Unidade 1

a)
HypertextFirst definition (from T. Nelson in Literary Machines) :
Collection of written materials or images connected together in a so complex way which they could not be suitably presented or represented on paper.
Other definitions :
- database technology typically characterized by textual fragments of small size connected together by links managed by the computer.
- analogy with the semantic network, diagram of representation of knowledge employed in artificial intelligence, directed graph where concepts are “nodes” and their relations are “links”. The indexing is being done on the semantic contents of the concepts rather than on an arbitrary order (alphabetical for example).
Hypermedia :
Multimedia dimension of the hypertext = integration of still and animated images (videos, animations…), sounds and texts in hypertexts.
Typology of hyperlinks :
Several typologies exist. Among them :
- according to CONKLIN (Conklin, J., 1987, “Hypertext: An introduction and survey”. IEEE Computer, 20 (9)) : referential links (notes, quotations, annotations…= non-hierarchical links, connecting two parts of the hyperdocument) and organisational links (hierarchical links = presenting materials at various levels of detail, increasingly large, and connecting a link-“father” with his “sons”)
- according to VERCOUSTRE (Vercoustre, A.-M., INRIA, France) : explicit links (defined by the author in the hypertext), implicit links (dynamic ones : unexpectable), executables links (which are lauching scripts or programs)
- according to PARUNAK (Parunak, H.V.D., 1991, “Ordering the Information Graph”. in Hypertext/Hypermedia Handbook, Berk, E. & Devlin, J., eds., Intertext Publications/McGraw-Hill Publishing Co., Inc., New York) : associative links (from a word in the hypertext to the proposals, from proposal to other proposals, paraphrase, illustration), links of aggregation (connect a whole to its parts and vice-versa, like a metonymy), links of revision (connect a node to its previous or posterior versions).
Comments :
It’s important to understand that hypertexts are based on concepts of network (semantic network between nodes) and association (with links).
Complementary links to consult :
- "Definition of hypertext" (by J. Blustein, Faculty of Computer Science Dalhousie University, Halifax, Canada) : more definitions and points of view about it (click on the links in the text : Ted Nelson's definition...)
- "Fourth generation hypermedia: some missing links for the World Wide Web" (by M. Bieber, Hypermedia Information Systems Research Laboratory, New Jersey Institute of Technology, University Heights, U.S.A) : more explaination about typology of links in hypermedia systems.
Complementary texts to read :
- "As we may think", Vannevar Bush, Athletic Monthly, 1945
b)
Vannevar Bush faces the problem of the sum of the experiments and human knowledge which increase at extraordinary intervals, making their manual exploitation impossible. Technical solutions appear - photo cells, advanced photography, thermal and cathodic tubes - to allow a transformation of the scientific recordings. For instance :
- combination of optical projection and photographic reduction on microfilms (ratio from 1 to 20),
- possibilities of reducing this ratio by 1 to 10000 between the original and the recording on microfilm (= compression associated with a possibility of consultation),
- possibility of recording the document oneself directly, without having to use various intermediaries.

Vannevar Bush thus puts ahead the importance of the selection, currently difficult because of the mechanisms of alphabetical or alphanumeric indexing. The human spirit works by association of ideas, with an intricate collection of trails carried by the cells of brain, the transitory statute of memory, the fast action, the details of the mental images. It is impossible to translate accurately but we can be inspired some :
- selection by association first (rather than by indexing)
- selection to be mechanized (without reaching the flexibility and speed of human spirit).

Vannevar Bush’ proposal of the “Memex” is an individual bookstore allowing to store personal files, supplement of memory of the user with office, screens, keyboard and levers to question, record and consult its documents. It offers the possibilities of :
- visualizing various positions of projection on documents
- adding notes and comments on documents
- linking two items together, thus building a track, naming it and indexing it
- marking the items by a code (a pointer will draw attention to them).
- going directly from one item to another and coming backward

