Editing CSNET (section)

==History==
<!-- Deleted image removed: [[File:CSNET Geographic Map 1987.jpg|thumb|right|CSNET Geographic Map circa 1987]] -->

By 1986 about 150 computers connected to [[ARPANET]], and about 2000 computers to the larger Arpa [[Internet]]. Membership of ARPANET was restricted to universities and companies with [[United States Department of Defense]] contracts. Other commercial, educational, and nonprofit organizations, ineligible for membership, wanted to interchange information with them.<ref name="fisher19880425">{{Cite magazine |last=Fisher |first=Sharon |date=1988-04-25 |title=The Largest Computer Network |url=https://books.google.com/books?id=pj0EAAAAMBAJ&pg=PA57#v=onepage&q&f=false |access-date=2025-04-17 |magazine=InfoWorld |pages=S7-S10}}</ref>

[[Lawrence Landweber]] at the [[University of Wisconsin–Madison]] prepared the original CSNET proposal, on behalf of a consortium of universities ([[Georgia Tech]], [[University of Minnesota]], [[University of New Mexico]], [[University of Oklahoma]], [[Purdue University]], [[University of California, Berkeley]], [[University of Utah]], [[University of Virginia]], [[University of Washington]], [[University of Wisconsin–Madison|University of Wisconsin]], and [[Yale University]]). The US [[National Science Foundation]] (NSF) requested a review from [[David J. Farber]] at the [[University of Delaware]]. Farber assigned the task to his graduate student Dave Crocker who was already active in the development of [[email|electronic mail]].<ref>{{Cite web |title=Impact of Email Work at The Rand Corporation in the mid-1970s |author=Dave Crocker |date=August 18, 2008 |url=http://bbiw.net/articles/rand-email.pdf |accessdate=September 30, 2011 |archive-url=https://web.archive.org/web/20120510112529/http://bbiw.net/articles/rand-email.pdf |archive-date=May 10, 2012 |url-status=dead }}</ref> The project was deemed interesting but in need of significant refinement. The proposal eventually gained the support of [[Vinton Cerf]] and [[DARPA]]. In 1980, the NSF awarded $5 million to launch the network. It was an unusually large project for the NSF at the time.<ref>{{Cite journal |title=History and overview of CSNET |author=Douglas Comer |journal=Communications |date=October 1983 |volume=26 |pages=747–753 |number=10 |doi=10.1145/358413.358423 |s2cid=11943330 |author-link=Douglas Comer |doi-access=free }}</ref>
A stipulation for the award of the contract was that the network needed to become self-sufficient by 1986.<ref name="nsf"/>

The first management team consisted of Landweber (University of Wisconsin), Farber (University of Delaware), [[Peter J. Denning]] ([[Purdue University]]), [[Anthony C. Hearn]] ([[RAND Corporation]]), and Bill Kern from the NSF.<ref>{{Cite book |author1=Peter J. Denning |author2=Anthony Hearn |author3=C. William Kern |title=Proceedings of the symposium on Communications Architectures & Protocols - COMM '83 |chapter=History and overview of CSNET |publisher=[[SIGCOMM]], [[Association for Computing Machinery]] |date=April 1983 |volume=13 |number=2 |page=138 |doi=10.1145/1035237.1035267 |isbn=978-0-89791-089-7 |s2cid=28021088 |url=http://www.isoc.org/internet/history/documents/Comm83.pdf |access-date=2011-09-30 |archive-url=https://web.archive.org/web/20111026035120/http://www.isoc.org/internet/history/documents/Comm83.pdf |archive-date=2011-10-26 |url-status=dead |author1-link=Peter J. Denning }}</ref> Once CSNET was fully operational, the systems and ongoing network operations were transferred to a team led by Richard Edmiston<ref>{{Cite journal|last1=Partridge|first1=Craig|last2=Blumenthal|date=March 2006|title=Data Networking at BBN|journal=IEEE Annals of the History of Computing|volume=28|issue=1|pages=63|doi=10.1109/MAHC.2006.7|s2cid=16881178}}</ref> at [[BBN Technologies|Bolt Beranek and Newman]] (BBN) of [[Cambridge, Massachusetts]] by 1984.<ref>{{Cite web |title=CSNET Transition Plan Bulletin #1 |author=Rick Adrion |date=October 5, 1983 |work=email message |publisher=National Science Foundation |url=http://www.rfc-editor.org/in-notes/museum/csnet-transition-bulletin.n1.1 |accessdate=September 30, 2011}}</ref>

The Purdue team, consisting of Peter Denning, Douglas Comer, and Paul McNabb, was responsible for designing and building the kernel interfaces that would allow sites outside of the ARPANET infrastructure to connect via public [[X.25]] networks, such as [[Telenet]]. The mechanism allowed systems with TCP/IP network stacks to use an X.25 network device, with IP datagrams being sent through dynamically allocated X.25 sessions. Purdue and other sites with ARPANET access would act as gateways into the ARPANET, allowing non-ARPANet sites to have email, telnet, ftp, and other forms of network access directly into the ARPANET.

By 1981, three sites were connected: University of Delaware, [[Princeton University]], and Purdue University. By 1982, 24 sites were connected expanding to 84 sites by 1984, including one in Israel. Soon thereafter, connections were established to computer science departments in Australia, Canada, France, Germany, Korea, and Japan. CSNET eventually connected more than 180 institutions,<ref>[http://www.livinginternet.com/i/ii_csnet.htm CSNET History]</ref> 60% universities, although {{asof|1988|lc=y}} most members only had email connections. Sites exchanged data via either X.25 or Phonenet<!--Not Apple Phonenet-->, a [[store and forward]] protocol.{{r|fisher19880425}}

One of the earliest experiments in free software distribution on a network, [[netlib]], was available on CSNET.<ref>{{Cite journal |title=Distribution of mathematical software via electronic mail |author1=Jack J. Dongarra |author2=Eric Grosse |journal= Communications |date=May 1987 |volume=30 |pages=403–407 |number=5 |doi=10.1145/22899.22904 |citeseerx=10.1.1.73.7679|s2cid=52895937 |author1-link=Jack J. Dongarra }}</ref>

CSNET was a forerunner of the [[National Science Foundation Network]] (NSFNet) which eventually became a backbone of the Internet. CSNET operated autonomously until 1989, when it merged with [[BITNET]] to form the [[Corporation for Research and Educational Networking]] (CREN). By 1991, the success of the NSFNET and NSF-sponsored regional networks had rendered the CSNET services redundant, and the CSNET network was shut down in October 1991.<ref>[ftp://athos.rutgers.edu/resource-guide/chapter6/section6-6.txt CSNET-CIC Shutdown Notice]{{dead link|date=May 2025|bot=medic}}{{cbignore|bot=medic}}</ref>