If IPv4 were land
- ID
- 41127
- Status
- summarized
- Published
- 29 Oct 2026, 8:19 PM
- Fetched
- 30 Oct 2026, 5:18 PM
- Provider
- Hacker News
- Category
- dev-community
- Original URL
- https://iptrust.co/blog/if-ipv4-were-land
- Source URL
- https://hnrss.org/best
Summary
- Score
- 5.5
- Created
- 30 Oct 2026, 5:19 PM
- Tags
- Audience
- developerssaas_founders
What happened
IPTrust redrew the world map so each country's area is proportional to its share of the 3.6 billion locatable IPv4 addresses, against roughly 135 million km² of land outside Antarctica, using its October 2026 geolocation database. Countries qualify only if they have at least 100,000 IPv4 addresses and 20,000 km² of land, and about 70 small territories with no outline at the 1:110m Natural Earth resolution are dropped — roughly 52 million addresses, mostly Hong Kong, Singapore and Mauritius. The Hacker News thread drew 282 points and 395 comments.
Why it matters
The chart's raw input is a commercial IP geolocation database, which is the same class of data behind IP-based geo-routing, fraud scoring and regional pricing — so the 'where is this user' answer your app gets is a vendor inference, not a fact, and this map shows how lopsided that inference is by allocation rather than population. The explicit exclusion of Hong Kong, Singapore and Mauritius (~52M addresses between them) is the practical warning: dense, high-allocation territories fall out of coarse geographic models, so don't assume a country-level IP lookup covers your Southeast Asian users cleanly.
Discussion angle
If the same exercise were run on IPv6, the map would be meaningless — so ask the room where they still hard-depend on IPv4-derived signals (geo-blocking, pricing, fraud rules, rate limits) and what breaks when CGNAT or a VPN makes the address a poor proxy for the user.