anoni.net

ProjectsOONI Coverage

OONI coverage in Hong Kong

On this page
  1. Networks with the most users
  2. Networks without measurements
  3. Where measurements come from
  4. Daily measurements
  5. Users covered
  6. Messaging app tests
  7. Across regions
  8. Glossary
  9. How the numbers are calculated
  10. Help fill the gaps
  11. Reading the measurement data

OONI's censorship measurements come from volunteers running OONI Probe on their own connections. Within one region, each provider and network (ASN, autonomous system number) may block different sites, so when only one or two providers are measured, blocking on the others goes unrecorded. This page sets OONI's measurement counts against APNIC's estimates of how many users each network has, to show how many users the measurements cover, which networks they concentrate on, and which networks have none.

The page is regenerated every hour and OONI's measurements are read every six hours. Taiwan is shown by default, with other regions as reference points; switch between them with the region list below. The glossary at the end of the page explains the fields. Each quarter's changes are written up in the quarterly reports.

ChinaMyanmarCambodiaPakistanBangladeshTaiwan Users covered 88.6%Macau Users covered 85.9%Japan Users covered 84.4%South Korea Users covered 94.5%Singapore Users covered 82.1%Vietnam Users covered 99.0%Thailand Users covered 97.5%Malaysia Users covered 96.5%Indonesia Users covered 80.6%Philippines Users covered 89.9%India Users covered 91.6%Hong Kong Users covered 66.6%
Users covered<75%75–85%85–90%90–95%≥95%Hong KongReference (app table only)Select a region on the map to open its page. Reference regions appear only in the messaging app table, for contrast with their records of blocking.

Measurements up to 2026-10-19 UTC · user estimates from 2026-10-15

Networks measured 47 +1vs previous 30 days
Users covered 66.6% -7.7 ppvs previous 30 days
Measurements 111,395 -28,522vs previous 30 days
Top two networks 45.1% -4.9 ppvs previous 30 days

Cards show the last 30 days, and the small charts show each day over the last 60. One measurement is OONI Probe's result for one website or service, so a single run of the website test produces hundreds. Users covered is the share of the region's internet users on networks with measurements. A lower share for the top two networks means the measurements are more spread out. pp means percentage points, the difference between two percentages.

Networks with the most users

The ten networks with the most users by the estimates of APNIC, the registry that allocates IP addresses and ASNs in the Asia Pacific. Users is the network's share of the region's internet users, Measured is its share of the region's measurements over the last 30 days, and Days is how many of those days it had any. Names are copied from registry records; in Taiwan, for example, HINET is Chunghwa Telecom's fixed network and EMOME-NET its mobile network. Networks with many users and few measurements are where running OONI Probe helps most.

Networks with the most usersUsersMeasured
ASNNameUsersMeasuredDays
AS4760 AS4760HKTIMS-AP - HKT Limited 17.1% 22.8% 30
AS9269 AS9269HKBN-AS-AP - Hong Kong Broadband Network Ltd. 11.7% 2.9% 11
AS9231 AS9231IPEOPLESNET-AS-AP - China Mobile Hong Kong Company Limited 6.1% 0.0% 0
AS38819 AS38819HKCSL-AS-AP - HKCSL GPRS NETWORK 5.7% 0.3% 7
AS9304 AS9304HUTCHISON-AS-AP - HGC Global Communications Limited 4.7% 3.5% 19
AS10118 AS10118HTCL-IAS-HK-AP - Hutchison Telephone Company Limited 4.0% 22.3% 27
AS17924 AS17924SMARTONE-MB-AS-AP - SmarTone Mobile Communications Ltd 3.6% 14.5% 26
AS62610 AS62610ZEN-DPS - Zenlayer Inc 3.0% 0.0% 0
AS134972 AS134972IKUUU-AS-AP - IKUUU NETWORK LTD 2.7% <0.1% 1
AS58453 AS58453CMI-INT-HK - China Mobile International Limited 2.6% 0.4% 6

Networks without measurements

54 networks with at least 0.1% of users had no measurements in the last 30 days, about 28.4% of users in total.

