Special issue of HT'19 selected papers

Introduction

Many hypertext publications mention Vannevar Bush's Memex (Bush, 1945) as one of the original ideas of hypertext. Memex is an acronym for Memory Extender. One of its core features is to store “trails” of thoughts persistently over documents such that a user may follow them at a later point of time. Bush only described Memex, but never built it physically. This was the time before the rise of digital computers, and Bush's machine was a mechanical device built around documents stored on microfiche.

Bush's ideas were been taken up again in the 1960s by hypertext pioneers such as Douglas Engelbart, Theodor Nelson, or Andries van Dam. Computers, although expensive, were already available at that time, making hypertext as software systems possible. This was a necessary prerequisite for further developments in the field. In fact, Nelson, who coined the term hypertext realised their necessity: “Let me introduce the word ‘hypertext’ to mean a body of written or pictorial material interconnected in such a complex way that it could not conveniently be presented or represented on paper.” (Nelson, 1965)

At that point of time, the focus in the field was primarily on nodes interconnected by links. Discussions took place mainly among academics—the industry was not yet broadly interested.

The situation changed with the rise of personal computers, which were affordable by ordinary people and organisations, and by the 1980s, several hypertext applications have been developed by academics and software companies. It was a time with many competing hypertext approaches. For example, Eastgate Systems released Storyspace (Bernstein, 2002; Joyce, 1991), a hypertext system that offers a 2D space for writing hypertext fiction; Brown University developed Intermedia with the promise to provide link creation mechanisms that would be as easy as copy & paste (Meyrowitz, 1986, 1989); and the hypertext system Guide (Brown, 1987) was one of the first cross-platform hypertext applications that ran on Macintosh and Windows PCs. There were many other examples: KMS (Akscyn et al., 1988), Hyperties (Shneiderman, 1987), NoteCards (Halasz et al., 1987) and HyperCard (Smith & Bernhardt, 1988), all of which explored different models and functionalities.

The Dexter Hypertext Reference Model (Grønbæk & Trigg, 1992; Halasz & Schwartz, 19942) aimed at providing a common scheme and terminology for all of these new systems. It was an effort to unify a community that was so innovative that it had a tendency to drift apart. In the 1990s, new types of hypertext (other than node–link structures) began to appear, including spatial hypertext (Marshall et al., 1991), taxonomic hypertext (Parunak, 1991) or argumentation supporting structures (Conklin & Begeman, 1987). The perception of hypertext evolved beyond nodes and links. From then on, hypertext researchers also called spatially represented nodes or taxonomies “hypertext”, for example. The community benefited from diversity, but at the cost of a loss of focus, and a shrinking common ground.

The need to support arbitrary structure types as well as multiple foundation services, such as concurrency control, structure storage, or versioning control, inspired system developers to come up with Open Hypermedia Systems (OHS), Component-based OHS (CB-OHS) and the idea of structural computing. It was another “synchronization point” (Wiil, 2005) that has drawn a boundary around the field of hypertext.

The situation changed radically with the arrival of the World Wide Web in the 1990. The huge success and wide adoption of this system wiped out many hypertext systems or reduced them to academic playgrounds. Even though academics worked on fusing OHS and WWW (e.g. Bouvin, 2000), from a hypertext perspective, it created a monoculture and many hypertext ideas were lost. The Web community went along a more data-centric path and many hypertext researchers followed.

In the following years, the hypertext community lost members to other surrounding communities too, such as Adaptive Hypermedia and User Modelling, the Semantic Web, and Web Science. All became conferences in their own right, while the artistic and literary parts of the community found new homes at places like the Electronic Literature Organisation. In 2012, the ACM Conference on Hypertext and Hypermedia was firstly called the ACM Conference on Hypertext and Social Media due to an increasing number of contributions from that field. However, in total, over the years, a decreasing number of attendees participated at the ACM Hypertext Conference itself (ACM's flagship conference on this topic), and an increasing sense of separation between the various parts of the Hypertext Diaspora developed.

With the motto “Hypertext—tear down the wall” the 30th anniversary ACM Conference on Hypertext and Social Media¹, which took place at Hof University, Germany in 2019, aimed at bringing these communities back together.

This special issue of the New Review of Hypertext and Multimedia journal includes extended versions of selected ACM Hypertext 2019 publications that were highly rated by reviewers. They were presented in different tracks of the conference, but all represent the interdisciplinary nature of modern hypertext research that makes it so valuable for both industry and academia.

Two publications look at hypertext as literature and explore how it extends and challenges our notions of writing.

Sam Brooker (Richmond, The American International University in London) explores the complex relationships between author and reader in interactive texts, arguing that the author's intentions should have more of a place in our hypertext critical theory.

Stacy Mason (University of California Santa Cruz) and Mark Bernstein (Eastgate Systems) investigate the poetics of contemporary link usage, describing the link as “the most important new punctuation mark since the invention of the comma”, with the aim of informing and inspiring new writers and creating a better vocabulary of linking.

Two more are concerned with hypertext in-the-wild, and in particularly the phenomenon of fake news.

Jacob Wobbrock and his collaborators (University of Washington) examine the credibility perceptions of news articles based on their presentation features independently of content. Their work shows how susceptible we are to manipulation, and how important it is to be aware of when our heuristics work against us.

Jakub Simko and his collaborators (Kempelen Institute of Intelligent Technologies and the Slovak University of Technology) present work looking at how experts judge the veracity of online news content, in order to identify new heuristics and to help create automated systems that make that judgement easier.

Our final two papers seek hypertext in other systems and media.

Isaac Alpizar-Chacon and Sergey Sosnovsky (Ultrecht University), are doing this explicitly by extracting the knowledge structures embedded in traditional textbooks, and using this information to enrich them with links and annotations. Their work shows how we might utilise the hypertext features embedded in non-hypertext content to synthesise knowledge and automatically create more interactive experiences.

Claus Atzenbeck and his collaborators (Hof University and KUNGFU.AI) go even further and challenge the view of hypertext as a type of system, and instead present it as a way of viewing systems. On this basis, they strive for a synthesis of augmentation and automation, which becomes visible in the intersection of human users and intelligent machines.

The work collected here, and presented at the 30th ACM Hypertext conference, demonstrates not only that hypertext work is broad and interdisciplinary, but that it has become an essential tool in our lives, and a critical technology and approach for building and understanding our digital world. By tearing down the walls between our communities we can benefit from each other's perspectives and views of hypertext, and look forward together to the next decade of this critical and important technology.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 https://ht.acm.org/ht2019/