Editing Datagram (section)

== History ==
{{See also|Packet switching#History}}<!--consider merging overlapping content at some point-->

In the early 1970s, the term ''datagram'' was created by combining the words ''data'' and ''telegram'' by the [[CCITT]] rapporteur on packet switching,<ref>{{cite journal|title=The CCITT studies packet switching as part of public data network development|date=1975 |doi=10.1145/1024916.1024918 |url=https://www.deepdyve.com/lp/association-for-computing-machinery/the-ccitt-studies-packet-switching-as-part-of-public-data-network-TU05NzohPP |last1=Bothner |first1=H. |journal=ACM SIGCOMM Computer Communication Review |volume=5 |issue=2 |pages=9–17 }}</ref> [[Halvor Bothner-By]].<ref>{{cite journal|title=X.25 virtual circuits — Transpac in France — Pre-Internet data networking|author1=Rémi Després|journal=IEEE Communications Magazine|date=November 2010|issue=10|volume=48| doi=10.1109/MCOM.2010.5621965 |url=https://ieeexplore.ieee.org/document/5621965|url-access=subscription}}</ref><ref>{{cite web|url=http://old.open-root.eu/la-doc/histoire-des-reseaux/le-datagramme/|url-status=dead|archive-url=https://web.archive.org/web/20190228065633/http://old.open-root.eu/la-doc/histoire-des-reseaux/le-datagramme/|archive-date=2019-02-28|title=Comment j'ai inventé le Datagramme|language=fr}}</ref> While the word was new, the concept had already a long history.

In 1964, [[Paul Baran]] described, in a [[RAND Corporation]] report, a hypothetical military network having to resist a nuclear attack. Small standardized ''message blocks'', bearing source and destination addresses, were [[Store and forward|stored and forwarded]] in computer nodes of a highly redundant meshed computer network. Baran wrote: "The network user who has called up a ''virtual connection'' to an end station and has transmitted messages ... might also view the system as a black box providing an apparent circuit connection".<ref>{{cite web|title=On distributed communications networks|url=http://pages.cs.wisc.edu/~akella/CS740/F08/740-Papers/Bar64.pdf|url-status=dead|archive-url=https://web.archive.org/web/20161026145448/https://pages.cs.wisc.edu/~akella/CS740/F08/740-Papers/Bar64.pdf|archive-date=2016-10-26}}</ref> The concept of what we now call a [[virtual circuit]] appears in the design,<ref name=":0">{{Cite web |last=Pelkey |first=James L. |date=May 27, 1988 |title=Interview of Donald Davies |url=http://archive.computerhistory.org/resources/access/text/2017/11/102738594-05-01-acc.pdf |page=7}}</ref> although no network was built.

In 1967, [[Donald Davies]] published a seminal article in which he introduced the [[Network packet|''packet'']] and ''[[packet switching]]''. His proposed core network is similar to the one proposed by Paul Baran though developed independently. He assumes that "all users of the network will provide themselves with some kind of error control". His target is a "common-carrier communication network". To support remote access to computer services by user terminals, which at that time were transmitted character by character, he included, at the network periphery, interface computers that convert character flows into packet flows and vice versa.<ref>{{cite web|title=A digital communication network for computers giving rapid response at remote terminals|url=https://people.mpi-sws.org/~gummadi/teaching/sp07/sys_seminar/how_did_erope_blow_this_vision.pdf |archive-url=https://ghostarchive.org/archive/20221009/https://people.mpi-sws.org/~gummadi/teaching/sp07/sys_seminar/how_did_erope_blow_this_vision.pdf |archive-date=2022-10-09 |url-status=live}}</ref> Davies wrote: "we were really rather against the virtual circuit, because we believed that a communication network should only concern itself with packets, and that any protocols involved in assembling these packets should be done end-to-end, between the customers themselves."<ref name=":0" />
 
In 1970, Lawrence Roberts and Barry D. Wessler published an article about [[ARPANET]], the first multi-node packet-switching network.<ref>{{cite book|chapter-url=https://www.researchgate.net/publication/234815171|chapter=Computer network development to achieve resource sharing|author1=Lawrence Roberts|author2=Barry D. Wessler|title=Proceedings of the May 5-7, 1970, spring joint computer conference on - AFIPS '70 (Spring) |year=1970|page=543 |doi=10.1145/1476936.1477020|s2cid=9343511 }}</ref> An accompanying paper described its switching nodes (the [[Interface Message Processor|IMPs]]) and its packet formats.<ref>{{cite book|author1=Frank E Heart|author2=R E Kahn|author3=Severo M Ornstein|author4=William R Crowther|author5=David C Walden|title=Proceedings of the May 5-7, 1970, spring joint computer conference on - AFIPS '70 (Spring) |chapter-url=https://dl.acm.org/doi/10.1145/1476936.1477021|doi=10.1145/1476936.1477021|chapter=The interface message processor for the ARPA computer network|year=1970 |pages=551–567 |isbn=978-1-4503-7903-8 |s2cid=9647377 }}</ref> The network core performed datagram switching as in Baran's and Davies' model, but the service offered to hosts by the network was [[connection oriented]].<ref name="BBN1822">{{cite web |date=January 2014 |title=INTERFACE MESSAGE PROCESSOR Specifications for the Innterconnection of a Host |url=http://www.bitsavers.org/pdf/bbn/imp/BBN1822_Jan1976.pdf |quote="three parameters uniquely specify a connection between source and destination Hosts." "The destination IMP returns a positive acknowledgment for receipt of the message to the source IMP, which in turn passes this acknowledgment to the source Host." "Each link is unidirectional and is controlled by the network so that no more than one message at a time may be sent over it."}}</ref><ref name="Pelkey8.4b">{{cite book |last=Pelkey |first=James |title=Entrepreneurial Capitalism and Innovation: A History of Computer Communications 1968–1988 |chapter=8.4 Transmission Control Protocol (TCP) 1973-1976 |quote=Arpanet had its deficiencies, however, for it was neither a true datagram network nor did it provide end-to-end error correction. |chapter-url=https://historyofcomputercommunications.info/section/8.4/Transmission-Control-Protocol-(TCP)-1973-1976/}}</ref> A reliable message transfer service was thus offered to user computers, thus greatly simplifying the network design. This made the ARPANET what would come to be called a [[virtual circuit]] network.<ref name="LP2012">{{cite web |date=April 2012 |title=An Interview with LOUIS POUZIN Conducted by Andrew L. Russell |url=https://conservancy.umn.edu/bitstream/handle/11299/155666/oh416lp.pdf?sequence=3&isAllowed=y |quote="Arpanet was virtual circuit." "essentially a virtual circuit service using internal datagram"}}</ref>