He proposed a new profession of “trail b lazers” = to propose useful ways in broad common recordings of knowledge.
Comments:
The text of Vannevar Bush is considered as the first conceptualization of hypertext, without naming it. Bush insists on the importance of selection, annotation and combination of documents in a semantic network. His proposition of a new job called “trail blazers” can be linked to actual roles of search engines and portals on the web.
Complementary links to consult :
- Vannevar Bush's biography
- presentation of theoretical hypertext computer system Memex (Wikipédia)
c)
Source : “Hypertext and hypermedia”, book, Jacob Nielsen, 1989, Academic Press
History :
Here are important dates to retain in the history of the hypertexts:
- 1945 : “Memex” (Memory Extender) project of Vannevar Bush = knowledge management tool which prefigures the concept of hypertext
- 1965 : Ted Nelson invents and publishes the word “hypertext” for the first time. His objective was to create a document starting from a vast whole of varied ideas, not structured, non-sequential and multimedia. He conceives also the project “Xanadu” = large, worldwide and multiple hyperbase of multimedia datas, distributed and based on a worldwide network of computers (with franchise agreements), allowing consultation, publication and recovery of information. Each document can include some flexible links, omni-directional ones or pointing towards other informations, and some sensors (informations stressing a characteristic of the document).
- 1968: Douglas Engelbart creates the NLS, “super”- word processing software including a mailing tool
- 1986-87: the software “Guides” (on PC) and “Hypercard” (on Macintosh) democratize the creation of hypertexts for a large and non-expert audience.

Fields of applications
Ben Shneiderman (in 1989) has defines 3 main rules to characterize a hypertext system :
- a big quantity of informations organized in fragments
- the fragments are connected together by links
- the user needs only a part of the whole information.

We can identify 5 categories of hypertexts (according to D. Scavetta):
- literary hypertexts (examples: Augment, Xanadu, Intermedia) = adaptation of literary works in hypertexts, with annotations’ possibilities, prevalence of the links on the internal structure of the nodes
- structural hypertexts (examples: KMS, gIBIS, Notecards) = information management and assistance to the argumentation, with nodes more important than the links and reduced possibilities of annotation
- hypertexts of presentation (example: Hyperties) = technical documentation, with separation between “author” (creation) and “reader” (browsing) modules
- hypertexts of groupware (example: Augment) = software engineering, information management inside an organization, where links and nodes are important and annotations are free
- hypertexts of exploration (example: KMS, Intermedia) = research of ideas in the processes of writing, formulation and exploration of problems with space metaphors, to work with information elements like concrete entities.

The fields of application of hypertexts are thus numerous:
- data processing : on line documentation, assistance for users, software engineering
- business : handbooks, dictionaries and reference books (encyclopedias), audits, commercial demonstrations, catalogs of products, ads
- intellectual work : organization of ideas, support for brainstorming, journalism, research
- education : self-training, supports of course, languages’ training, traditional literature, culture, museums
- entertainment : tourist guides, bookstores, bibliographical research tools, interactive fiction.

Considered prospects
In 1989, J. Nielsen imagined the following prospects for the future of hypertexts :
- in the short term (3 to 5 years) : emergence of a mass market, integration of the hypertext in various data-processing equipments
- in the medium term (5 to 10 years) : widened publication of hypertexts and hypermedias, solution for the compatibility between hypertext systems, development of editors and distributors specialized in hypertexts
- in the long term (10 to 20 years) : emergence of very large hypertexts, spaces of information shared between universities and companies, arrival of specialists and advisers in hypertexts who make a selection of recommended/official nodes and links, recording of users’ votes, impacts of non-sequentiality on the society, in particular in the training.
Complementary links to consult :
- Xanalogical structure (Xanadu project) : very interesting article about the project + link to Ted Nelson's homepage (he's still alive !) at the top of the page
- hypertext of presentation "Hyperties" : description and screenshots
- presentation of "Hypercard" (Wikipédia)

d)
This part is a small (and not exhaustive) list of hypertext/hypermedia tools, to illustrate previous conceptual definitions of hypertext.

KMS
Reference : “KMS: a distributed hypermedia system for managing knowledge in organizations”, Robert Akscyn, Donald McCracken, Elise Yoder,1987, Proceeding of the ACM conference on Hypertext, Chapel Hill, North Carolina, pp.1-20, ACM Press
“KMS is a commercial hypermedia system developed by Knowledge Systems for networks of heterogeneous workstations. It is designed to support organization-wide collaboration for a broad range of applications, such as electronic publishing, software engineering, project management, computer-aided design and on-line documentation. KMS is a successor to the ZOG system developed at Carnegie Mellon University from 1972 to 1985. A KMS database consists of screen-sized WYSIWYG workspaces called frames that contain text, graphics and image items. Single items in frames can be linked to other frames. They may also be used to invoke programs. The database can be distributed across an indefinite number of file servers and be as large as available disk space permits. Independently developed KMS databases can be linked together. The KMS user interface uses an extreme form of direct manipulation. A single browser/editor is used to traverse the database and manipulate its contents. Over 85% of the user's interaction is direct—a single point-and-click designates both object and operation. Running on Sun and Apollo workstations, KMS accesses and displays frames in less than one second, on average.(…)” (abstract of the article published by ACM Press)
KMS is a famous example of hypertext system for knowledge management in companies, to improve individual and collective productivity and to limit efforts devoted to shared databases’ maintenance. Its interface does’nt distinguish navigation and edition of contents (nodes are directly editable at the screen, like a Wiki : you can move, copy-paste, re-organize them…). This paradigm of direct manipulation and time multiplexing (very fast display on screen) is associated with collaboration tools (simultaneous and multiple access to the same project, communication tools and direct annotations).

