https://d2ng68cf36kxsk.cloudfront.net/store/public/post/1195/post_images/Slide375.JPG">

Huge investments in Soviet cybernetics followed. However, although a large number of computer centers were created around the Soviet Union, they were never connected. Benjamin Peters tells a painful story of military opposition (they had their own secret network which they wouldn’t share), bureaucratic territorialism (enterprises whose autonomy and capacity for number-rigging would be reduced), and a layer of the informal economy (a large unsanctioned market economy and widespread corruption, vested interests that such a system would expose and possibly eliminate).

In the United States, comprehensively networked computing did not come until the Advanced Research Projects Agency Network (ARPANET). This began its life as a military project in 1969. Academic access was allowed in 1981, and the internet protocol TCP/IP^§MS1.1.1f established in 1982. Peters calls this the “American military-industrial-academic complex,” where he says the actors behaved like socialists. Meanwhile, “the Soviet networked command economy fell apart not because it resisted the superior practices of competitive free markets but because it was consumed by unregulated conflicts among institutional and individual self-interests.”⁴⁶⁷ The Soviets, ironically, were behaving like capitalists. Or, we would say by way of homage to Deng Xiaoping, this part of the Cold War was won by a socialism that had American characteristics.

As for the distributed logic of the network, system architect and inventor of “packet switching,” Polish-born Jewish refugee Paul Baran, had nuclear war in mind. A distributed network was more likely to survive nuclear devastation than a centralized one. He published widely hoping the Soviets would take note, because the best deterrent for nuclear war, it seemed to him at the time, would be the indestructability of the other side.⁴⁶⁸

Then, “hypertext,” to get all the meanings connected that were referenced in computers. Ted Nelson is credited with coining the word. “Ordinary text,” he said in 1967, “called by McLuhan and others ‘linear,’ is a continuing sequence in a fixed order … Hypertext diverges from ordinary text in that the readers’ possible sequences diverge from plain sequence.” He predicted “new forms of writing, appearing on computer screens, that will branch to perform at the reader’s command. A hypertext is a non-sequential piece of writing; only the computer display makes it practical. Somewhere between a book, a TV show and a penny arcade, the hypertext can be a vast tapestry of information.”⁴⁶⁹

In this definition, the origins of hypertext were in the sixteenth century, with the invention of page numbers, indexes, tables of contents, headings, figures cross-referenced in text, and beyond the covers of the book, library catalogues, shelving systems, bibliographies, and citation protocols.^§AS1.4.7c All of these were designed for non-linear readings, accommodating the interested and interpreting reader.^§AS1.2.3a They also encapsulated the foundational epistemologies of modern science and scholarly knowledge, a social intertextuality acknowledging the sources of quotations, the provenance of ideas.

The integrated network was to depend on precise text-mediated, text-to-text connections in the form of Uniform Record Locators (URLs),^§MS1.1.1f names for places where copyable sources were to be located, in highly regulated conditions where no two names could ever be the same. When hypertext became a reality in HTML, it was with the <link> tag, though this had not been in Berners-Lee’s 1991 version and was only to appear from HTML 2 in 1995.^§AS1.4.7d Nelson had defined this in 1973: “link – connection between two points … which may serve as a jump from one to the other.”⁴⁷⁰ The innovation here was not the textuality of interconnection. It was the now universal computer protocols of its activation and the speed with which it could operationalize these.

Next came the web browser, by means of which files could be located and viewed. The first widely used web browser, Mosaic, was developed in the National Center for Supercomputing Applications at the University of Illinois and released in 1993.

Then, in 1998, Stanford University students Larry Page and Sergei Brin presented a paper about their doctoral research project at the Seventh International World-Wide Web Conference in Brisbane, Australia, “The Anatomy of a Large-Scale Hypertextual Web Search Engine.” In their paper, they described the “full text and hyperlink database of at least 24 million pages” that they had developed, and the “PageRank” algorithm which used “the link structure of the Web to calculate a quality ranking for each web page.” Keyword search would bring up a huge number of pages in no priority order. But by counting the number of backward links – other pages that had already cited a particular page – they were able to rank pages.⁴⁷¹

A patent application for Page and Brin’s invention had been filed several months before, with ownership assigned to Stanford University according to the practice required of graduate students in the United States.⁴⁷² Their research had been supported by a grant from the National Science Foundation under the US Government’s Digital Library initiative. The University did a deal in which it took Google shares instead of royalties, then sold its shares for a tiny fraction of their subsequent value. This is how oligarchies are created in the twenty-first century.

