Control, Change and the Internet

Patrick Maslen (patrick@quest-bird.com)

Chapter 4: "IP on Everything"

"Just as within the last thirty years the United States has become a 'national society', so in the next thirty years we will have an international society - not as a political order, but at least within the space-time framework of communications."

- Daniel Bell, The Social Framework of the Information Society (1980)

By the mid-1990s, the Protocol Wars were over and the Internet Protocol had become a global standard (taking the English language with it). The countries not connected to the Internet became a minority. There were several possible reasons for their abstention. Economics, national religion and hostility to the United States were factors, but most of all was fear of the subversive effect of the Internet's global society upon local populations. This global society, or 'trans-nation' had transcended its 1980s limitations and become an international player in its own right, sometimes even challenging the sovereignty of individual states.

As the Internet expanded beyond the United States' borders in the late 1980s, its English-language bias became a problem. The idea of an international language is as old as the Tower of Babel. So is international disagreement about what that language should be. Disputes of this kind led to the creation of artificial languages like Esperanto in the 1920s, which received little support. International argument about language was moot, however, since there were not enough people from different language groups engaged in regular discourse for a standard language to be really necessary. When the Internet made such discourse possible, the advantages of a standard language became obvious to the users of the net. The Internet's language was English. Just as TCP/IP became the world standard protocol by use rather than declaration, so did the English language, riding piggy-back on Internet message packets.

English, the 'language of democracy,'[1] was unavoidably part of the Internet Protocol, and it imparted its own cultural legacy to Internet society. English was the de facto language of trade and diplomacy. Even before the Internet it was a major language spoken by computer scientists. The American company IBM was one of the main worldwide suppliers of computer equipment. Knowledge of English was therefore a prerequisite for understanding many computer languages. English became the language of computing.

This 'common language' asserted itself across the networks of the Internet, in a way that the TCP/IP protocols did not. Netizens' use of English was enforced on both social and technical levels. The Internet protocols did not force anyone to speak English, but since it was the language of the American netizen majority, full participation in Internet discussions and newsgroups - the essence of Internet society - was difficult without it. The use of English on the Internet was a factor which favoured those who spoke it (like the Americans), and to some extent, alienated those who did not.

Another reason for the proliferation of English across the Internet was technical. All American computer equipment used a standard code called ASCII (American Standard Code for Information Interchange) to represent letters, number symbols and other characters. The trouble with this system, which only became apparent when dealing with languages other than English, was that it could not cope well with non-Roman characters. The ASCII character set only allowed an absolute maximum of 256 different characters (it normally used half that). Anglocentric ASCII did not account for accented Roman characters as used in western European languages like French, let alone other writing systems.

Two hundred and fifty-six different characters were nowhere near enough to cope with Asian languages. There are up to 7,500 Japanese kanji ideographs alone, let alone the thousands more characters used in China, Taiwan and Korea. In the case of JUNET (the Japanese Unix Network) which connected to the Internet in 1984, ASCII limitations and lack of local standards for kanji character conversion forced Japanese researchers to use English, the 'lowest common denominator,'[2] in their dealings with the global network community.

ASCII was a myopic standard which absolutely failed to address the 'peculiarities' of non-English languages.[3] Unlike TCP/IP, ASCII was an American standard which did not travel the globe gracefully. It hard-wired language intolerance into the system. Within the confines of the United States it was useful, but as soon as TCP/IP foisted it on the rest of the world, ASCII became a political manifesto. Still, whether it enforced or encouraged the use of English, ASCII did provide the global Internet community with a common language. It is unlikely, that the world would have adopted one otherwise.

French computer scientists started the protocol wars partly because they had developed the Internet's rival system themselves, but also to contain the 'contagion' of the English language spreading further out into the international computing community. Even though the peer-to-peer nature of the Internet protocol made global centralised control of computers impossible, it increased the homogenisation of English as an international language. This process accelerated as Europeans, many of whom spoke English as a second language anyway, adopted the Internet protocol over OSI. The Internet accelerated and developed but did not of itself cause the internationalisation of English[4]; that was a process already well advanced. In spite of ASCII's language bias, the Internet continued to expand internationally.