NoteCards
Reference : “Reflections on NoteCards: seven issues for the next generation of hypermedia systems”, Halasz, Frank G., 2001, ACM Journal of Computer Documentation, Vol. 25 , 3, pp.71-87, ACM Press.
“NoteCards, developed by a team at Xerox PARC, was designed to support the task of transforming a chaotic collection of unrelated thoughts into an integrated, orderly interpretation of ideas and their interconnections. (….)” (abstract of the article published by ACM Press).
NoteCards is another famous hypertext system for workstations, to help brainstorming and structuration of ideas through a network of electronic cards. It offers various possibilities of organization, storage, information retrieval according to different ways of display, manipulation, modification and navigation. The links are named by the user, anchored via an icon towards another card. A browser represents a diagram of the structure of cards’ network and repertories propose to gather cards together. It is a 2nd generation hypertext system (the 1st one corresponds to only hypertextual systems with simple interfaces, like NLS or ZOG) which supports multimedia and offers several overall views of the network, for an individual or collective use, with a more advanced interface.
The paper of Frank Halasz however mentions several weaknesses in NoteCards and proposes solutions for hypertext systems of 3rd generation :
- System little adapted to networks of big size, broad and heterogeneous. To use preferably for on-line interactive presentations with small scales. Solution: fish-eyes views, graph flyovers, “contents” (individual nodes) or “structure” (networks) oriented retrieval informations’ system ;
- Static and split up structural model, not being able to reconfigure itself with each change in the contained informations. Not-virtual system, not-dynamics, imposing a premature organization of datas at the beginning (cards, titles and fileboxes quickly become obsolete). Solution: segmentation, naming and filling “delayed” in various ways, virtual or dynamic structures based on a description of components and a mechanism of research allowing to generate new compositions of nodes and links, to which the user can add properties (“virtual” links).
- Storage and passive recovery system. Solution: system including processes, maintenance and various computational engines
- No versioning to maintain and handle a history of modifications in a network. Solution: to allow consulting simultaneously various configurations of the same network, to offer a personalized context for each user of the same network.
- Weak possibilities of groupware : to create annotations, to maintain various organizations of the same network, to transfer messages between users in an asynchronous way. Solution : simultaneous multi-user access to the same network, notification of a material change to all users, social interactions to manage.
- Limited extensibility and flexibility : currently too generic system, not managing a work on the significance of datas and links (semantic approach). Solution: to allow adapting, personalizing, parameterizing the system with a particular task via an application interface accessible to non-expert persons.

FreeMind
This software belongs to category of intellectual hypertext systems, like you can find easily on the web (for free in general). For example (according to its official website), it helps you in many uses :
- Keeping track of projects, including subtasks, state of subtasks and time recording
- Project workplace, including links to necessary files, executables, source of information and of course information
- Workplace for internet research using Google and other sources
- Keeping a collection of small or middle sized notes with links on some area which expands as needed. Such a collection of notes is sometimes called knowledge base.
- Essay writing and brainstorming, using colors to show which essays are open, completed, not yet started etc, using size of nodes to indicate size of essays. I don't have one map for one essay, I have one map for all essays. I move parts of some essays to other when it seems appropriate.
- Keeping a small database of something with structure that is either very dynamic or not known in advance. The main disadvantage of such approach when compared to traditional database applications are poor query possibilities, but I use it that way anyway - contacts, recipes, medical records etc. You learn about the structure from the additional data items you enter. For example, different medical records use different structure and you do not have to analyze all the possible structures before you enter the first medical record.
- Commented internet favorites or bookmarks, with colors and fonts having the meaning you want.
Ideas are organized like a map including many nodes and links to different ressources.

