the links effect

Frames p68, Understanding Comics by Scott McCloud

The link dissolves physical geography and institutes a new order, an order where proximity is defined by interest. An order in which everything is as close as everything else so the only things you keep near are those that you need or like. A site's centrality to its neighborhood, its standing as a 'net citizen', can be expressed by the number of other sites that connect to it. Some search engines make this explicit; Google, for example, ranks search returns, in part, by how many other sites link to it.

The link is what puts the 'hyper' into hypertext -- the practice that is at the heart of the Internet and which constitutes computer-mediated communications as a truly new medium. It is Vannever Bush who is credited with first speculating about hypertext in an article called "As We May Think" (July, 1945, The Atlantic Monthly), although he did not use the term and it would be two decades before anyone else did. Bush had been troubled by the discrepancy between humanity's mushrooming storehouse of knowledge and the inadequate tools for accessing that knowledge. Bush wrote:

The means we use for threading through the consequent maze to the momentarily important item is the same as it was used in the days of square-rigged ships... Our ineptitude in getting at the record is largely caused by the artificiality of systems of indexing. When data of any sort are placed in storage, they are filed alphabetically or numerically, and information is found (when it is) by tracing it down from subclass to subclass...The human mind does not work that way. It operates by association. With one item in its grasp, it snaps instantly to the next that is suggested by the association of thoughts, in accordance with some intricate web of trails carried by the cells of the brain.

An information retrieval system, Bush thought, should display a similar facility, being able to freely link associated subjects. His solution was an analogue storage device for multiple-media, the Memex. The Memex would allow users to construct trails of associations, based on their own cognition processes, concatenating data into personally meaningful sequences.

Hypertext links create associations between between data chunks precisely as Bush envisaged. What he hadn't envisaged, however, was that each person's Memex would be connected to and accessible to each others. The link makes any web session potentially a foray into 'hyper-montage'.

Montage was first theorised in relation to film. In a famous experiment in 1919, film director Lev Kuleshov edited footage of an actor's face before three other images; a bowl of hot soup, a dead woman in a coffin, and a little girl playing with a toy. Each sequence was shown to a separate audience, and each audience saw three very different emotions -- hunger, horror, parental love -- expressed by identical images of the actor. Kuleshov had demonstrated that editing, the concatenation of disparate shots, was more powerful than the content. Eisenstein went on to further explicate that narrative is created not by the content of the individual shots in a film but by their interaction with each other -- their sequence, contiguity and relationships.

This has important repercussion for the web author. In cinema the director controls the montage and thus the production of meaning: it's HER story she wants to tell.* The Internet hypes montage into a higher gear in which there is no way of really controlling what the user experiences or even in what order. Traditional modes of conveying meaning may be characterise as disciplining the audience, allowing some readings and not others, encouraging or disallowing behaviors. The author maintains a direct, if tenuous, control.

The Internet is free of many of the behavioral and, as we've seen, interpretive restrictions of older media. Navigation bars, site maps and annotated links only gesture towards controlling the users experience and, thereby, interpretation. The user can click out at any time, can choose to access parts of a site at their own whim or be skimming several sites simultaneously. This reality requires new ways to engage the users' attention and keep it; a new rhetoric needs to be developed appropriate to the medium.*

It is the fact of links that makes users exhausted, frustrated and 'lost' online. Links are seldom differentiated and a click may lead a user to another part of the site, to some related data or image, or to another site entirely. The primitive conventions that exist are honoured more in the breach than in practice. Few web designers stick with the default link settings (underlined blue for unvisited, red for active, purple for visited links) -- and some don't indicate links at all. Whilst some sites adhere rigerously to an internal schema (annotating links that lead to other sections of the site and to external data, using stylesheets to visually differentiate between link types, etc), such formula are not exportable to the rest of the Web. Part of the problem is that if it can be clicked, it can be a link.

In the early years of the Web the hypertext community, in particular, made many suggestions for making links more informative. These included instituting simple universal visual signifiers to differentiate between link types -- for example, one colour for links to footnotes and glossaries, another colour for sequential material, another for related material and another for link to offsite and so on. Another suggestion was to display the age and usage of links graphically; for example, by fading the colour over time (like autumn leaves), or in response to low usage (using the metaphor of foot-trails gradually disappearing through lack of use). Currently, many Javascripts and applets exist which will allow site authors to add metainformation about link destinations to appear on mouse-over -- these are seldom used.

Most web designers and theorists have given up trying to combat the problem of uninformative links at an Internet-wide level and have concentrated on what can be done to improve user experience within sites. Usability and design experts rant against 'mystery meat' interfaces. They suggest standardised schemas for use within sites -- consistant navigation bars and common metaphors, limiting the number of link options available at any one time, not linking to outside sites, etc etc. Usability guru Jakob Neilsen has been amongst the most vocal and has been instrumental in shifting the focus from controlling users to responding to their needs. It is unfortunate, however, that the end result of much of this advice is radically conservative and boils down to 'stick with what users already know and feel comfortable with'.

Editorialized lists of links, from the huge Yahoo-style sites to more humble 'favorite links' pages, have always been one of the most utilised aids for finding stuff online -- but at the cost of leaving the user at the mercy of someone else's taste and judgment. In addition, link indexes are tedious and time-consuming to maintain, requiring frequent checking to ensure that they remain topical as new sites go up and linked-to sites unexpectedly disappear. Most don't bother -- hence the prevalence of error 404s.

One of the most effective solutions has turned out to be one of the simplest. WebRings extended the concept of site link pages, which tend to promote the web author's friends' or related sites. Whilst the basic WebRing concept is very simple -- related topic sites join together in a ring, joining one to the next through a conventional hyperlink -- the ambition is much greater.

WebRings offer a method of following through on discussion particularly developed in relation to hypertext, but with important implications for web organisation generally. Hypertext theorist, Jay David Bolter refers to electronic writing as "topographic," as "both a verbal and a visual description." By this he means "not the writing of a place, but rather a writing with places, spatially-realised topics."*

Hypertext is topographic in the sense that it both describes a place and constitutes a space. Howard Rheingold noted, reflecting on his experience in and with the WELL, that users navigate the Net in ways that give a sense of 'neighborhood', particularly as the route becomes routine:... Because topic is the basis for choosing with whom to communicate, topic is of the utmost importance in structuring and navigating the vast electronic net.... Topic control is of such burning interest precisely because topic is place -- from topos, the Greek word for place. ...topic control might be seen as cartographic."*

What WebRings do is keep you in the neighborhood of the topic of interest -- no matter how far apart the actual nodes that store the material are.

Operated by Starseed, Inc. since August, 1997, WebRing was first created in June, 1995, by Sage Weil and Troy Griffith, launched in March, 1996. It drew its initial inspiration from a similar structure called EUROPa (Expanding Unidirectional Ring of Pages), a system servicing three primary World Wide Web groups: visitors, member sites and advertiser-merchants,

WebRings are open and free of charge to both visitors and members. The WebRing system can support a nearly unlimited number of separate and distinct rings across the Internet. This unique structure allows the creation and evolution of tens of thousands of different web communities. Each ring was started and is maintained by an individual website owner. Through navigation links found most often at the bottom of member pages, visitors can travel all or any of the sites in a ring. It's a simple idea but obviously very effective -- in little more than two years of operation (March 1996- April 1998), over 40,000 separate rings were instigated and connected to WebRing.

Next
Visualizing the Datasphere