con§tel: sharing marginalia

con§tel: an environment to support creativity through shared marginalia


This paper examines the CON§TEL1 prototype, a web-based service that supports collective construction and dialogue within a community. This prototype explores the poetic and the dialectic relationship between reader and author mediated by a web system. This relationship takes place in a social environment where interaction design explores different ways of empowering readers —as scholars— by enabling significant collaboration.

The scholarly community formalizes its knowledge contributions through conventional text formats, such as academic papers; this standardized mediation works as a bricks for building on the top of others as a concrete manifestation of the interconnectedness of the knowledge, both within specific or multidisciplinary domains. Shared ideas and underlying relationships between texts and authors are formalized through citation and commentary; and by these means, academic papers are inscribed in a textual corpus.

Because knowledge is primarily shaped around domains or specific subject matters, we could think of all of its formal mediations as a collective attempt for creating a shared space for fixing memory. As a social construction, it aims for integrity and consistency defining shared pathways and lexicons, common fundaments and also new lines of exploration.

With this idea, I will examine other web-based systems that afford human metadata and encourage collaborative knowledge generation. This inquiry is focused in the system’s interaction design as how users are empowered for reflecting and responding over the given, what tools are available for the user’s understanding and what feedback users get as they modify and participate in new constructive ways.

(1) CON§TEL 2004-2006, e.[ad] School of Architecture & Design, PUCV, Chile. Spencer, Sanfuentes


Nos parece que la condición humana es poéica, vale decir que por ella el hombre vive libremente y sin cesar en la vigilia y coraje de hacer mundo2

The above statement, besides warning the reader about the nature of the upcoming text, it holds the vision by which CON§TEL was designed. We think of people as enablers of poetry3, not in the sense of promoters of lyrical poetry, but rather as enablers of invention, pushing reality towards the unknown, questioning and redefining language, building on the shoulders of others and permanently reinventing the world: this is our broad poetic principle. From this vision, human beings are understood as complex individuals that are capable of transforming reality.

Then, our challenge as designers lies in our capacities for enabling meaningful and constructive interaction, were people can participate in the creation of a shared vision. This project is focused on the social space of written language but more precisely, in how the personal act of reading —and making meaning out of texts through annotation— can be transferred to the public realm, turning this private and asymmetric relationship between the author and the reader into a social dialogue of peers, reflecting individual contributions into a coherent and living corpus. We are also interested in how shared ideas constitute spontaneous social organizations around communities of practice, how the act of reading can be turned into a conversation and how this conversation can be the germ of an argument of innovation. CON§TEL formalizes these relationships and tries to outline this dynamic cycle with a simple set of operational rules; therefore, it is an attempt of formalizing the complexity of interactions around of social text that aims to define the fundaments for collaborative marginalia as an epistemological environment for creation.

(2) “It seems to us that the human condition is poetic, that is to say that for it man lives freely and without end in the vigil and courage of making a world” Alberto Cruz, e.[ad] School of Architecture & Design, Chile.

(3) The word poetic is rooted in the Greek poiesis, which means turning something from not being, into being.

The CON§TEL environment

CON§TEL is a web-based platform using three integrated components: a text editor, a set of annotation tools and a typographic map that functions as visual orienting reference for the overall interaction. Initially, this platform was meant to serve our own scholar community, in the e.[ad] School of Architecture and Design. This school is founded upon a particular vision that ties craftsmanship with poetry seeking to define our identity as americans (understanting America as the American continent and not as a single country). This vision holds the community together with a consistent uniqueness and provides a solid ground for dialogue. This vision is manifested and formalized in a collection of texts —the fundamentals— that have been written by the community members during these last fifty years. CON§TEL considers people as creative transforming agents, where the initial inquiry in the reading process is interpreted as a constructionist understanding of the narrative structures in the given textual universe. The reader always transforms what he reads by his mere understanding and reflecting. Interaction that affords formalizing these kinds of responses in a social environment would create a new manner of dialogue, mediated by the openness of knowledge.

All annotations and marginalia that were earlier performed over these texts as a form of individual study, where now meant to participate of a common digital environment. The hypothesis behind this initiative was the belief that this new kind of social annotation would produce a representative and coherent figure that could reflect the current zeitgeist of the community in relation to the original founding vision. In this sense, CON§TEL is built over a paradox: how can people be contemplative witnesses of their own simultaneously performed action? This question addresses an identity issue; the user will find himself in relationships with others mediated by the textual corpus —over which he has a new power— reinforcing the image of a communal body. This paradoxical situation is also made possible because of the internal circularity of the annotation act; the system will provide immediate feedback to the annotator in a way the their contribution is displayed in relationship with the existing corpus.