CD-ROM : Little Women (by Louisa May Alcott)
This CD-Rom, published by in 1995 by BookWorm, is a good example of literary hypertext, based on a novel (firstly published as a book). Hypertext version gives you different ways of reading and using text and multimedia materials : annotations, clickable words (linked to a glossary), vocal reading of the text, links to additional multimedia datas (pictures, sounds, animations), margin notes, index, research tool, interactive table of contents, highlighter, bookmarks.
Please look at my Powerpoint presentation including screenshots of this CD-ROM, as examples of annotation index (slide 1), highlighting text and bookmarking pages (slide 2), annotation and cross referencing tool (slide 3), discussion editor (slide 4) and dictionary (slide 5).

Complementary links to consult :
- presentation of the “Mind Maps” method, including a video (in YouTube) from Tony Busan (inventor of Mind Maps) and a list of mind mapping softwares.
- very interesting interactive presentation (like a map !) of FreeMind (in French) : click on the differents nodes to open new parts of the map and select icons (near titles of nodes) to open new documents.
Complementary texts to read :
- “Control Choices and Network Effects in Hypertext Systems” (E. James Whitehead, Jr., 1998) : compares 3 categories of hypertext systems (monolithic ones, open ones and WWW) from the perspective of control decisions embedded in their architectures. This text gives a good perspective on 3 generations of hypertext systems we have mentioned in our course.

e)
I’ve collected below different definitions (and typologies) of interactivity in multimedia applications, according to various researchers in France in the domain of Information and Communication Sciences.
Definition 1
According to Isabelle Rieusset-Lemarié
Reference : « Esthétique de l'interactivité: approche historique », 1995, in Ecrit, Image, Oral et Nouvelles technologies, MC Vettraino-Soulard, dir., Actes du Séminaire 1994-1995, Publications de l'Université Paris 7 - Denis Diderot
Interactivity is opposed structurally to the dimension of spectacle because it moves away the spectator from any position of actor (in the sense taken in the Greek tragedy). The tragedy induces the active participation of the citizens, a phenomenon of “desindividuation” (mythical relation with the universe, allowing to transcend the individuals and to found the social body). On the contrary, according to the author, interactive television is an imitation of interactivity because it is reign of the individual, generalized insulation. The esthetics of interactivity must transform in-depth the social link to make emerging a symbolic dimension able to generate a way of being in the world. For instance, the rites make it possible to create links of a true community.
Definition 2
According to Roger Odin
Reference : « Cinéma et interactivité », 1993, Cahiers de recherche du CIRCAV, n° 3, Université de. Lille 3, pp. 19-26
4 criteria are needed to define an interactive media :
- direct contact between speaker and receiver
- interruptibility (to be able to act on course of events)
- reversibility between speaker and receiver (role reversal)
- possibility of acting on contents of the communication.
Definition 3
According to Pierre Lévy
Reference : « Remarques sur les interfaces », 1989, Réseaux, n° 33
Interactivity is a mutual and simultaneous action from two participants who can pursue together the same purpose but not necessarily. 4 consequences:
- the interruptibility (model of the conversation)
- the interaction should not be blocked by a request impossible to satisfy
- the “limited horizon line” (impossibility of knowing if the initial goal will be achieved, not too much solidified communication and interaction’s horizon line)
- feeling of an infinite database (15 choices at least in front of the user).
Then, Levy takes 3 metaphors of animals (which he borrows from Heyer) to express three types of interactivity:
- the herbivore grazes (inactive)
- the bee gathers (very selective but without a priori on the objective, systematic overflight and zapping)
- predatory seeks a specific prey.
An “intelligent” device will thus integrate 2 dimensions :
- the interactivity one (conversation, play, exploration and gathering, with hypermedia)
- the selection one (to benefit from the quantity by selecting, structuring and organizing the interesting documents among the network).
We go from devices of reception (like TV) to interfaces of selection, recombining and interaction.