Any culture which expands very rapidly and contacts people from different cultures is likely to change. This is especially true of fast-forward Internet culture, whose population increased faster than any other society. In spite of its large Western population and majority culture, the Internet's peer-to-peer, two-way communication structure and its vociferous yet open USENET culture accepted and was transformed by newcomers from all over the world. The Internet had always been international, but its expansion in the 1990s allowed it to become truly global.

Worldwide Network Growth
No. of Countries connected to:
Date Internet BITNET UUCP FidoNet OSI
Sep-91 31 47 79 49
Dec-91 33 46 78 53
Feb-92 38 46 92 63
Apr-92 40 47 90 66 25
Aug-92 49 46 89 67 26
Jan-93 50 50 101 72 31
Apr-93 56 51 107 79 31
Aug-93 59 51 117 84 31
Feb-94 62 51 125 88 31
Jul-94 75 52 129 89 31
Nov-94 81 51 133 95 - -
Feb-95 86 48 141 98 - -
Jun-95 96 47 144 99 - -
(source: Zakon, Hobbes' Internet Timeline v.2.4a, 22 Feb 1996.)

By 1989, the Internet connected 100,000 computers. ARPANET ceased to exist, swallowed by the Internet, a 'happy victim of its own overwhelming success.'[5] By mid-1992 there were a million hosts; the following year, two million. Meanwhile, the protocol wars were dwindling in their intensity and relevance to the ever-growing Internet population. Between 1992 and 1994, thirty-five countries connected to the Internet, twenty-five of them in 1993. During this period, the number of OSI- connected countries increased by just six, and then froze. By the end of 1994, the protocol wars were over and the Internet had doubled the number of countries it connected. Three years was all it took thanks to the Internet's enormous rate of change.

In mid-1994, the Internet was growing faster outside the U.S. than inside it, but the United States still possessed the majority of connected computers and population in the burgeoning worldwide community. 27% of the three million Internet connected computers were operated by U.S. higher educational institutions, 24% by U.S. commercial sites. Other United States operators including government and military nodes accounted for a further 12% of the total. Western European nations operated another 23% of the network, with 10% shared by Australia, Canada and Japan. The remaining four percent of all connected nodes, some 128,000 computers, were shared between the remaining 62 connected countries.[6]

map of Internet connectivity in June 1996

Most countries in the world were connected to the Internet by mid-1996. Many countries which had no Internet access could still connect to the global e-mail network via cheaper technologies such as UUCP and FidoNet. The countries which were not connected to the global net at all were a small minority. A glance at the above map reveals two immediate facts about worldwide Internet distribution.[7] Countries with no Internet connection are in the minority and Africa stands out as the worst-connected continent, with significant areas of poor or non-existent Internet connectivity (only five African countries were connected to the Internet before 1994.)[8] What the map does not show is the degree of connection, the number of sites in each connected country, which in some cases is very small or restricted.

In June 1996, just twenty countries remained with absolutely no e-mail or Internet connection: Afghanistan, Bhutan, Burma, Burundi, Congo, Gabon, Guinea Bissau, Iraq, Liberia, Libya, Mauritania, North Korea, Oman, Rio Muni , Rwanda, Somalia, Syria, Western Sahara, Yemen, and Zaire. A few years before there had been many countries like them. By 1996, their absence from Internet was notable; their people were outside the global network community. Of these countries, more than half are in Africa, in most cases only recently (fifty years or less) free from colonial rule.[9] Many of these countries are not wealthy, but there are surprising exceptions. Libya, Western Sahara, Oman and Gabon are relatively affluent, so their abstention from the Internet cannot be explained in purely economic terms.

Politics, economics and international diplomacy have always been involved in the spread of the Internet. Sixteen of these twenty 'unwired' countries are Islamic, or have large Islamic populations. Saudi Arabia, another Islamic nation, has Internet access, but only just: a few nodes restricted to hospitals and universities. Algeria and Iran are similar. Egypt, Indonesia and the Philippines (also with significant Islamic populations) have fewer Internet connections than average, relative to their GNP.[10] Governments of Muslim countries seem shy of the Internet, distrusting its American influences.

At least six Internet-free countries have been involved in destructive internal or external conflicts in the last ten years, often involving the United Nations. Iraq, Libya, North Korea and Somalia (a fifth of the no-Internet countries) have all been on poor (or hostile) diplomatic terms with the United States since the early 1990s. Hostility to the United States by these countries, or vice versa, could help to explain their reluctance to connect to the Internet.

