it'' | 0.00001 | CED Digital & Servizi S.r.l. (Caltagirone Editore group) | n/a -- registrant unchanged since domain creation 2014-09-12 per .it registry WHOIS | absent | error | absent | [[whois://whois.nic.it/cedscdn.it|register]] | |
| | +| ''cjponyparts.com'' | 0.00001 | CJ Pony Parts, Inc. | n/a -- independent, continuously CJ Pony Parts, Inc. since at least 2015 | absent | current | absent | [[https://web.archive.org/web/20150405015158id_/http://www.cjponyparts.com/terms-of-use/a/110/?SID=|archived-legal-doc]] | |
| | | ''coverartarchive.org'' | 0.00001 | MetaBrainz Foundation, Inc. | n/a -- unchanged since inception; run jointly with Internet Archive | absent | current | absent | [[https://metabrainz.org/projects|parent-site]] | |
| | | ''dnb.com'' | 0.00001 | Dun & Bradstreet, Inc. | n/a -- operating/trading name unchanged; ultimate parent went private via an investor consortium in Feb 2022 | current | current | absent | [[https://web.archive.org/web/20241002130956/https://www.dnb.com/utility-pages/privacy-policy.html|legal-doc]] | |
| | | ''experian.com'' | 0.00001 | Experian Information Solutions, Inc. (brand: Experian; ultimate parent: Experian plc, LSE-listed) | n/a -- long-standing corporate structure | current | granularity | current | [[tls://experian.com:443|register]] | |
| | | ''farfetch-contents.com'' | 0.00001 | Farfetch UK Limited (part of the Coupang, Inc. group since its Jan 2024 acquisition of Farfetch Holdings) | Farfetch Holdings acquired by Coupang, Inc. 2024-01-30; Farfetch UK Limited (no. 06400760, formerly Farfetch.com Limited) remains an active UK subsidiary | absent | current | absent | [[https://find-and-update.company-information.service.gov.uk/company/06400760|register]] | |
| | -| ''i.ua'' | 0.00001 | //unresolved// | UNRESOLVED -- no primary legal document reachable; site is a JS-rendered shell with no static footer/legal text, and /legal returned 404 | absent | absent | unknown | //none// | |
| | +| ''i.ua'' | 0.00001 | ТОВ «КЕПРЕЙТ ПАРТНЕРС» (LLC Keprait Partners), Ukrainian register code 33500955 | n/a -- registrant since 2019-07-21 per the registry record, no evidence of a later change | absent | absent | current | [[https://help.i.ua/agreement/|legal-doc]] | |
| | | ''jdpower.com'' | 0.00001 | J.D. Power | n/a -- current | absent | current | current | [[tls://jdpower.com:443|register]] | |
| | | ''jobs2careers.com'' | 0.00001 | Talroo, Inc. | n/a -- Jobs2Careers is an active Talroo brand/channel, not a former or renamed entity | absent | current | absent | [[https://www.talroo.com/|parent-site]] | |
| | | ''myaccountaccess.com'' | 0.00001 | U.S. Bank National Association | n/a -- current | absent | current | absent | [[tls://myaccountaccess.com:443|register]] | |
| | @@ -320,28 +290,28 @@ |
| | |
| | === webXray === |
| | ^ Verdict ^ Domain-level ^ 95% CI ^ Encounter-weighted ^ 95% CI ^ |
| | -| current | 69.9% | 55.5%–83.1% | 93.9% | 80.9%–98.8% | |
| | -| granularity | 10.2% | 2.1%–20.9% | 0.2% | 0.0%–0.8% | |
| | -| stale | 19.8% | 9.2%–32.3% | 5.9% | 1.0%–19.1% | |
| | +| current | 69.2% | 56.1%–81.4% | 93.8% | 80.9%–98.6% | |
| | +| granularity | 8.8% | 2.1%–17.5% | 0.2% | 0.0%–0.7% | |
| | +| stale | 22.0% | 11.4%–33.8% | 5.9% | 1.1%–18.7% | |
| | | error | 0.0% | 0.0%–0.0% | 0.0% | 0.0%–0.0% | |
| | -| current+gran. | 80.2% | 67.7%–90.8% | 94.1% | 81.4%–99.0% | |
| | -| stale+error | 19.8% | 9.0%–32.6% | 5.9% | 1.1%–19.4% | |
| | +| current+gran. | 78.0% | 65.8%–88.8% | 94.1% | 81.4%–98.8% | |
| | +| stale+error | 22.0% | 11.0%–33.8% | 5.9% | 1.1%–18.9% | |
| | |
| | === Tracker Radar === |
| | ^ Verdict ^ Domain-level ^ 95% CI ^ Encounter-weighted ^ 95% CI ^ |