ASNNameUsers
AS9231 AS9231IPEOPLESNET-AS-AP - China Mobile Hong Kong Company Limited 6.14%
AS62610 AS62610ZEN-DPS - Zenlayer Inc 2.96%
AS48266 AS48266Catixs - Catixs Ltd 1.53%
AS63888 AS63888DATAWINGLIMITED-AS - DATAWING LIMITED 1.46%
AS401443 AS401443WAP-AC - WAP.AC LLC 0.94%
AS201217 AS201217radishcloud - RADISHCLOUD TECHNOLOGY LLC 0.88%
AS153371 AS153371BACKWAVESLIMITED-AS-AP - BACK WAVES LIMITED 0.85%
AS134196 AS134196ANYUN-INTERNET-TECHNOLOGY-HK-CO-LIMITED - ANYUN INTERNET TECHNOLOGY (HK) CO.,LIMITED 0.85%

Where measurements come from

Each network's share of measurements over the last 30 days. When a few networks account for most of them, the rest can only be judged from scattered data.

Measurements by networkAS4760AS10118AS9908AS17924AS24544Others
  • AS4760HKTIMS-AP - HKT Limited22.8%
  • AS10118HTCL-IAS-HK-AP - Hutchison Telephone Company Limited22.3%
  • AS9908HKCABLE2-HK-AP - HK Cable TV Ltd18.6%
  • AS17924SMARTONE-MB-AS-AP - SmarTone Mobile Communications Ltd14.5%
  • AS24544LAWSCLOUDINFRASTRUCTURELIMITED-AS-HK - Law's Cloud Infrastructure Limited4.8%
  • Others17.0%

6 more networks are not in APNIC's user estimates and account for 0.8% of measurements; most are academic, hosting, or cloud networks.

Daily measurements

Measurements OONI Probe sends from this region each day, split by result. OK means the result matched the control. An anomaly means it did not, which can be blocking but also an unstable connection or a problem with the site itself. Confirmed blocking is an anomaly that matches a known blocking pattern, such as a redirect to a block page. Failure means the test did not complete, for example because the connection dropped midway.