A 1995 report by the United States' RAND corporation showed a definite (two-way) correlation between countries' levels of democracy[11] and Internet connectivity. Sixteen of the twenty unconnected nations possessed very low levels of democracy. The Internet Society (ISOC), a non-profit organisation dedicated to supporting the Internet, suggests that Internet diffusion is aided by local availability of reasonably priced computer equipment, telecommunications lines, and skilled individuals and institutions. ISOC mentions, but does not dwell upon regulatory constraints placed upon computer networks.[12]

These regulatory constraints are usually political in origin, from governments which fear the subversive effects of Internet connection upon their populations. The government of Burma provides an extreme example of this sort of legislation. In 1996 it imposed a 15-year prison sentence on anyone who owned, imported, borrowed or used a modem, or set up a computer network without the permission of the Ministry of Communications, Posts and Telegraphs.[13] The Burmese government, one of the countries which scored the lowest "democracy rating" in the RAND report, has reason to fear the Internet.

The great number of people talking to each other through computer networks such as the Internet led to the emergence of what Howard Frederick, former director of PeaceNet, terms 'the global civil society'[14] If nations are 'communities that define themselves or are defined through cultural features,'[15] then it is possible to view Internet culture as a nation, or a 'trans-nation'. It has its own history, a common language. USENET has its own cultural features, such as netiquette and flaming, and its exclusively written medium 'keeps much of the three-dimensional external world out'[16] Henry Hardy claims that the Net is 'a nation of first allegiance for many of its members.'[17] By the 1990s, the anarchic 'net-nation' Internet did not concern itself with crotchety system administrators restricting news feeds. It had grown large enough to interfere with the power of states to control their populations.

In August 1991, information about the attempted KGB coup d'état in Russia leaked out of Moscow via UUCP to the Internet, much to the chagrin of the KGB. Anatoly Voranov was a system administrator of GLASNET in Moscow at the time:

I remember a posting from a Chinese student in America, a participant in the Tiananmen Square events in Beijing, offering to share his personal experiences of how to beat tanks in the heart of the city. People wondered why the KGB didn't cut our connection. I wonder too. I think they simply didn't know that we existed. And we had a trick: the UUCP connection originated in San Francisco, because at that time a non-authorised person or organisation could not call abroad from Moscow. And it was impossible even for the KGB to cut the phone link of the whole of Moscow.[18]
GLASNET was not the only network which got news of the coup out of the country. Although KGB coup-supporters neglected computer networks in their plans, it is unlikely that they would have been able to completely stifle them; as Voranov says, even the KGB could not silence Russia's entire telecommunications network. It was not simply the result of an oversight, or a fluke that information escaped Russia electronically. The loyalty of network users lay with the network, with the world community, and not with the State.

Groups of people with similar interests, previously isolated from each other, connected to the Internet. These groups ranged from soap-opera lovers to peace activists to criminals. Some of these people used the Internet to form organisations which owed allegiance to no particular government or corporation, such as PeaceNet, a global human-rights movement. PeaceNet was part of a worldwide non- governmental organisation of inexpensive, local networks (GLASNET was one) called the Association for Progressive Communications (APC). The APC is dedicated to freedom of information; one of its boasts is that it 'provides the first free flow of information between the United States and Cuba in thirty years.'[19]

Non-governmental organisations (NGOs) on the computer nets were new players in international relations, joining grudging governments and transnational corporations. NGOs included not only computer networks like the Association for Progressive Communication, but also more 'traditional' extra-governmental organisations such as Amnesty International, Greenpeace and the United Nations, who used the Internet to forward their own agendas. The effects of the Internet (and vice versa, for the Internet works both ways) on some of the world's oldest non- governmental organisations - organised religions (the Vatican has been on-line since 1995) - remain to be seen.[20]

Some non-governmental organisations used the Internet to further goals which directly opposed those of national governments. The Zapatista rebels of Mexico used the Internet to attract worldwide and local support for their struggle against the Mexican government. The Internet allowed the Zapatistas to spread information about their situation to the world - and to their own country - in spite of a Mexican military cordon, and the government's control of traditional media in the country.[21] In response the Mexican government began to use its own Internet resources to attempt to counter the Zapatistas' "propaganda". The Mexican 'cyberwar' continues. The Internet enabled new forms of social organisation, but also new forms of conflict.[22]

J. C. R. Licklider, one of the conceptual founders of computer networking wrote in 1980: 'If networks are going to be to the future what the high seas were to the past, then their control is going to be the focus of international competition.'[23] But the Internet's experience of the 1980s demonstrated its resistance to control of various kinds. Its wild exponential growth and international spread made it even less controllable. The Internet and its related networks are the new high seas. Controlled by no-one, they are the domain of international players. These players include the global network society, transnational corporations, governments and even organised religion, woven together in a complex interdependence based on the Internet Protocol.