| | -| current | 73.7% | 60.8%–85.8% | 68.6% | 29.3%–97.0% | |
| | -| granularity | 2.8% | 0.0%–8.3% | 0.1% | 0.0%–0.4% | |
| | -| stale | 22.1% | 10.5%–35.1% | 26.8% | 0.4%–64.6% | |
| | -| error | 1.5% | 0.0%–4.4% | 4.5% | 0.0%–19.3% | |
| | -| current+gran. | 76.5% | 63.5%–88.0% | 68.7% | 29.2%–97.7% | |
| | -| stale+error | 23.5% | 11.7%–36.1% | 31.3% | 2.9%–71.2% | |
| | +| current | 73.9% | 61.6%–85.3% | 74.3% | 41.1%–97.8% | |
| | +| granularity | 2.5% | 0.0%–7.5% | 0.1% | 0.0%–0.3% | |
| | +| stale | 19.2% | 9.1%–30.5% | 21.6% | 0.3%–53.1% | |
| | +| error | 4.4% | 0.0%–9.8% | 4.0% | 0.0%–15.7% | |
| | +| current+gran. | 76.4% | 64.6%–87.5% | 74.4% | 41.6%–97.8% | |
| | +| stale+error | 23.6% | 12.5%–35.6% | 25.6% | 2.2%–58.5% | |
| | |
| | === Disconnect === |
| | ^ Verdict ^ Domain-level ^ 95% CI ^ Encounter-weighted ^ 95% CI ^ |
| | -| current | 94.4% | 86.9%–100.0% | 83.1% | 53.7%–100.0% | |
| | -| granularity | 4.4% | 0.0%–11.2% | 16.5% | 0.0%–45.8% | |
| | -| stale | 1.2% | 0.0%–3.8% | 0.4% | 0.0%–1.5% | |
| | -| error | 0.0% | 0.0%–0.0% | 0.0% | 0.0%–0.0% | |
| | -| current+gran. | 98.8% | 96.2%–100.0% | 99.6% | 98.5%–100.0% | |
| | -| stale+error | 1.2% | 0.0%–3.8% | 0.4% | 0.0%–1.5% | |
| | +| current | 89.2% | 80.5%–96.8% | 82.8% | 53.8%–100.0% | |
| | +| granularity | 7.6% | 1.1%–15.2% | 16.8% | 0.0%–45.6% | |
| | +| stale | 1.1% | 0.0%–3.4% | 0.4% | 0.0%–1.5% | |
| | +| error | 2.1% | 0.0%–6.2% | 0.0% | 0.0%–0.0% | |
| | +| current+gran. | 96.8% | 91.5%–100.0% | 99.6% | 98.5%–100.0% | |
| | +| stale+error | 3.2% | 0.0%–8.5% | 0.4% | 0.0%–1.5% | |
| | |
| </code> | </code> |
| |
| ===== O. Unedited output: scripts/report_tail_pass.py ===== | ===== Q. Unedited output: scripts/report_tail_pass.py ===== |
| |
| Run as ''python3 scripts/report_tail_pass.py --sample out/owner_sample.json --before out/adj_rows_scored.json --after out/adj_rows_tail_final.json --before-output out/owner_random_sample-output.txt --after-output out/owner_random_sample-tail-output.txt''. | Run as ''python3 scripts/report_tail_pass.py --sample out/owner_sample.json --before out/adj_rows_scored.json --after out/adj_rows_tail_final.json --before-output out/owner_random_sample-output.txt --after-output out/owner_random_sample-tail-output.txt''. |
| |
| Raw counts, unweighted, because this is a comparison between two groups of | Raw counts, unweighted, because this is a comparison between two groups of |
| drawn rows and not an estimate for the list. current+granularity against | drawn rows and not an estimate for the list. Fisher's exact, two-sided. A |
| stale+error; Fisher's exact, two-sided. A significantly worse recovered | significantly worse recovered group would support the worry that the rates |
| group would support the worry that the rates are flattered. | are flattered. |
| |
| List already-eligible good/n recovered good/n p (Fisher) | **Both metrics, because the answer depends on which one you ask.** The first |
| webXray 39/49 5/7 0.635 | version of this script tested current+granularity here and current-only in |
| Tracker Radar 36/47 7/9 1.000 | section E2 -- each the choice that makes its own claim mildest -- and did not |
| Disconnect 41/42 11/12 0.398 | say so. A review pass caught it. Neither metric is privileged: `current` is |
| pooled 116/138 23/28 0.782 | the stricter reading, `current+granularity` is the one the content page says |
| | is right when the unit of analysis is 'which company'. |
| | |
| | metric: current only |
| | List already-eligible good/n recovered good/n p (Fisher) |
| | webXray 35/49 5/7 1.000 |
| | Tracker Radar 35/47 7/9 1.000 |