Measurements per dayOKAnomalyConfirmedFailure
05k10k08/2109/2010/192026/08/21 OK 6,1512026/08/21 Anomaly 4172026/08/21 Confirmed 02026/08/21 Failure 4512026/08/22 OK 6,1322026/08/22 Anomaly 3722026/08/22 Confirmed 02026/08/22 Failure 5312026/08/23 OK 4,0322026/08/23 Anomaly 1382026/08/23 Confirmed 02026/08/23 Failure 1942026/08/24 OK 2,8122026/08/24 Anomaly 992026/08/24 Confirmed 02026/08/24 Failure 402026/08/25 OK 3,9982026/08/25 Anomaly 782026/08/25 Confirmed 02026/08/25 Failure 842026/08/26 OK 4,2212026/08/26 Anomaly 742026/08/26 Confirmed 02026/08/26 Failure 3282026/08/27 OK 7,3192026/08/27 Anomaly 2462026/08/27 Confirmed 02026/08/27 Failure 1722026/08/28 OK 3,4902026/08/28 Anomaly 822026/08/28 Confirmed 02026/08/28 Failure 592026/08/29 OK 4,2862026/08/29 Anomaly 1292026/08/29 Confirmed 02026/08/29 Failure 962026/08/30 OK 3,5792026/08/30 Anomaly 792026/08/30 Confirmed 02026/08/30 Failure 602026/08/31 OK 3,6942026/08/31 Anomaly 1092026/08/31 Confirmed 442026/08/31 Failure 1672026/09/01 OK 4,1332026/09/01 Anomaly 792026/09/01 Confirmed 02026/09/01 Failure 682026/09/02 OK 5,0972026/09/02 Anomaly 1722026/09/02 Confirmed 02026/09/02 Failure 762026/09/03 OK 3,6262026/09/03 Anomaly 1082026/09/03 Confirmed 02026/09/03 Failure 742026/09/04 OK 3,7102026/09/04 Anomaly 922026/09/04 Confirmed 02026/09/04 Failure 652026/09/05 OK 6,3112026/09/05 Anomaly 1462026/09/05 Confirmed 02026/09/05 Failure 942026/09/06 OK 5,7622026/09/06 Anomaly 1432026/09/06 Confirmed 02026/09/06 Failure 1222026/09/07 OK 5,5702026/09/07 Anomaly 2442026/09/07 Confirmed 02026/09/07 Failure 1062026/09/08 OK 3,4122026/09/08 Anomaly 1312026/09/08 Confirmed 02026/09/08 Failure 532026/09/09 OK 3,8232026/09/09 Anomaly 2152026/09/09 Confirmed 162026/09/09 Failure 2262026/09/10 OK 3,4462026/09/10 Anomaly 1002026/09/10 Confirmed 02026/09/10 Failure 692026/09/11 OK 6,0692026/09/11 Anomaly 2112026/09/11 Confirmed 32026/09/11 Failure 1172026/09/12 OK 3,5362026/09/12 Anomaly 1072026/09/12 Confirmed 02026/09/12 Failure 1832026/09/13 OK 5,5482026/09/13 Anomaly 872026/09/13 Confirmed 02026/09/13 Failure 952026/09/14 OK 4,2212026/09/14 Anomaly 1352026/09/14 Confirmed 492026/09/14 Failure 1242026/09/15 OK 3,9082026/09/15 Anomaly 772026/09/15 Confirmed 02026/09/15 Failure 482026/09/16 OK 5,6132026/09/16 Anomaly 2522026/09/16 Confirmed 02026/09/16 Failure 832026/09/17 OK 3,8202026/09/17 Anomaly 972026/09/17 Confirmed 02026/09/17 Failure 612026/09/18 OK 1,1122026/09/18 Anomaly 392026/09/18 Confirmed 02026/09/18 Failure 152026/09/19 OK 3,1442026/09/19 Anomaly 552026/09/19 Confirmed 02026/09/19 Failure 562026/09/20 OK 4,4812026/09/20 Anomaly 1002026/09/20 Confirmed 02026/09/20 Failure 632026/09/21 OK 3,3212026/09/21 Anomaly 1002026/09/21 Confirmed 02026/09/21 Failure 612026/09/22 OK 3,5762026/09/22 Anomaly 1512026/09/22 Confirmed 02026/09/22 Failure 662026/09/23 OK 2,3772026/09/23 Anomaly 1182026/09/23 Confirmed 02026/09/23 Failure 472026/09/24 OK 1,5592026/09/24 Anomaly 612026/09/24 Confirmed 82026/09/24 Failure 1692026/09/25 OK 4,5272026/09/25 Anomaly 2472026/09/25 Confirmed 52026/09/25 Failure 3052026/09/26 OK 2,9372026/09/26 Anomaly 832026/09/26 Confirmed 02026/09/26 Failure 792026/09/27 OK 2,5682026/09/27 Anomaly 562026/09/27 Confirmed 02026/09/27 Failure 392026/09/28 OK 2,7622026/09/28 Anomaly 662026/09/28 Confirmed 92026/09/28 Failure 982026/09/29 OK 4,0672026/09/29 Anomaly 652026/09/29 Confirmed 02026/09/29 Failure 1522026/09/30 OK 5,1212026/09/30 Anomaly 1252026/09/30 Confirmed 