Not that was there anything particularly original about the PageRank algorithm. Larry Page was son of a Michigan State University computer science professor, and professors are of course conscious of their citations.⁴⁷³ In 1955, Eugene Garfield began to develop a database that counted scientific article citations as a measure of their impact – the backward links from later work that acknowledge antecedent works.⁴⁷⁴ The Lithuanian-Italian-Jewish Garfield had studied in the same linguistics department at the University of Pennsylvania as Noam Chomsky. Then, in 1960, Garfield created a company to sell these data, the Institute for Scientific Information. Page and Brin did not cite Garfield in their Brisbane paper, ironically demonstrating that the source of ideas is not necessarily explicitly linked to their later impact. But they did cite Garfield in another, lesser known paper dated to that year,⁴⁷⁵ and in their patent application. This is how the idea of backward linking slipped into the hands of the company they founded, Google, as private intellectual property.

Actually, it was not the algorithm that proved significant, but the copy of the 24 million pages Page and Brin had copied and indexed to test the algorithm. Nor is the link count a measure of quality, as their Brisbane paper claimed. It’s a measure of a quantity: popularity counted by links. Accentuating this, a link from an already well-linked page is ranked above a link from a page less linked. And the more people searching something, the more the algorithm recommends the same link to others. This is not information quality, but populism on an industrial scale.

Page and Brin founded the web search company Google in 1998. Google has a copy of the whole web – there are several others, including those held by Microsoft and Yahoo, but these pale into insignificance by comparison. Clickstream data independently collected in 2018 showed that Google had 92.96% of market share. The second largest provider is Yahoo, at 1.57% of the market, followed by Microsoft Bing at 1.48%.⁴⁷⁶ Google was the first copy and remains the most used. These copies, incidentally, were made without ever asking permission of creators or publishers, so contravening the foundational principles of private ownership of intellectual property. Private property is supposedly inviolate, but some violators the system tolerates more than others. But then, Google, Microsoft, and Yahoo are today megacorporations that are too big to be criminal.

When you search at Google, you are not actually searching the web – that would be impossibly slow. You are searching Google’s indexed copy of the web. The logic of web search that Google and the other companies offer is principally a business of counting collocations of letters as proxies for the meanings of words, then using hyperlinks as a measure of popularity. This is what brings you to the first page of a search listing. Hence the junk populism of the internet. Then there is a cultural and epistemic version of corruption, where you can pay to be placed ahead of others, where money can trump popularity. This is Google’s business model, and for that matter today, the ailing business model of liberal democracy.

Several decades later, Google has made a copy of everything, not just the whole internet, but every book for Google Book Search, and every journal article for Google Scholar. Google is just as interested in copying your private ephemera – every web search you happen to make, every email you send by GMail or through its hidden proxies, and everything you write in Google Docs or save to Google Drive. Nothing is too trivial for Google to know, for its figuring exactly what each one of us is meaning, in our finely nuanced differences.^§AS2.2.2 It’s a well-targeted advertising business.

Google’s transpositions are fundamentally algorithmic. Its main technology is to count the graphemes where statistical similarities might add up to meanings that are never more than latent.⁴⁷⁷ Even though its main business is words, it has precious little reliance on linguistics.^§AS2.2.1a Instead, it relies on ontologies that make sense of the words by locating them in the universal encylopedia of human experience.^§MS3.4

As for the distributed network, the reality is the increasingly opposite. The cloud is a network of highly centralized server farms, not only at Google (for search), but Facebook (for social media), and that cloud of clouds, Amazon Web Services. Google has laid undersea cables around the world, creating its own private internet. The power of these companies comes from their ownership of industrial plant, to even greater monopolistic effect than the industrial robber barons of early modernity.