A Space for Languaging

ahora está el lugar para poetrí­as ni hipnotizantes ni
poetrí­as que transforman cada momento al tacto
en nuevos momentos de nuevas poetrí­as

What is peculiar to us human beings is that we exist as such in language, the living operation space of all our interactions. We exist in the flow of living together in the recursive coordinations of behaviors that language holds inside as meaning. Language coordinates all consensual behaviors in all our specific and different kinds of human activities, concrete or abstract, manipulative or imagined, practical or theoretical; they occur as domains of consensual coordinations of doings in the different domains that arise in our living in language. So, languaging is our manner of existence as human beings5.

Before language, the first human beings begun to live by coordinating their behaviors through their consensual and innate sharing of emotions. This primarily layer of coordination is the foundational basis for language. So language was naturally established as a linking space/form between personal emotion and social response: in this realm conversation arises. From this standpoint, we could define Culture as all the closed network of conversations together.

Language, then frames our thinking and shapes our interactions. We need to be aware of the plastic nature of it so that we are not imprisoned but, on the contrary, liberated by the possibilities that it allows us to accomplish.

Our challenge is to be constantly pushing forward the consensual space of conventional language so that new interactions are made possible.

(4) “now is the place for poetry not hypnotizers nor / comforters / poetries which transform each moment into the tactile / in new moments of new poetries” Iommi, Godofredo. amereida, 74.

(5) Humberto Maturana, Biology of Language.

Language as Foundation

su tierra así­ transida
¿no expondrá en la carne
un ritmo
que mueva a lenguaje
porque sin lenguaje
todas las rutas hacia nuestra intimidad
aunque se adueí±en
deforman y engaí±an?

Social communities exist in the sharing act of communication itself, in the ways of preserving and renewing their own memory7, in the ways of storing and protecting their cultural heritage —as in their particular traditions and costumes— as well as their rules and places for depositing new contributions to their knowledge. But maybe the most distinctive attribute of communities would be the ability for inventing new and particular lexicons8.

Language offers its plastic power for inventing new words, new common places, new definitions and ideas embodied in new expressions. Some of these could be mere jargon or technical minutia, but others could be real verbal necessities fulfilling the need for new vocabularies in the community’s ever-growing lexicon. These new words inaugurate new vertices to be explored; they expand knowledge pushing it into new frontiers.

In this case, CON§TEL had to define the initial set of conditions in terms of a collection of classic readings or —fundamental stones for the memory palace— for a given community9. This set of readings is relatively small in number but, given the nature and complexity of them, they hold original statements and principles from where newer knowledge can be built from, or be in reference with: new words are inserted in this constellation of texts as new stars that draw new figures with the older drawing.

When we talk today about academic communities, the attention towards the present of their knowledge gravitates much more than any other concern10. The idea of having a centralized space for this purpose could appear as an answer to this —necessity of the latest—. Every contribution is made standing over the shoulders of others; everyone is concerned of not being redundant with something already mentioned. This memory palace11 could melt the differences between foundation and innovation, between production and reception, between the teacher and the student, between the author and the reader. I’m thinking in a reader that meets people within his reading, shares his own disquisitions with his community in a way that his own word —in a prior moment, intimate— now turns refulgent, and in this way he shares the founding lexis of this neo-alphabetized community. That’s why besides the intrinsic and particular textual dimensions implied in this creative reading process, we should reinvent the ways for marking, annotating, entitling and making new connections; also the way in which we could navigate within this environment —universe of words— and how the reader by means of this visualization could be placed at and in a knowledge that wants to be understood, shared and transmitted.

(6) “the earth so overcome / is there not exposed in the flesh // a rhythm / which moves to language / because without language // all roads to our intimacy / even as they are possessed // deform and deceive” Iommi, Godofredo. amereida, 45.

(7) John Dewey; much of his “Democracy and Education” is devoted to exploring the damage done by the assumption that knowledge is a commodity, and that the goal of education is to cram as much of that commodity as possible into a person’s memory. He claims that knowledge is acquired by the dynamic and transforming experience.

(8) This is tied back to our poetic principle and the human ability of naming.

(9) The community around the poetic vision of Amereida, The Open City Group and the School of Architecture and Design, PUCV, Chile.

(10) This is especially true in scientific communities, where building new knowledge is mostly over the shoulders of the older.