Roberts presented the idea of packet switching to the communication professionals and faced anger and hostility. Before ARPANET was operating, they argued that the router buffers would quickly run out. After the ARPANET was operating, they argued packet switching would never be economic without the government subsidy. Baran faced the same rejection and thus failed to convince the military to construct a packet-switching network.<ref>{{Citation |last=Roberts |first=L. |title=The arpanet and computer networks |date=1988-01-01 |url=https://doi.org/10.1145/61975.66916 |work=A history of personal workstations |pages=141–172 |access-date=2023-11-30 |place=New York, NY, USA |publisher=Association for Computing Machinery |doi=10.1145/61975.66916 |isbn=978-0-201-11259-7}}</ref>

In 1973, [[Louis Pouzin]] presented his design for [[CYCLADES]], the first large-scale network implementing the pure Davies datagram model.<ref name="Pouzen">{{cite web|first=Louis |last=Pouzen |title=Presentation and major design aspects of the Cyclades network |url=http://rogerdmoore.ca/PS/CYCLB.html |archive-url=https://web.archive.org/web/20070927205826/http://rogerdmoore.ca/PS/CYCLB.html |archive-date=September 27, 2007}}</ref> The CYCLADES team has thus been the first to tackle the highly complex problem of providing user applications a reliable [[virtual circuit]] service<ref>{{cite IETF|rfc=1379|title=Extending TCP for transactions -- Concepts}}</ref> while using the [[end-to-end principle]] in a network service known to possibly produce non-negligible datagram losses and reordering.<ref name="Bennett2009">{{cite web |last1=Bennett |first1=Richard |date=September 2009 |title=Designed for Change: End-to-End Arguments, Internet Innovation, and the Net Neutrality Debate |url=https://www.itif.org/files/2009-designed-for-change.pdf |access-date=11 September 2017 |publisher=Information Technology and Innovation Foundation |pages=7, 11}}</ref> Although Pouzin's concern "in a first stage is not to make breakthrough [sic] in packet switching technology, but to build a reliable communications tool for Cyclades",<ref name="Pouzen" /> two members of his team, [[Hubert Zimmerman]] and [[Gérard Le Lann]], made significant contributions to the design of Internet's TCP that [[Vint Cerf]], its main designer, acknowledged.{{Ref RFC|675}}

In 1981, the Defense Advanced Research Projects Agency ([[DARPA]]) issued the first specification the [[Internet Protocol]] (IP). It introduced a major evolution of the datagram concept: ''[[IP fragmentation|fragmentation]].''{{Ref RFC|791}} With fragmentation, some parts of the global network may use large packet size (typically [[local area network]]s to minimize processing overhead), while some others may impose smaller packet sizes (typically [[wide area network]]s to minimize response time). Network nodes may fragment a datagram into several smaller packets.

In 1999, the [[Internet Engineering Task Force]] (IETF) sanctioned the use of the already largely deployed [[network address translation]] (NAT) whereby each public address can be shared by several private devices.{{Ref RFC|2663}} With it, the forthcoming [[IPv4 address exhaustion|Internet Address exhaustion]] was delayed, leaving enough time to introduce [[IPv6]], the new generation of Internet Protocol supporting longer addresses. The initial principle of full [[End-to-end principle|end to end]] network transparency to datagrams was for this relaxed: NAT nodes had to manage per-connection states, making them in part [[connection oriented]].

In 2015, the [[IETF]] upgraded its ''informational'' 1998 RFC 2309<ref>{{Cite web |url=https://datatracker.ietf.org/doc/html/rfc2309 |title=Recommendations on Queue Management and Congestion Avoidance in the Internet |last=Zhang |first=Lixia |last2=Partridge |first2=Craig |last3=Shenker |first3=Scott |last4=Wroclawski |first4=John T. |last5=Ramakrishnan |first5=K. K. |last6=Peterson |first6=Larry |last7=Clark |first7=David D. |last8=Minshall |first8=Greg |last9=Crowcroft |first9=Jon |date=1998-04-01 |publisher=Internet Engineering Task Force |id=RFC 2309}}</ref> that datagram switching nodes perform [[active queue management]], to make it a  stronger and more detailed ''[[best current practice]]'' recommendation through the publication of RFC 7567. While the initial datagram queueing model was simple to implement and needed no more tuning than queue lengths, support of more sophisticated and parametrized mechanisms were found necessary "to improve and preserve Internet performance" ([[Random Early Detection|RED]], [[Explicit Congestion Notification|ECN]] etc.). Further research on the subject was also called for, with a list of identified items.{{Ref RFC|7567}}