Footnotes

  1. Donald Wilhelm, Global Communications and Political Power, 1990, p.133.
  2. Jeffrey Shapard, 'Islands in the (Data)Stream: Language, Character Codes and Electronic Isolation in Japan', in Linda M. Harasim (ed.), Global Networks, 1993, p. 265-8.
  3. Since 1993, some improvements have been made in this area. See K. Moore, 'MIME (Multipurpose Internet Mail Extensions) Part Two: Message Header Extensions for Non-ASCII Text', RFC 1522, [NIS.NSF.NET/internet/documents/rfc1522.txt], 23 September 1993.
  4. F. J. M. Laver, 'Computers + Communication + People = ?', in Williams (ed.), Pathways to the Information Society, 1982, p. 4.
  5. Sterling, 'Short History of the Internet', [gopher://gopher.isoc.org:70/00/internet/history/short.history.of.internet], February 1993.
  6. Internet Society Press Release, 'Latest Internet measurements reveal dramatic growth in 1994', [gopher://info.isoc.org/00/isoc/pub/press-releases/host-count.txt], 4 August 1994.
  7. There is no "BITNET but not Internet" on this map; BITNET has been connected to the Internet since 1990.
  8. South Africa, Tunisia (1991), Egypt, Ghana, Kenya (1993), from Zakon, Hobbes' Internet Timeline v.2.4a, [http://info.isoc.org/guest/zakon/Internet/History/HIT.html], February 22 1996.
  9. Rex Beddis, A Sense of Place, 1982, pp. 10, 84.
  10. 'Internet Global Infrastructure Diffusion' graphic [globdif1.gif] at [http://info.isoc.org], 1994.
  11. Christopher Kedzie, 'International Implications for Global Democratization', ch. 6 of Robert H Anderson, Tora K. Bikson, Sally Ann Law, Bridger M. Mitchell, 'Universal Access to Email: Feasability and Societal Implications', RAND report, [http://www.rand.org/publications/MR/MR650/index.html], 1995.
  12. 'Factors Promoting Internet Diffusion', graphic [globdif3.gif] at [http://info.isoc.org], 1994.
  13. 'Burma Ban on Modems', Age, 1 October, 1996, p. D2.
  14. Frederick, 'Networks and the Emergence of the Global Civil Society', in Harasim, Global Networks, 1993, pp. 284-5.
  15. Braman, 'Horizons of the State', in Journal of Communication, vol. 45, no. 4 (Autumn 1995), p. 14.
  16. MacKinnon, Richard C., 'Searching for the Leviathan in USENET', in Jones (ed.), CyberSociety, 1995, p. 134.
  17. Hardy, 'The History of the Net', [http://www.ocean.ic.net/ftp/doc/nethist.html], 1993.
  18. Anatoly Voronov, GLASNET operator during 1991 Russian coup attempt, quoted in LaQuey, The Internet Companion, 1994, pp. 48-9.
  19. Howard Frederick, 'Networks and the Emergence of the Global Civil Society', in Harasim, Global Networks, 1993, p. 291
  20. at [http://www.vatican.va/], from Zakon, Hobbes' Internet Timeline v.2.4a, [http://info.isoc.org/guest/zakon/Internet/History/HIT.html], 22 February 1996.
  21. Harry Cleaver [hmcleave@eco.utexas.edu], 'Working The Net for the Zapatistas', in Chiapas95 [chiapas-l@profmexis.dgsca], 7 February 1996.
  22. The U.S. is taking cyberwar seriously. See John Arquilla and David Ronfeldt, 'Cyberwar is Coming!', [http://gopher.well.sf.ca.us/70/0/Military/cyberwar], 1993.
  23. J. C. R. Licklider, 'Computers and Government', in Dertouzos and Moses (eds.), The Computer Age, 1980, p. 118.

Where to next?