(11) Here, I use the term memory palace in the sense of a collective construction of interrelated knowledge that defines it’s own representational space: the space as the continent topos of all the multiple topics (content) of the particular domain. This use could lead to confusion because the term is widely used in psychology as a way to referring to a spatial pneumonic technique.

Social Networking Applications

Metadata creation on a broad level is a precondition for semantic-based information retrieval. In the Web medium metadata refers to the information that is laid on top of previous information. Today’s systems are progressively opening into the dynamics of social annotation and the general public getting more enthusiastic about their new potentials of influencing the web through their contributions.

But how could we construct new dynamic connections in contribution with others? Would we be willing to participate in a massive —and therefore, more anonymous— knowledge-sharing community?

To estimate the value of annotation or metadata the context of its creation always has to be considered. Otherwise metadata may be ambiguous, it may contradict other statements, it even may be wrong. The question is how individual metadata can be valuable for a broader audience while minimizing these problems.

Social networks could be one solution. If people with a shared understanding about a topic are brought together in a group, subjective information is more likely to be understood by other group members. The existence of a shared lexicon my open that possibility.

First of all, digital communities are very different from physical communities. Probably the emphasis for digital communities is upon the subject matter than rather a particular philosophy or methodology12. Further more, in digital communities every social relation would be mediated by the subject matter, because every action would be pointing towards the original argument of engagement. In this way, the invention of new words would need to be much more concise and specific, tempering the voice to a much broader community, a global one.

A folksonomy13 is the result of personal free tagging of information or objects in digital environments for one’s own later retrieval. The tagging is done in a social environment shared and open to others. The act of tagging is a very basic annotation, done with the intention of adding a personal handler to the new information. This could be seen as consuming information as well, if we believe information is commodity, but also it could be interpreted as the act of responding with your own voice as a conversation between reader and author around a specific body of knowledge.

The value of personal tagging in digital communities is about keeping independency between

  1. the person tagging;
  2. the object being tagged; and
  3. the tag being used on that particular object.

If you know the tagged object14 and the tag word you can find other individuals who use the same tag on that object, which may lead —with a little more investigation— to somebody who has the same interest and vocabulary as you do. That person can become your filter for items on which they use that tag. You then know an individual and a tag combination to follow.

In a further step, you could define your specific people as your special filterers, either because you knew them previously or because you trust and respect their way of annotating; they could be widely recognized experts, influential leaders or just colleagues or peers, but the central scheme is that you could define your social folksonomic radius. Tags in this second layer —meta— organizes information from a Boolean perspective of overlapping hierarchies, where each concept defines it own “radius of influence”, allowing much more organic arrangements around single concepts (as the new atoms of the system) in contrast to classic hierarchical and categorical way of classification.

(12) It seems that digital gathering is more about the whatness that about the howness of certain knowledge. This may be explained by the fact that physical communities are held together by the —practice of doing— which is primarily not-mediated meanwhile digital communities live through all sorts of mediations.

(13) A combination of the words folk (or folks) and taxonomy, the term folksonomy has been attributed to Thomas Vander. So “folksonomy” literally means “people’s classification management”. The features that would later be termed “folksonomy” appeared in in late 2003 and were quickly replicated in other social software. Thomas Vander Wal has stated that folksonomy is a subset of tagging and it is “tagging that works”. (edited fragment from Wikipedia)

(14) in (a web repository for personal bookmarking websites) it is the web page being tagged by one or more words or phrases.

Annotation as a Dialectic Interaction

The learning process is deeply related to reading, in terms of getting informed in the first place about the existent knowledge, and then becoming aware of the different connections between underlying ideas. These connections are made in the same reading process as far as the reader has acquired certain background understanding. This kind of dialogue between reader and text occurs in the intimacy of the studying act and usually takes place as marginalia, or the annotations in the margins of the books. Here, the reader responds to the author in a parallel discourse that runs in this marginal white space of the page: by this act, the reader comprehends the typographic symbols by turning himself from passive reader into an active author that finds new connections and completes unfinished relations15. This formal interaction embodies the internal process of the reader that generates a personal image or self-generated representation of the newly acquired knowledge.

Later, these personal annotations, now in the evolved shape of a model or diagram, will be presented for discussion in a larger group for seeking approval, validation, and as a way of contributing with a personal approach enriching the collective process of understanding.

Discreet units of meaning —concepts, headings or loose ideas— distinguished through annotation and arranged in personal diagrams act as interchangeable units within the group discussion. New boundary objects will be created from the merging of individual diagrams in a way the group is properly represented.

