Noortje Marres, Richard Rogers 1 Jan 1999

To Trace or to Rub

Screening the Web Navigation Debate

Ever since 'space' has become the ruling metaphor for the electronic realm, geographers have taken it upon themselves to plot the web. They have traded in their conventional tools and have begun developing new techniques to facilitate navigation and orienteering.



1 See the Geographies of Cyperspace project, led by geographer Martin Dodge, University College London, at ^^

As geographical models are informing the web journey of the future, it is worthwhile to explore which other models are available to turn the conventional search 'n surf practice into a more rewarding activity. Here we wish to focus on the expedition and the archeological models, put forward by the 'tracers' and the 'rubbers', respectively. The tracers and the rubbers share the conviction that there are social affiliations out there on the web. New models of these social affiliations should inform web itineraries - the meaningful journeys of the future.

There are two schools of thought in the social navigation debate on the web, represented by the 'tracers' and the 'rubbers'. The tracers view the surfer as the best guide to the web. Much like nineteenth-century polar explorers desiring to establish whether and how their compatriots before them actually found some sense of 'the Pole', the tracers seek evidence of previous surfer presence and pathways to determine whether and how they found some sense of 'artificial intelligence', 'botany' or whatever other topic they were researching.22 F. Spufford I May Be Some Time London 1996
Tracers look for vestiges of how surfers have found what they were after. They then plot these pathways through the sources, and search for meaning. The meaning may be the best bi-ways to the information, which is useful for search engine logicians, or the meaning may be particular stories. Just as one may have asked whether a particular explorer 'authored' a (story of the discovery of the) North Pole, one now may ask whether a surfer authored a (storyline through and on the) web. What is the story of 'artificial intelligence' the surfer has just authored in his or her click-through? From surfers' 'footprints in the snow', tracers read surfer stories and pass ‘knowledge route diagrams' along to other surfers, who may read and follow them or leave them aside in lieu of their own path- and story-making. Agency lies first of all with the surfer.

Contrary to the information expedition model, the 'rubbers' take an archaeological tack. They see storylines within the existing webworks. Instead of looking for the ice pick, the piece of clothing or other traces left behind by previous expeditioners, they locate the graveyards and the gravestones, and 'rub' them for meaning. Web sites are the stones, and hyperlinks connect them in a yard. Taken together, the text and dates on the stones in the graveyard presumably would provide the raw material for a story, for example, of occurrence clusterings (all those fallen in a battle or stricken by the plague) or of occurrence distributions (all those gradually falling to natural causes). Rubbers look for meaning given not by the surfer routings, but by webmasters' markings, i.e., the text and hyperlinks on the sites they administer. To find the meaningful stories, the rubbers take out their white paper and black crayon, and 'rub' the markings on the stones, and in the yards. Storylines of, say, 'Global Positioning System' are found within a certain set of sites, demarcated by occurrences of links between the sites writing about gps.3

3 Of the population of sites mentioning the 'Global Positioning System', those receiving disproportionate numbers of ‘links in' are counted in the sample, thus demarcating the discourse.

The links and claims on the sites as narrational elements. Agency lies first of all with webmasters.

Tracers are currently the best known of the two schools, for their methods fit well into dominant narratives about the web. As the new realm is thought to challenge traditional notions of authorship and control, so surfers ‘write' the web and have volition over their own routes. The knowledgeable, free-spirit user determines his course, that is to say, the user has the privileged position and vantage point.

The method and technique thus fit well into the notion of the web as jungle. It takes the web as an urban-like environment offering a jumble of signs, doors and alleys, but no clues as to its dead ends and the whereabouts of its treasures.4

4 R. Rogers 'Playing with Search Engines and making lowly information into knowledge', Mediamatic, 9#2/3, Amsterdam 1998.

Here as well, it is the cunning and bold surfer making the journey. Against the backdrop of empowerment and chaotic systems, it makes sense to welcome the user as the locus of meaningful action.

Tracings are valuable resources, which could reveal individual patterns of behaviour and eventually ‘preferences'. Should surfers leave identity vestiges along with their pathways from jumpstation to jumpstation, tracers may begin to track surfer types, just as the Albert Heijn supermarket once wished to track grazer types with their preferred customer cards, before the age, gender and postal code fields were deleted from their database. While tracings might serve as customer profiles, e.g. informing garden centres about the virtual excursions of the amateur botanist, tracers have set their sights on profiling knowledge gatherers. Knowing the path taken by a computer scientist (or the aggregate computer scientist) through the artificial intelligence ‘literature' on the web would be useful to all future ai surfers, it is thought.

Rubbers are less interested in what surfers find on their own, or aggregate surfer preferences. They are intrigued by what webmasters like surfers to know, and by webmasters' ‘election' of relevant websites by means of making a hyperlink. Rubbing the linkage of websites treating a particular topic, they aim to reveal potential routings through a thematic network given by the web. The 'Global Positioning System' network, for example, discloses aspects that are currently defining the phenomenon (navigation, us military, consumer product, terrorist asset, environmental monitoring), as well as the relations between some of the more fascinating players.

