March 20, 1970 — The First ARPANET RFC Series Begins to Take Shape

On March 20, 1970, a small set of documents appeared that would later become part of the foundation of the modern Internet: RFC 36, RFC 37, RFC 38, and RFC 39. These were part of the early Request for Comments (RFC) series — an experiment in open collaboration that changed how protocols are designed forever.

At the time, ARPANET was still young. There was no “Internet standard” in the modern sense. Instead, there was a group of researchers trying to figure things out in real time.

What RFC was back then

The RFC series started in 1969 thanks to Steve Crocker and a small group of engineers working on ARPANET. Instead of writing strict formal specifications, they shared ideas as working notes.

The name “Request for Comments” was intentional. These documents were not final. They were invitations to discuss.

By March 1970, this process was already working — and RFC 36–39 are a good snapshot of that early phase.

What RFC 36–39 were about

These documents were not “standards” yet. They were closer to engineering conversations.

• RFC 36 continued discussions about host-to-host communication — the core problem of ARPANET
• RFC 37 and RFC 38 included feedback and clarifications between different teams working on network nodes
• RFC 39 reflected ongoing coordination — people reacting to each other’s ideas, sometimes disagreeing, sometimes refining concepts

At this stage, even basic things were still unclear: how machines should talk to each other, how to structure messages, how to manage connections.

Why this matters

Today, protocols like TCP/IP, HTTP, and DNS feel solid and inevitable. But RFC 36–39 remind us that:

• The Internet was built through discussion, not top-down design
• Early engineers shared unfinished ideas openly
• Standards emerged gradually from real usage, not theory

This process became one of the most important traditions in computing. Even today, the RFC system is still used by the Internet Engineering Task Force (IETF).

The most interesting part

What stands out when you read these early RFCs is how human they are. They include informal language, direct responses to other engineers, and open uncertainty.

In some cases, you can literally see the network being invented line by line.

There was no guarantee it would work. No guarantee it would scale. Just people trying things and writing them down.

Early RFC archive · RFC 36 · RFC 37 · RFC 38 · RFC 39