62026/09/30 Failure 2112026/10/01 OK 4,8052026/10/01 Anomaly 1192026/10/01 Confirmed 72026/10/01 Failure 2662026/10/02 OK 3,0572026/10/02 Anomaly 972026/10/02 Confirmed 92026/10/02 Failure 902026/10/03 OK 2,6442026/10/03 Anomaly 802026/10/03 Confirmed 22026/10/03 Failure 1122026/10/04 OK 2,7762026/10/04 Anomaly 1062026/10/04 Confirmed 72026/10/04 Failure 1252026/10/05 OK 1,4952026/10/05 Anomaly 282026/10/05 Confirmed 42026/10/05 Failure 632026/10/06 OK 3,3942026/10/06 Anomaly 742026/10/06 Confirmed 02026/10/06 Failure 602026/10/07 OK 2,6462026/10/07 Anomaly 1092026/10/07 Confirmed 02026/10/07 Failure 1512026/10/08 OK 2,7852026/10/08 Anomaly 722026/10/08 Confirmed 02026/10/08 Failure 372026/10/09 OK 3,7672026/10/09 Anomaly 882026/10/09 Confirmed 02026/10/09 Failure 722026/10/10 OK 2,7332026/10/10 Anomaly 412026/10/10 Confirmed 02026/10/10 Failure 552026/10/11 OK 3,9212026/10/11 Anomaly 732026/10/11 Confirmed 02026/10/11 Failure 522026/10/12 OK 5,0142026/10/12 Anomaly 1252026/10/12 Confirmed 02026/10/12 Failure 772026/10/13 OK 3,2312026/10/13 Anomaly 1002026/10/13 Confirmed 72026/10/13 Failure 3072026/10/14 OK 4,1552026/10/14 Anomaly 2012026/10/14 Confirmed 52026/10/14 Failure 2172026/10/15 OK 4,7682026/10/15 Anomaly 1502026/10/15 Confirmed 32026/10/15 Failure 1512026/10/16 OK 2,5962026/10/16 Anomaly 1042026/10/16 Confirmed 42026/10/16 Failure 2022026/10/17 OK 4,0962026/10/17 Anomaly 1532026/10/17 Confirmed 02026/10/17 Failure 2742026/10/18 OK 5,5392026/10/18 Anomaly 2122026/10/18 Confirmed 72026/10/18 Failure 2722026/10/19 OK 3,3472026/10/19 Anomaly 1402026/10/19 Confirmed 102026/10/19 Failure 120
Networks measured per day
0102008/2109/2010/192026/08/21 Networks measured 92026/08/22 Networks measured 102026/08/23 Networks measured 102026/08/24 Networks measured 92026/08/25 Networks measured 92026/08/26 Networks measured 122026/08/27 Networks measured 82026/08/28 Networks measured 82026/08/29 Networks measured 102026/08/30 Networks measured 72026/08/31 Networks measured 132026/09/01 Networks measured 92026/09/02 Networks measured 122026/09/03 Networks measured 102026/09/04 Networks measured 72026/09/05 Networks measured 62026/09/06 Networks measured 82026/09/07 Networks measured 72026/09/08 Networks measured 42026/09/09 Networks measured 62026/09/10 Networks measured 92026/09/11 Networks measured 112026/09/12 Networks measured 62026/09/13 Networks measured 72026/09/14 Networks measured 102026/09/15 Networks measured 62026/09/16 Networks measured 72026/09/17 Networks measured 72026/09/18 Networks measured 52026/09/19 Networks measured 62026/09/20 Networks measured 72026/09/21 Networks measured 62026/09/22 Networks measured 52026/09/23 Networks measured 62026/09/24 Networks measured 82026/09/25 Networks measured 72026/09/26 Networks measured 52026/09/27 Networks measured 52026/09/28 Networks measured 112026/09/29 Networks measured 72026/09/30 Networks measured 82026/10/01 Networks measured 92026/10/02 Networks measured 92026/10/03 Networks measured 92026/10/04 Networks measured 112026/10/05 Networks measured 92026/10/06 Networks measured 102026/10/07 Networks measured 92026/10/08 Networks measured 82026/10/09 Networks measured 92026/10/10 Networks measured 72026/10/11 Networks measured 92026/10/12 Networks measured 102026/10/13 Networks measured 112026/10/14 Networks measured 102026/10/15 Networks measured 112026/10/16 Networks measured 102026/10/17 Networks measured 102026/10/18 Networks measured 92026/10/19 Networks measured 7