Rubbers are interested in the distributed lives topics may lead on the web and in web actors' roles in staging them. Contrary to chaos and jungle theories of the web, they discern networks that are more or less articulated, depending on the degree to which websites share a discourse and are connected through links. To the knowledgeable, free-spirit user, they add the wandering spirits of topics that may be encountered on a diversity of web sites. Navigable, thematic networks emerge when sites are mapped.

Rubbings are valuable resources, which could reveal indications of relative organisational presence on the net. A specific set of organisations may appear as central in a particularly interesting storyline about 'climate change', say a intergovernmental consultative body, a scientific institute contracted by that body and an on-line newspaper archive. Other organisations, like a leading oil and gas lobby or an activist .org, turn out to take more marginal positions. Such recordings of players' positions could lead organisations to reconsider the wording and hyperlinking on their site. They may wish to appear closer to central players, to expand on their contribution to a storyline or to disappear from the map altogether.5

5 Pornography is a good example of a storyline in which any number of organisations would not wish to appear. In March of 1999, the internet intelligence organisation, announced that America's top brands * linked to porn. Of course, porn sites have linked to the top brands, not vice versa.

Where not organisations but surfers are concerned, the rubbings' active contextual framings may serve as point of reference for playful web research or for more obstinate route following. The latter could potentially result in storyline interference via a mail-the-master-button or such like. Rubbers have set their sights on adding value to web searches and on monitoring collectively staged thematic networks.

Tracers and rubbers share the conviction that a fruitful web search is informed by social affiliations. Where tracers find these affiliations in the shared interests of users, rubbers locate them in the common linkage of organisations, protagonistic and antagonistic alike. Tracers employ a community formula, according to which a durable base is to be set up for like-minded roamers. For tracers, the cultivated web of the future shows a parcelled landscape, where each parcel houses a community base where fellows request directions from fellows. Rubbers, conversely, advocate a formula that we wish to dub neo-pluralist. Rings of stakes may be set in the web, within which the unfolding of an event can be witnessed. Each of the players within provides its own version of the event. As the rubbings indicate where passageways between parcels have been laid out (or are absent),6

6 The neo-pluralist formula thus involves a typology of links: the allying link, the hostile link, the appropriative link, the supportive link and, ultimately, the absent link.

site visitors will be able to approach the heterogenous elements as a continuous configuration. According to rubbers, the web of the future may very well show a parcelled landscape, but at the intersections neo-pluralist sites will emerge. Here diverging positions may lock into a tight linkage, like when Goethe, Napoleon and, we may add, the unknown soldier, met at the battlefield of Jena.

On a more formal level, tracers can be said to build on the sociology of identity, while rubbers put forward a politics of association. Tracers assume the web to be peopled by individuals who adhere to one or more 'common interest' types. For these types to emerge, contact is to be established between the like-minded, as in 'Six Degrees'7

7 Idea developed by Guglielmo Marconi that every person is linked to the rest of humanity through six people or fewer. Popularised in John Guare's play Six Degrees of Separation (filmed in 1993 by Fred Schepisi and inspiration for the game Six Degrees of Kevin Bacon, playable on the Internet and elsewhere). On the Internet the concept achieved some notoriety through its application by the site (where by supplying the correct personal information, one could contact said relations). For a critical note on the site in question see note.

As a form of contact and bonding, tracers have singled out the exchange of preferences, a voluntary process free of the obtrusiveness implicit in the group-norm. In a way similar to's * who bought this book also bought..., it is through the appropriation of personal preferences that identity becomes articulated: there follows the prospect of focused self-assertion.

Rubbers depart from the assumption that with the variety of players now present on the web, joint platforms are emerging where diverse websites are being drawn together. In the way that has established links with thousands of formerly unrelated non-commercial websites,8

8 See D. Schiller Digital Capitalism, Networking the Global Market System Cambridge, Mass. 1999. The links from the grassroots organisation Friends of the Earth to Ford, Texaco and Shell are an other example; see

common grounds are being established between the maybe-not-so-like-minded. These platforms are brought into existence by hyperlinking and appropriating wordings; websites join the platforms on webmasters' own initiative (link out, quoting) or by 'force' (link in, quoted). Understanding both active and passive involvement in such platforms as defining a website's role and place on the web, rubbers single out a politics of association as the process by which web presence becomes articulated.

In sum, tracers and rubbers are the two goal posts between which the social navigation debate is played out. From the tracer's point of view, surfers hold the equivalent of an etch-a-sketch, which allows him or her to retrieve a portraiture of other surfers' lines of inquiry into information. From the rubber's point of view, thematic networks provide the dots to be connected (just as, elsewhere in this magazine, connecting the dots results in a Degas picture), allowing surfers to survey given lines of association. ‘Etch-a-sketching' or ‘Connecting the Dots' of web discourse already inform emerging search engine logics and net preferencing models, which are otherwise intentionally kept mysterious.