Definition 4
According to Geneviève Jacquinot
Reference : "L'interactivité ou le dernier des postulats fondamentaux et menteurs", 1998, in Cinéma et dernières technologies, Leblanc, Beau, dirs, INA/De Boeck
We have entered the era of electronic “intertextuality” = direct access with images and sounds stored in memory, contextualisation at will, mix of images and various real-time handlings, simulation of virtual worlds… The digital image does not represent any more reality but translates an algorithm (calculated, exact image, without out-field). The digital image founds new methods of how to explore the seeing/hearing/knowing and is based on an intermediate semiotic statute (between “index” and “symbol” in Pierce terminology), playing an important role in structuration of knowledge, comprehension, memory and imagination.
Do multimedia devices propose a fusion between the understandable and the sensitive ? According to the author, the interactive devices do not allow oneiric consumption (implying and identification-making as in cinema). They block the emotion because of the need for acting (to select icons and buttons). We slip thus from interactivity into interaction.
It is also necessary to distinguish 2 kinds of interactivity:
- mechanical, functional, transitive one (actions on keyboard and screen) = functional level of the interactive terms (human-computer interaction)
- mental, intentional, intransitive (communication between user and software through choices of contents, structure, navigation) = meaning level of interaction’ scenarios.
It is the second type which makes it possible to develop a sensory, emotional, intellectual activity. The specificity of the interactive training experiment (with multimedia) is “to try out by”, which replaces “to live again with” (like with the cinema, TV and photography) = more directly operational relation with the learning experience, articulation between learning “by doing” and learning “by observation and instruction” between the physical and the abstract knowledge.

Definition 5
According to Jean-Louis Weissberg
Reference : « La simulation de l'autre, approche de l'interactivité informatique », 1989, Réseaux, n° 33, CNET, Paris
2 kinds of interactivity are existing :
- conversational interactivity = linguistic communication where the machine can analyze messages and send an identifiable message back to the user (anthropomorphism projected on the machine who “includes/understands” the messages).
- “interactivity of order” = role simulation, selection of pictograms, direct intervention on the image to realize the interactivity (consisted mediation by the software to organize materials and to produce the simulation effect).
Interactive communication is the relationship between an user, a software and its author = “auto-communication”, since the user realizes a conversation with himself (and the software makes it possible to activate).
The contact between user and interactive contents has 4 characteristics:
- not complete (it’s impossible to know all the possible combinations in a hypermedia)
- mediatization (no direct access to contents)
- segmentation (segmented contents)
- latency (the “spect-actor” must act “to wake up” the interactive contents).
The user becomes thus his own programmer of information taken in the stock produced by the author.

Definition 6
According to Sylvie Leleu-Merviel, Jean-Marc Laubin and Alain Durand
Reference : “Vers une classification des procédés d’interactivité par niveaux corrélés aux données”, 1997, in Hypertextes et hypermédias: H2PTM'97, Balpe, Jean-Pierre; Lelu, Alain; Nanard, Marc; Saleh, Imad, éd., Paris, Hermès, pp. 367-382
We can classify the interactive supports according to 3 types of datas :
- fixed datas,
- evolutionary datas (multiples combinations are possible),
- generative datas (not created in advance)
and according to 6 levels of “scénation” (= organized structure with which the user is put in interaction):
- audio-vision (linear and continuous diffusion: no material or conversational interactivity, but a communication interactivity) = cinema, TV
- reading (interruptible but sequential linear diffusion) = video tape recorder, books.
- consultation (indexed structure, sequential reading by request) = directories
- navigation (sequence of paths designed in advance) = tree structure
- exploration (networks of links allowing to individualize the paths) = encyclopedia
- virtual visit (not sequential diffusion, all depends on the user’s actions) = virtual guided visits.
Several levels are combinable in the same multimedia application.
Definition 7
According to Philippe Quéau
Reference : “Alteraction”, 1989, Réseaux, vol. 7, n° 33, pp. 27-46
Interaction is a potential of non foreseeable states by their author. 3 types of interaction are possible:
- heteronomous interaction (fixed laws)
- autonomous interaction (simulation of artificial life)
- “alteraction” (altering interaction, where the model controling the interaction deteriorates itself successively during the time of interaction).
Comments
You will notice that many scientific points of view have been published on this concept of interactivity. The question is not so homogeneous as for hypertext and hypermedia, and debate is always open.
Complementary links to consult :
- Svanaes, D., 2000, Understanding Interactivity: Steps to a Phenomenology of Human-Computer Interaction, NTNU, Trondheim, Norway. PhD (in public domain) : please see the chapter 2 (pp. 20-86) about additional definitions of interactivity in Human Computer Interface litterature (another scientific domain). This text is a little hard to read (it's a PhD), but it gives you other scientific points of view on the subject, not always specialized in multimedia products.