Users covered

The share of this region's internet users, by APNIC's estimates, on networks that had at least one measurement that day. One measurement is enough to count, so this is an upper bound rather than a sign that every network is measured enough.

Users covered (%)
05010008/2109/2010/192026/08/21 Users covered 41.62026/08/22 Users covered 41.72026/08/23 Users covered 41.92026/08/24 Users covered 51.92026/08/25 Users covered 43.42026/08/26 Users covered 49.52026/08/27 Users covered 39.32026/08/28 Users covered 42.82026/08/29 Users covered 40.22026/08/30 Users covered 41.12026/08/31 Users covered 55.32026/09/01 Users covered 42.92026/09/02 Users covered 45.92026/09/03 Users covered 51.92026/09/04 Users covered 45.52026/09/05 Users covered 42.82026/09/06 Users covered 42.82026/09/07 Users covered 27.52026/09/08 Users covered 25.92026/09/09 Users covered 37.62026/09/10 Users covered 49.12026/09/11 Users covered 37.12026/09/12 Users covered 39.02026/09/13 Users covered 43.32026/09/14 Users covered 47.82026/09/15 Users covered 31.92026/09/16 Users covered 29.32026/09/17 Users covered 31.32026/09/18 Users covered 26.62026/09/19 Users covered 33.72026/09/20 Users covered 39.32026/09/21 Users covered 31.12026/09/22 Users covered 29.52026/09/23 Users covered 26.62026/09/24 Users covered 31.82026/09/25 Users covered 28.22026/09/26 Users covered 27.52026/09/27 Users covered 26.52026/09/28 Users covered 31.92026/09/29 Users covered 32.12026/09/30 Users covered 26.52026/10/01 Users covered 38.32026/10/02 Users covered 33.82026/10/03 Users covered 43.72026/10/04 Users covered 36.02026/10/05 Users covered 29.12026/10/06 Users covered 39.22026/10/07 Users covered 43.42026/10/08 Users covered 31.92026/10/09 Users covered 45.52026/10/10 Users covered 27.82026/10/11 Users covered 43.22026/10/12 Users covered 43.32026/10/13 Users covered 48.22026/10/14 Users covered 38.42026/10/15 Users covered 44.42026/10/16 Users covered 35.42026/10/17 Users covered 48.62026/10/18 Users covered 33.42026/10/19 Users covered 29.4

Steady coverage

Coverage counts a network as soon as it has a single measurement, even if it was measured on only a day or two. Steady coverage counts only networks measured on at least 15 of the last 30 days. The wider the gap between the two, the more users sit on networks with only scattered measurements.

Users covered 66.6% Measured on at least one day
Users steadily covered 31.2% Measured on at least 15 days

Messaging app tests

OONI Probe tests whether the servers of Signal, WhatsApp and Telegram can be reached. An anomaly only means the result differed from what was expected, which an unstable connection can also cause. An anomaly rate close to 100% usually means the app is blocked there. Rates are left out when there were fewer than 30 measurements in the last 30 days.

Anomaly rate in Hong Kong, last 30 days
  • Signal1.0%
  • WhatsApp0.7%
  • Telegram0.5%

Up to 731 measurements per app

Across regions

Users covered by networks measured in the last 30 days. Each small chart shows the daily coverage over the last 60 days on one shared scale, so the levels compare directly. Open a region for its full charts.

TaiwanTW 88.6% Hong KongHK 66.6% MacauMO 85.9% JapanJP 84.4% South KoreaKR 94.5% SingaporeSG 82.1% VietnamVN 99.0% ThailandTH 97.5% MalaysiaMY 96.5% IndonesiaID 80.6% PhilippinesPH 89.9% IndiaIN 91.6%

Small-chart scale: 0% to 100%.

RegionCoveredSteady
TaiwanTW 88.6% 54.1%
Hong KongHK 66.6% 31.2%
MacauMO 85.9% 0.0%
JapanJP 84.4% 72.9%
South KoreaKR 94.5% 81.8%
SingaporeSG 82.1% 35.2%
VietnamVN 99.0% 93.9%
ThailandTH 97.5% 48.9%
MalaysiaMY 96.5% 95.1%
IndonesiaID 80.6% 69.7%
PhilippinesPH 89.9% 86.9%
IndiaIN 91.6% 86.3%

Messaging app anomaly rates across regions

Anomaly rates over the last 30 days, with rates above half highlighted. The last rows are reference regions without their own pages, which have records of blocking to compare against.

