VC
Vint Cerf
Crocker's high-school friend and UCLA colleague. He later co-designed TCP/IP.
~/arpanet
The memo that became the Internet's rulebook
In 1969 a handful of graduate students were asked to work out how computers on a brand-new network should talk to each other. Steve Crocker wrote down their first ideas, and started a tradition that still governs the Internet.
Seconds since the first ARPANET message, UCLA → SRI, 29 Oct 1969:
—
$ head -n 20 rfc1.txt Network Working Group Steve Crocker Request for Comments: 1 UCLA 7 April 1969 Title: Host Software Author: Steve Crocker Installation: UCLA Date: 7 April 1969 INTRODUCTION The software for the ARPA Network exists partly in the IMPs and partly in the respective HOSTs.
RFC 1 · Host Software
A deliberately humble beginning
No one had been put in charge, and no one wanted to seem to claim authority. So the group's notes were offered as Requests for Comments: ideas put forward for discussion, not decrees.
Crocker's RFC 1, dated 7 April 1969, described how host computers would work with the network's Interface Message Processors (IMPs). Bill Duvall of SRI wrote the companion RFC 2. Jon Postel later became the long-serving RFC Editor.
“Very little of what is here is firm, and reactions are expected.” Steve Crocker, RFC 1 · 7 April 1969
1968 – 1983
How the first protocols came together
-
AUGUST 1968
Santa Barbara kickoff
ARPA brings together representatives of the first host sites to start planning how their computers will use the new network.
-
1969
The Network Working Group forms
Crocker helps organize the graduate students and researchers who will design the host protocols. It is the forerunner of today's IETF.
-
7 APRIL 1969
RFC 1 is issued
“Host Software” · Steve Crocker, UCLAThe first Request for Comments begins the document series that still defines Internet standards.
-
29 OCTOBER 1969
The first message
Charley Kline at UCLA sends the first ARPANET message to SRI, where Bill Duvall receives it.
-
EARLY 1970s
The Network Control Program
Crocker leads the design teamNCP, the ARPANET's host-to-host protocol, standardizes the network interface and makes it far easier for new sites to connect.
-
1972
International Network Working Group
Crocker forms INWG to take networking work across borders, then turns his own research toward artificial intelligence.
-
1 JANUARY 1983
Handing over to TCP/IP
After more than a decade of service, NCP gives way to TCP/IP. The protocols change; the RFC process carries on.
The Network Working Group
Among the people in the room
The ARPANET's host protocols were the work of a small group of young engineers. Crocker led the UCLA team.
JP
Jon Postel
UCLA team member and later the long-serving RFC Editor, who wrote nearly 200 RFCs himself.
BD
Bill Duvall
SRI researcher, author of RFC 2, and receiver of the first ARPANET message.
CK
Charley Kline
UCLA programmer who sent the first ARPANET message on 29 October 1969.
The legacy
From one memo to 9,000+
The series Crocker began now numbers well over nine thousand documents, published by the IETF and its RFC Editor.
The Internet's core protocols were all published as RFCs: IP, TCP, the Domain Name System, email, the Web's HTTP and modern encryption. Each went through the same open process of proposal, comment and refinement that began in 1969.
RFC 791 · IP
RFC 793 · TCP
RFC 1035 · DNS
RFC 5321 · SMTP
RFC 8446 · TLS 1.3
RFC 9000 · QUIC
The hardware
The Interface Message Processor
Every ARPANET site connected through an IMP, a ruggedized minicomputer that handled packet switching.
RFC 1 was about the other half of the problem: the software on each site's host computer that talked to its IMP and, through it, to every other host on the network. Working out that host-to-host conversation was the Network Working Group's job.
Read RFC 1 at the RFC Editor
“We felt that the network services should be expandable.”