Table of Contents
Paul Mockapetris
Who is Paul Mockapetris?
Paul Mockapetris is the inventor of the Domain Name System (DNS). Before his work, if you wanted to connect to another computer on the internet, you had to know its numerical IP address. There was a single text file called HOSTS.TXT, maintained by hand at Stanford Research Institute, that mapped names to numbers. Every computer on the network had to download a fresh copy regularly. By the early 1980s, this system was falling apart. The file was getting too big, updates were too slow, and name collisions were becoming a real problem.
Mockapetris designed a distributed, hierarchical naming system that solved all of these problems at once. He published the specification in 1983 (RFC 882 and RFC 883), and it has been running the internet ever since. Every time you type a domain name into your browser, his system translates it into an IP address. It happens billions of times per day, usually in under 50 milliseconds, and most people have no idea it exists.
Early Life and Education
Paul Mockapetris was born in 1948 in Boston, Massachusetts. He studied electrical engineering at MIT, earning his bachelor's degree, and went on to get a Ph.D. in Information and Computer Science from the University of California, Irvine in 1982.
While at UC Irvine, he worked at the Information Sciences Institute (ISI) at the University of Southern California. ISI was one of the key research labs in the early internet, and it was there that Mockapetris designed DNS.
The Problem DNS Solved
In the early ARPANET, the naming system was simple: one file, one authority, one list. Elizabeth Feinler at Stanford's Network Information Center (the NIC) maintained HOSTS.TXT. If you wanted to add a name, you called or emailed her team. They would update the file, and everyone would download the new version.
This worked fine for a few hundred computers. By 1983, there were thousands, and the system was breaking down:
- The file had to be updated manually, which meant delays of days
- Every computer had to download the whole file, even if only one entry changed
- Name conflicts were constant because there was no structure to prevent them
- The single point of control could not scale
Several people proposed solutions. Jon Postel, who managed much of the early internet's infrastructure, asked Mockapetris to evaluate the proposals and pick one. Mockapetris looked at them, decided none of them were good enough, and designed his own.
How DNS Works
The system Mockapetris built is deceptively simple in concept but powerful in practice:
- Domain names are organized in a hierarchy: .com, .org, .net at the top, then second-level domains (google.com), then subdomains (mail.google.com)
- No single server knows everything. Root servers point to TLD servers, which point to authoritative servers for each domain
- Results are cached at every level, so the same lookup does not have to travel the full chain every time
- The system is distributed globally, with no single point of failure
He wrote the first DNS implementation himself, called JEEVES, running on a DEC PDP machine at ISI. He also wrote the Unix implementation called BIND (Berkeley Internet Name Domain), which became the reference implementation and is still one of the most widely used DNS servers in the world.
RFC 882 and RFC 883, published in November 1983, defined the system. They were later superseded by RFC 1034 and RFC 1035 in 1987, which remain the core DNS specification to this day.
Career After DNS
After creating DNS, Mockapetris continued working in networking. He served as program manager at DARPA, where he funded research in networking and security. He was chairman of Nominum, a DNS infrastructure company, and has served on the boards of various internet governance organizations.
He has also been involved in DNS security extensions (DNSSEC), which add cryptographic authentication to DNS responses to prevent spoofing and cache poisoning attacks.
Notable Awards
| Year | Award |
|---|---|
| 2003 | ACM SIGCOMM Award |
| 2005 | IEEE Internet Award |
| 2012 | Inducted into Internet Hall of Fame |
| 2019 | IEEE Medal for Internet Technology |
Legacy
DNS is one of those systems that is so fundamental, so reliable, and so invisible that people forget it exists. Every domain name resolution, every website visit, every email delivery depends on it. The system handles over a trillion queries per day across the global network. And the core design has not changed since 1983. The hierarchy, the delegation, the caching, the distributed authority: it all still works exactly the way Mockapetris drew it up over forty years ago.
When you type a domain into a tool like ipwhois.net and get back registration data, nameservers, and IP addresses, every step of that process starts with DNS doing its job silently in the background.
References
- Paul Mockapetris, Wikipedia
- RFC 882: Domain Names, Concepts and Facilities (1983)
- RFC 1034/1035: Domain Names, Implementation and Specification (1987)
- Internet Hall of Fame, Paul Mockapetris
Sources
Information compiled from Wikipedia, IETF RFC archives, Internet Hall of Fame, ISI/USC records, and networking history resources up to 2026.