- February
Posted By : Adrian Chan
SXD: The construction of objective relations and operations

This piece has been adapted from a white paper I have in progress on the social web and social media. The paper concerns the deep relationality of social media. This is an excerpt on the construction of relations.

The world of the web is built on data that has neither fixed position nor place in terms of physical reality, but which exists by dint of its accessibility. This world of information available as you go, on an “as needed” basis, is constructed out of links. Because it depends on links for its reality and availability, it is deeply relational.

All relations are constructed and subject to modification as those relations themselves develop (or lose) connections. This world is a never-ending proliferation of references whose value is contingent on their visibility, because their visibility is the condition of their existence, and thus their use, and hence relevance.

In social media, participation in this world is always a construction of the world at the same time as it is a mode of consumption. And the social web is a “space” in which the connections formed by use are in some ways analogous to the connections formed inside a brain and from which comes mind. Subjectivity is the emergence of subjective mind — of socially relevant and valid associations. These associations among data elements employ distinctions that connect and differentiate data. Selections of data count as choices — individual user choices as well as machine-made operations.

Operations capture relations
Social media systems, like any computer-based application, perform a variety of functions. These include database queries, filtering results, and then sorting them for some kind of ordered display on the page. Functional operations may connect, distinguish, add to, or extend items of data and their associations. Users sometimes see these functions in the form of UI elements such as menu items, buttons, ratings, votes, and so on.

These operations indeed make use of calculations and algorithms, but in the social web they usually appear quite socially meaningful. In spite of their technical basis, operations are performed to produce and mediate the social, by way of a constructed presentation interface. In other words, technical operations underlie many of the constraints on social interaction online.

If these various operations are a necessary means of producing the content of social media information and content, what kinds of operations are involved and what kinds of information selections do they make? Social media are not entirely constructed out of user participation and content (communication). So then surely these operations shape the social that may form around an application. And if that is the case, then perhaps this social includes and manifests bias of selection at its foundation.

A basic associative operation is required for anything to be found (the web is built on links; data is based on links). The associative operation connects. Connections make navigation end-user possible. Votes, ratings, and other selections and expressions of user tastes and preferences all require operations.

Because the web uses links for associating objects as well as for providing navigation to them, these associative connections directly apply to visible information and content forms. In this way, associations reflect subjective preferences, or values. Captured in code and codified, qualitative selections enable quantitative operations. The ambiguity of subjective choice is sacrificed for the stability of binary selections. The social web is differently social — a cornucopia of social selections rendered coherent by selective switches.

In the formal organization of information online, and this holds for social media, most activity depends upon some very basic operations. Associative connections produce the coupling operation of a link. The operation is conjunctive: one thing and another.

Sometimes a pair remains just a pair. But a pair can be extended with further conjunctions into a series. A series of joined things might be grouped into a set, that set then being tagged (named) or categorized and thus given additional identity. Here the series is more than a series, for it has a designation applying to each element in the series. And for this the elements in the series need not be navigated by “next and next,” but may be shown in any order.

Subtraction and exclusion suffer from the fact that the subtractive operation normally leaves no record (unless data is rendered historically). So we tend to always add and join content online. Put differently, links are positive associations — the web grows connections along series.

Operations take form
There is a logic to the forms in which these operations are presented as information — communication included. This logic expresses the intrinsic (if you will) arrangement of conceptual relations among the individual elements of a pair, series, or set. Magnitude, or quantity, is the most common, and covers operations and relations involving more or less, greater or lesser quantities.

Ranking and rating, ordered lists (top tens), and so on would be examples of values arranged by magnitude. Magnitude works well to create social distinctions and to differentiate social membership. For example, it might be used to articulate popularity, importance, or trending, all of which will strike us as eminently “social.” In our culture, and in our mediated world of social news and information, magnitude is a widely-used social differentiator.

The fact that these operations are normally presented as a call to action, for the purpose of engaging users, engenders further common social practices. Adding friends, adding posts, adding links, adding tags, songs, products, become social operations when given recognizable and familiar forms. And because addition is captured and rendered more easily than subtraction, the common logic of making connections among objects online privileges increasing values over decreasing values. (And magnitude conceptually wants to increase and grow.)

But there are more operational orders available in the real world world. Operations essential to cultural expression and representation, but which are difficult to codify and operationalize online. There are, for example, symbolic and signifying orders in which one thing stands in for another. Analogies and metaphor also produce relations among items, without reducing that relation to simple associative connection. You might call this fuzzy logic — it’s the greater range of subjective association possible with mind and communication.

In social organization and in the world of human communication, there are connections that require more than a direct coupling or connection. For example, triangulation, which permits us to select one thing for the purpose of some other relation. In relational terms this is would be shown as a relation between A and B that has an effect on the relation B and C (in a set of relations A-B-C).

Triangulation is a fundamental feature of social organization, allowing for indirect relationships and actions. Human relationships are the subject of a great deal of this indirect, triangulated interaction, and while pairs, or dyads, are the basis of a meaningful exchange, it can be argued that it is the triad that forms the basis of groups and social networks.