For all the talk of the free market, there are by now just a handful of digital behemoths,⁴⁷⁸ and as a consequence we live our lives in pseudo-public places of the internet, where we consent to certain kinds of control and an imbalance of benefits that guarantees corporate profits on a scale not seen even in the industrial era. Why would a megacorporation give you anything for free? Because the value of what they get back out of you is greater than the cost of their “giving.”

Today we have a massively interlinked but also highly centralized, universal infrastructure of meanings. This is held together by an oligarchy of centrally controlled entities who come together when needed to agree on protocols of standardization in this new, stateless regime for the governance of our meanings.

This was what Viktor Gluskov was thinking for his network design of 1963 – and not.

---

⁴⁶⁴ Anthony Grafton, 1997, The Footnote: A Curious History, London: Faber and Faber.

⁴⁶⁵ Roger Chartier, 1992 [1994], The Order of Books: Readers, Authors, and Libraries in Europe between the Fourteenth and Eighteenth Centuries, translated by L.G. Cochrane, Stanford, CA: Stanford University Press, pp. 77ff.

⁴⁶⁶ Benjamin Peters, 2016, How Not to Network a Nation: The Uneasy History of the Soviet Internet, Cambridge, MA: MIT Press, pp. 107–08.

⁴⁶⁷ Ibid., pp. 197, 187.

⁴⁶⁸ Ibid., pp. 94–97.

⁴⁶⁹ Theodor H. Nelson, 1977, Selected Papers, 1965–77, Swarthmore, PA: Swathmore College, pp. 15–16.

⁴⁷⁰ Ibid., p. 105.

⁴⁷¹ Sergy Brin and Larry Page, 1998, “The Anatomy of a Large-Scale Hypertextual Web Search Engine,” paper presented at the Seventh International World-Wide Web Conference, 14–18 April, Brisbane (http://ilpubs.stanford.edu:8090/361/).

⁴⁷² Lawrence Page, 2001, “Method for Node Ranking in a Linked Database,” edited by U.S. Patent Office. Stanford, CA: The Board of Trustees of the Leland Stanford Junior University.

⁴⁷³ David A. Vise, 2005, The Google Story, New York: Delacorte Press, p. 37.

⁴⁷⁴ Eugene Garfield, 1955 [2006], “Citation Indexes for Science: A New Dimension in Documentation through Association of Ideas,” International Journal of Epidemiology 35(5):1123–27, doi: 10.1093/ije/dyl189; Eugene Garfield, 2006, “Commentary: Fifty Years of Citation Indexing,” ibid.:1127–28; Bill Cope and Mary Kalantzis, 2014, “Changing Knowledge Ecologies and the Transformation of the Scholarly Journal,” pp. 9–84 in The Future of the Academic Journal (Edn 2), edited by B. Cope and A. Phillips, Oxford: Elsevier.

⁴⁷⁵ Lawrence Page, Sergey Brin, Rajeev Motwani, and Terry Winograd, 1998, “The Pagerank Citation Ranking: Bringing Order to the Web,” Stanford, CA: Stanford InfoLab.

⁴⁷⁶ https://sparktoro.com/blog/2018-search-market-share-myths-vs-realities-of-google-bing-amazon-facebook-duckduckgo-more/

⁴⁷⁷ Thomas K. Landauer, Danielle S. McNamara, Simon Dennis, and Walter Kintsch, eds. 2007, Handbook of Latent Semantic Analysis, New York, NY: Routledge.

⁴⁷⁸ Matteo Pasquinelli, 2009, “Google’s Pagerank Algorithm: A Diagram of Cognitive Capitalism and the Rentier of the Common Intellect,” pp. 152–62 in Deep Search, edited by K. Becker and F. Stalder, London: Transaction Publishers.

---

Kalantzis, Mary, and Bill Cope. 2020. Adding Sense: Context and Interest in a Grammar of Multimodal Meaning. Cambridge UK: Cambridge University Press. pp. 172–177. https://doi.org/10.1017/9781108862059

Larry Page and Sergey Brin’s “PageRank” Algorithm §AS1.4.7e