| | Disconnect 39/42 9/12 0.116 |
| | pooled 109/138 21/28 0.622 |
| | |
| | metric: current+granularity |
| | List already-eligible good/n recovered good/n p (Fisher) |
| | webXray 39/49 5/7 0.635 |
| | Tracker Radar 36/47 7/9 1.000 |
| | Disconnect 41/42 11/12 0.398 |
| | pooled 116/138 23/28 0.782 |
| |
| The pooled row counts a domain once per list that drew it, so it is not an | The pooled row counts a domain once per list that drew it, so it is not an |
| --- E. Rate movement, computed with owner_random_sample.py's estimator --- | --- E. Rate movement, computed with owner_random_sample.py's estimator --- |
| |
| Imported, not reimplemented. 95% intervals are stratified bootstrap | Imported, not reimplemented -- the point estimator is owner_random_sample.py's |
| percentiles, 10,000 resamples, seed 20260905 -- the estimator's own. | own `strat_estimate`. **Point estimates only.** The 95% intervals are in the |
| | estimator's own output and nowhere else: its bootstrap draws for several |
| | verdicts from one generator, so re-running it here from a generator seeded |
| | identically prints a different interval for the same estimate. |
| |
| webXray | webXray |
| current+gran. 80.2% 78.0% -2.2 94.1% 94.1% -0.0 | current+gran. 80.2% 78.0% -2.2 94.1% 94.1% -0.0 |
| stale+error 19.8% 22.0% +2.2 5.9% 5.9% +0.0 | stale+error 19.8% 22.0% +2.2 5.9% 5.9% +0.0 |
| current, after, 95% CI: domain [56.1%, 81.4%] encounter [80.9%, 98.6%] | |
| |
| Tracker Radar | Tracker Radar |
| current+gran. 76.5% 76.4% -0.1 68.7% 74.4% +5.7 | current+gran. 76.5% 76.4% -0.1 68.7% 74.4% +5.7 |
| stale+error 23.5% 23.6% +0.1 31.3% 25.6% -5.7 | stale+error 23.5% 23.6% +0.1 31.3% 25.6% -5.7 |
| current, after, 95% CI: domain [61.7%, 85.5%] encounter [41.4%, 97.7%] | |
| |
| Disconnect | Disconnect |
| current+gran. 98.8% 96.8% -2.0 99.6% 99.6% -0.0 | current+gran. 98.8% 96.8% -2.0 99.6% 99.6% -0.0 |
| stale+error 1.2% 3.2% +2.0 0.4% 0.4% +0.0 | stale+error 1.2% 3.2% +2.0 0.4% 0.4% +0.0 |
| current, after, 95% CI: domain [80.7%, 96.8%] encounter [53.6%, 100.0%] | |
| |
| --- E2. The three pairwise comparisons, recomputed --------------------- | --- E2. The three pairwise comparisons, recomputed --------------------- |
| Disconnect 39/42 48/54 92.9% 88.9% | Disconnect 39/42 48/54 92.9% 88.9% |
| |
| Pair p before p after survives 0.0167 after? | **The result depends on the metric, and publishing only one of them was |
| Disconnect vs webXray 0.014 0.031 no | this script's most misleading omission.** Counting `granularity` as |
| Disconnect vs Tracker Radar 0.025 0.083 no | acceptable -- which the content page says is right when the unit of analysis |
| Tracker Radar vs webXray 0.820 0.831 no | is 'which company' rather than 'which legal entity' -- BOTH Disconnect |
| | comparisons clear Bonferroni, before and after, and the second pass overturns |
| | nothing. On `current` only, Disconnect-versus-webXray stops clearing it. |
| | Report both or report neither. |
| | |
| | metric: current only |
| | Pair p before p after survives 0.0167 before / after? |
| | Disconnect vs webXray 0.014 0.031 yes / no (48/54 vs 40/56) |
| | Disconnect vs Tracker Radar 0.025 0.083 no / no (48/54 vs 42/56) |
| | Tracker Radar vs webXray 0.820 0.831 no / no (42/56 vs 40/56) |
| | |
| | metric: current+granularity |
| | Pair p before p after survives 0.0167 before / after? |
| | Disconnect vs webXray 0.010 0.008 yes / yes (52/54 vs 44/56) |
| | Disconnect vs Tracker Radar 0.004 0.004 yes / yes (52/54 vs 43/56) |
| | Tracker Radar vs webXray 0.807 1.000 no / no (43/56 vs 44/56) |
| |