RegionSignalWhatsAppTelegram
TaiwanTW 0.9%1.8%1.6%
Hong KongHK 1.0%0.7%0.5%
MacauMO –––
JapanJP 0.5%0.7%0.5%
South KoreaKR 0.2%2.7%2.4%
SingaporeSG 1.0%0.3%1.2%
VietnamVN 0.1%0.2%0.5%
ThailandTH 0.5%0.2%0.3%
MalaysiaMY <0.1%0.1%1.1%
IndonesiaID 3.0%0.5%0.7%
PhilippinesPH 0.2%0.1%2.6%
IndiaIN 1.9%0.4%1.7%
Reference regions
ChinaCN 98.6%98.2%99.1%
MyanmarMM 97.3%98.8%0.4%
CambodiaKH 29.6%35.4%2.4%
PakistanPK 97.2%0.6%99.4%
BangladeshBD 0.5%0.2%0.3%

Glossary

  • OONI Probe: an open-source app from OONI that tests from a phone or computer whether websites and services can be reached, with the results published in OONI's database
  • Measurement: one result of OONI Probe testing one website or service. A single run of the website test produces hundreds, so a high count does not mean many people ran tests
  • Autonomous system (ASN): a group of networks run by one organisation under one routing policy, such as a telecom, a university or a cloud provider, each with its own AS number. A telecom's fixed and mobile networks are often separate ASNs
  • Network names: copied from APNIC and RIPE NCC registry records and often abbreviated. In Taiwan, for example, HINET is Chunghwa Telecom's fixed network and EMOME-NET its mobile network
  • APNIC: the Asia Pacific Network Information Centre, which allocates IP addresses and ASNs in the region and estimates each network's users from sampled web ads
  • Users, Measured (%): Users is the network's share of the region's internet users, and Measured is its share of the region's measurements. The further apart the two are, the more the measurements are out of proportion to the users
  • Days: how many of the last 30 days had any measurements
  • Users covered: the share of the region's internet users on networks with measurements, where a single measurement is enough to count
  • Users steadily covered: counts only networks measured on at least 15 of the last 30 days
  • Top two networks: the share of measurements from the two most measured networks; lower means more spread out
  • pp (percentage points): the difference between two percentages. Coverage going from 94.6% to 93.9% is written as −0.7 pp
  • OK, anomaly, confirmed, failure: OK means the result matched the control. An anomaly means it did not, which can be blocking or an unstable connection. Confirmed blocking is an anomaly matching a known blocking pattern. A failure is a test that did not complete
  • Anomaly rate (messaging apps): the share of the app's measurements that were anomalies. Close to 100% usually means the app is blocked there
  • Reference region: a region without its own page, listed in a comparison table for contrast

How the numbers are calculated

  • Measurement counts come from OONI's aggregation API, across all tests, grouped by the day each measurement started (UTC)
  • Users per network are APNIC's estimates, derived from ad-based sampling and updated weekly
  • Network names come from the APNIC and RIPE NCC ASN name lists
  • A single measurement is enough for a network to count as measured; whether it is measured enough shows in its share of measurements and the number of days
  • Steady coverage counts only networks measured on at least 15 of the last 30 days
  • Messaging app results also come from the OONI aggregation API, counted by region; the reference regions at the end of the table have no pages of their own
  • Days without measurements are left as gaps in the charts rather than filled in

Help fill the gaps

  • Install OONI Probe and run tests on your own connection, on a phone or a computer. Mobile data and home broadband are usually different ASNs, so testing on both helps
  • The community's OONI Run link 10328 sets everything up in one tap; see the OONI Run v2 guide for the risks to consider before installing and what the list contains
  • Maintaining the OONI Test List: the list decides which websites get tested, and help with Taiwan's tw.csv is welcome
  • ASN Coverage: a command-line tool that downloads raw measurements and breaks them down by network type and blocking method

Reading the measurement data