If we understand the dynamics of these discreet semantic units as a result of human interaction —through dialogue and conversation— we would be surprised to recognize unique characteristics. These kind of semantic units behave as autonomous entities, even though they depend on a medium for their concrete existence and material interchange (spoken words, the wall, a whiteboard or the Web); all the phenomena related to them depend on the way their autonomy is realized. Also, their autonomy is the result of their organization as systems in continuous self-production: ideas lead to new ideas, knowledge flourished from previous knowledge, the publication of a paper will lead to more papers, etc.

The problem here is that these discreet units don’t mean the same thing to all the participants of the discussion; their significance is not supported by the same emotions or previous experiences. How could me step beyond mere labels and turn them into something unique and shared in the community?

Conversation functions here as a circular process of recursive iterations, gathering the multivocal units and fostering them in fixed representations —fixed by the medium and producing a boundary object— that initially work as crossroads for multiple interpretations but, through conversation, they start to become consensual references for collective orientation.

Thinking beyond tagging —as the empty label— demands a process, the dynamic act of conversation and through several iterations.

The major structural challenge would be making visible and legible all prior dispersed personal relationships between these semantic units. The map should be a consistent configuration that works as a central and truthful reference of the community’s present. Interaction design should be focused the subtle process of this reader —that without noticing— is turned into an author and witnesses how his small contribution slightly modifies the map. The map should place his contribution and validate him amongst his peers. In the same way, the system works for double purpose: for the internal community that builds from the inside and the external observers that would easily recognize in this configuration the main characteristics and hierarchies within the community.

Nevertheless, this cartographic-textual array implies certain conditions, like the declaration of a central theme, that is to say, the definition of a monolithic universe that traces its own —frontiers of concern— shaped by the own community of practice. I guess it would be inappropriate to try to arouse encyclopedic knowledge within this kind of organization: purity of ideas would blur away because of the scattering levels of meanings16. Maybe relying on the auto organization capacity of the initial architecture would demand in a future phase the figure of an administrator or a palace custodian to keep semantic principles loyal to the original statements of the community.

The visual representation of this knowledge becomes of crucial importance; the spatial dimensions can provide us with the simultaneousness of visual narratives. But this mapping is not the result of a single author, nor the schematic view built from a single point of view but the rendering of an algorithm that strives to reflect truthfully the current state of the dialogue. This image works as a metacognitive tool that presents the actual state of the system, which is vital for providing with valuable contributions. Obviously, the difficulty here strives in designing the adequate point of view that offers the spectator enough distance to access the overall consistency of the interconnected whole. This is the invention of the space of action.

(15) In simple words is thinking about thinking; metacognition refers to higher order thinking involving active control over the cognitive processes engaged in learning. It’s the final step in the learning process; it’s the moment where the student is aware of reading-writing strategies and able of continuing the learning process by him, because he has already produced a personal image of the whole.

(16) Maybe CON§TEL needs to remain pure in the definition of its central subject matter, encyclopedic knowledge would introduce noise as lateral semantic values were the need of further context could destroy our initial semantic units, that is, our smallest chunks.

System description

El acto engendra la forma; como el trazo que, al ser puesto a la luz, orienta la normal indiferencia de las direcciones17

(17) “The act engenders the form; like a pen stroke which, put to the light, orients the normal indifference of the directions”

Following the logic of game design, CON§TEL was modeled to operate over a specific field, with a given set of rules, specifying some constraints, providing the user with fruition and allowing the unexpected as the result of collective interaction. This space for marginalia follows the annotating traditions of the scholastic medieval monks, embodied in the three main labors of the scriptorium: the annotator, emendator and illuminator.

Twelfth century scholarship developed a highly sophisticated visual form in coherent page design, particularly in glossed bibles. Glosses promoted orderly access to complex information and provided tools for medieval scholars to organize, annotate, and cross-reference between different kinds (both textual and pictorial) of complex information working alone or in concert with others. They permitted different users to contribute new information over many years while keeping track of contributors. This was accomplished by the establishment of visual structures for organizing and arranging text and comments according to their informational roles in a visually elegant and comprehensible manner.