| --- E3. Sensitivity: the recovered rows that rest on the weakest source --- | --- E3. Sensitivity: the recovered rows that rest on the weakest source --- |
| dropped: cratecamera.com, sa-as.com | dropped: cratecamera.com, sa-as.com |
| |
| List eligible current (domain) vs. full pass | 1. uncorroborated registrant organisation only (2 row(s): cratecamera.com, sa-as.com) |
| webXray 56 69.2% +0.0 | the narrowest reading: a row whose ONLY evidence is a registrant name nobody validated. |
| Tracker Radar 55 73.3% -0.6 | List eligible current raw current (domain) vs. full pass |
| Disconnect 53 89.2% +0.0 | webXray 56 40/56 69.2% +0.0 |
| | Tracker Radar 55 41/55 73.3% -0.6 |
| | Disconnect 53 47/53 89.2% +0.0 |
| | Disconnect vs webXray, current only: p = 0.032 (does not survive 0.0167) |
| | Disconnect vs Tracker Radar, current only: p = 0.082 (does not survive 0.0167) |
| | |
| | 2. EVERY row resting on a source kind the first pass did not accept (11 row(s): acint.net, app-us1.com, blogblog.com, cjponyparts.com, cratecamera.com, gssprt.jp, lduhtrp.net, pages02.net, sa-as.com, travelpayouts.com, yceml.net) |
| | the same-bar comparison: what the rates would be if the second pass had used |
| | the first pass's bar exactly and settled only what that bar could reach. |
| | List eligible current raw current (domain) vs. full pass |
| | webXray 53 38/53 69.2% +0.0 |
| | Tracker Radar 51 37/51 72.1% -1.8 |
| | Disconnect 51 45/51 89.0% -0.2 |
| | Disconnect vs webXray, current only: p = 0.050 (does not survive 0.0167) |
| | Disconnect vs Tracker Radar, current only: p = 0.079 (does not survive 0.0167) |
| | |
| | 3. the i.ua rescore, under the reading this run rejected (1 row(s): i.ua) |
| | if 'I.UA' is held to name no company at all, the row is an error rather than |
| | current, which is the reading the hand correction argued against. Dropping the |
| | row entirely is the conservative stand-in, since scoring it `error` would lower |
| | Disconnect further than dropping it does. |
| | List eligible current raw current (domain) vs. full pass |
| | webXray 56 40/56 69.2% +0.0 |
| | Tracker Radar 56 42/56 73.9% +0.0 |
| | Disconnect 53 47/53 88.6% -0.6 |
| | Disconnect vs webXray, current only: p = 0.032 (does not survive 0.0167) |
| | Disconnect vs Tracker Radar, current only: p = 0.084 (does not survive 0.0167) |
| |
| --- F. Residue after the second pass, printed in full ----------------- | --- F. Residue after the second pass, printed in full ----------------- |
| </code> | </code> |
| |
| ===== P. Unedited output: scripts/check_tail_figures_mutations.sh ===== | ===== R. Unedited output: scripts/check_tail_figures_mutations.sh ===== |
| |
| Run as ''bash scripts/check_tail_figures_mutations.sh''. | Run as ''bash scripts/check_tail_figures_mutations.sh''. |
| caught delete the sensitivity-dropped ''sa-as.com'' | caught delete the sensitivity-dropped ''sa-as.com'' |
| caught delete the error domain ''km0trk.com'' | caught delete the error domain ''km0trk.com'' |
| | caught same-bar Tracker Radar 72.1% -> 72.6% |
| | caught same-bar Disconnect 89.0% -> 89.5% |
| | caught same-bar Tracker Radar delta (−1.8) -> (−1.9) |
| | caught delete the same-bar Disconnect delta (−0.2) |
| | caught i.ua alternative reading 88.6% -> 87.6% |
| | caught reintroduce the superseded Disconnect rate 89.2% -> 94.4% |
| | caught reintroduce the superseded Tracker Radar rate 73.9% -> 73.7% |
| | caught reintroduce the superseded webXray encounter rate 93.8% -> 93.9% |
| | caught revert ONE copy of the hand-changed count 30 -> 29 of the 525 |
| |
| mutations caught: 15; survived or broken: 0 | mutations caught: 24; survived or broken: 0 |
| </code> | </code> |
| |