This environment, far from being a nostalgic representation of monastic behaviors, tries to setup the basis allowing these kind of collaborative interactions between different voices and elements over the text. In that sense, the system recognizes three different actions over the text:

  1. Distinguishing key passages: The reader often highlights certain passages over the text assigning them higher relevance. By this action, we define the first textual entity of the system, the section [§]. By marking a section, the user adds it to his collection (like quotes of relevant ideas, central statements or definitions). He may use these argumental chunks later in the construction of his own texts.
  2. Adding headings: The reader also responds to the text adding personal headings through it. Usually pointing towards specific sections, these headings constitute a particular kind of tag, more complex and unique. This way of reflecting over text facilitates high-level reading and also provides of a good way of organizing the textual corpus. The heading [a] serves for handling many sections across different sources and it’s the user’s tool for administration and classification. The spatial arrangement of these elements constitutes the visualization map.
  3. Annotating: The user is also likely to engage in a deeper dialogue with the text through the insertion of notes [n]. These notes may vary from simple commentary to mayor arguments questioning or extending ideas.

Figure 1, The numeric relationships between the three notation entities that can be marked within a tabular document. The relationships define the ontology behind these —rules of the game—.

These three actions that point to three different textual entities define the relative levels within a textual string: the heading [a] is the superior level, the metatext that holds the action of labeling or defining the semantic handler for the following unit: the section [§] that embodies the medium level, which is the actual level of the text and holds the action of chunking. The note [n] is the infra level because it deepens the argument, it expand it.

Figure 2: Textual tabular structure applied to the annotation entities.

From this relationship, we define a basic circularity within the annotation entities because of their ability to entail one with another allowed by the numeric relationships defined in their ontology. This way of operating with the annotation elements opens collective dialogue into overlapping textual narratives and opens the possibility of spatial modeling.

Figure 3. The CON§TEL interface.

The graphical interface is structured as a vertical split screen where the left side is dedicated to the cartographic model. This map presents the semantic proximity and relevance of every heading [a] of the system determined by the quantity of sections [§] related and the subsequent expansions of it through potential notes [n]. In this interactive map, the selected heading will be displayed in the center; underneath the map we can see the list of all the sections related to that heading and the bibliographic information about each one of them. The right side displays the text selected from the previous list and allows the user to annotate over it. The new annotations will be reflected in the map, providing the user with immediate visual feedback.

Conclusions and Future Work

In this paper, We have explained an existing system that is being tested in a specific Design and Architecture community, relying in collaborative writing and using group consensual memory as the main construct. Adding, modifying, reassembling; all these processes of structuring, recombination and contextualization of texts no longer happen in the head of single authors but in a public writing area: new discourse technologies arise from this distributed collaborative exchange of ideas, text processing in the truest sense of the word.

There is still a great deal of research needed on what community conversation is, the interaction of collaborative annotation and how new argumetns can arise from it.

Further research is needed for defining how smaller groups inside a community would work as filtering agents for improving semantic accuracy —defining individual scopes— in a way that this model can be extended to broader audiences, interconnecting different constellations, different consensual domains around specific subject matters.


[1] Seely, Duguid (2002) The Social Life of Information – Business & Economics

[2] Bakken, Abstrakt, Liber (2003) Autopoietic Organization Theory: Drawing on Niklas Luhmann’s Social Systems Perspective, Copenhagen Business School Press.

[3] Bielenberg, Zacher (2005) Groups in Social Software: Utilizing Tagging to Integrate Individual Contexts for Social Navigation. Submitted to the Program of Digital Media, Universitaät Bremen in partial fulfillment of the requirements for the degree of Master of Science in Digital Media, 16. Aug. 2005.

[4] Neuwrith, Kaufer, Cavalier, Chandhok, Morris (1991) A Visual Design for Collaborative Work: Columns for Commenting and Annotation.

[5] Kaufer, Neuwirth, Chandhok, Morris (1994) Accommodating mixed sensory/modal preferences in collaborative writing systems. Computer Supported Cooperative Work, Volume 3 Issue 3-4, Kluwer Academic Publishers

[6] Hargraves, Haven (2004) Designing to Support the Social Aesthetics of Inquiry.

[7] Maturana (1978) Biology of Language: The Epistemology of Reality. Psychology and Biology of Language and Thought: Essays in Honor of Eric Lenneberg New York: Academic Press.

[8] Maturana (2002) Metadesign, Instituto de Terapia Cognitiva.

[9] Iommi (1980) Elogio a la Unidad Discreta, Taller de Investigaciones Gráficas, Esc. Arq. Dis. UCV, Valparaíso

[10] Pendleton-Julian (1996) The Road that is Not a Road, and the Open City, The MIT Press

[11] Hyerle (1996) Visual Tools for Constructing Knowledge. Association for Supervision & Curriculum.

 Other Project Links

Agregar un comentario

Su dirección de correo no se hará público. Los campos requeridos están marcados *