But indirect action is difficult to represent in a medium whose basic operation is explicit coupling (linking). The common link has only one reference, not two. Triangulating communication and action can be attempted with a series of two or more connections. If action A causes action B which leads to C, for example, a series might appear as triangulation. We have gift and pass-along transactions in many social media systems that illustrate this kind of activity. But triangulating activity that simply involves the social observation by C of activity between A and B will go unrecorded.

It’s likely that future innovation will give rise to more complex and differentiated operations than we have available to us today. Perhaps the link is just the beginning; the first and necessary means of populating the world with named objects and identities? Perhaps the future of social online will extend the web built on basic associative connections and quantifying operations, using smarter social algorithms that anticipate personal and social preferences and tastes. Upon which we might imagine future design possibilities that will allow more for changes in scale, intensity, changes of time frame and interval, speed, modality (voice to text, etc), and gaming as means of capturing and re-presenting information and communication.

Example of basic associative connections:

  • Add
  • Repeat
  • Confirm
  • Join
  • Subtract
  • Couple or pair
  • Chain

Types of operations
Here is a list of some of the common selections and operations on which social media are constructed.

  • Limiting operation: this operation limits elements (people, posts, objects) displayed. Although the limit has no value other than its number (say a Top Ten list), it can have social significance (most popular). Note that nothing changes about the item, no relationships are created among items in the limit set, and no semantic assignations accrue to items in the limit set. Limit operations are good when a social or cultural form is needed and when it can be created by subjecting membership to scarcity and competition.
  • Extensive operation: this operation takes an object, element, term, label, person and extends it. Tags are extended when they are applied to objects. People are extended when they are added as friends. An extensive operation creates additional connections that extend an identity. All concepts have infinite extension (for the concept “chair,” there is an infinite number of actual chairs); online, these extensions are established through links.
  • Proliferation operation: this operation, through actual copies of digital files, through embedded players and reference, or through links and messages, proliferates an object or element. Proliferation here is different from circulation. In the digital kingdom scarcity exists in the user’s attention and time, and on the on the display (screen) itself, but not in the world of data, files, and links. Proliferation operations send an object, image, person, link, or other element around, increasing its visibility and presence. Viral operations are proliferations.
  • Increase (additive) operation: operations that increase a stock of anything by creating more of it are basic increase operations. This operation has a value for reproduction, and though it doesn’t create significations itself (addition of the same to the same has no meaning) we have a cultural bias that favors accumulation. Adding friends doesn’t change the nature of the term friend, doesn’t change the friends themselves, but does signify popularity (and for no reason other than a socio-cultural one). Whether or not the number itself (of elements) matters depends on whether the operation constructs a number or a series. If it constructs a number, the total may matter (either by signifying or by tending towards a limit). If it constructs a series, then the operation functions across time and is for all intents and purposes unlimited.
  • Series operation: this operation creates a series out of steps. Series are not sequenced, and have no intrinsic order. In other words nothing changes in the going from one to the next, logically or conceptually. The arrangement is simply a series of connections and serves purposes of navigation.
  • Pass along: this operation simply involves passing an item along. It is useful for circulation where proliferation (which involves duplication or linked reference) is not desired.
  • Scale operation: there are two operations that involve scale, one in which scaling changes the thing (non-linear), and one in which it doesn’t (linear). Most social psychological factors change as scale increases (a group of five, ten, 25, 50, 500, 5,000…).
  • Bifurcation (either/or): this operation is used in voting and in exclusive choices. It is one of the few operations in which exclusive connections are made. It is worth noting that exclusion is not visible, though quantities can show the balance of yes/no or accept/reject selections in toto.
  • Combinative operation: operations that combine and/or join items are commonplace and are necessary to cementing connections between things that are alike, similar, or related in other ways (price, location, etc).
  • Semantic assignation operation: operations that assign semantic meaning, such as categories, labels, tags, priorities, and so on, are critical for the production of meta data. Search engines wouldn’t work without this operation. Indeed, the difficulty of merging socially-constructed meanings (folksonomies, tags) and taxonomic (hierarchical taxonomies) meanings will continue to confound designers.
  • Move operation: this operation repositions an element, on the page or across pages, or among domains. Moving elements is a bit strange in that it the online world cares little “where” something “is” (how to get to it matters more).
  • Self-reflexive operation: Social media capture user input and display the results. Thus some links change what they point to according to use. Because their referent or value changes based on how many times they are clicked relative to other links point to measure their own activity, they will have changing destinations or referents. (A “most viewed member” link will point to the most viewed member, whoever that is at the time).

Bear in mind that these are relations among data selections, and not relationships between people. Interpersonal and social relationships of course also take different forms, but are not formalized or expressed by means of logical operations. Rather, they are expressed through subjective interest, action, and communication. Furthermore, inter-subjective relations involve subjective interpretation — people create relationships around relational moves that are far more complex and which are richer in meaning. (Psychological relational types include self-reference, introjection, internalization, projection, transference, and more. Social relational types are built around dyadic and triadic units, and vary in social form along axes of trust, commitment, dependence, and more.)

Relations described here and in these operational logics are objective, not subjective. Yes, they may obtain values through selections that originate with user choices and activities, but codification in data and meta data objectify relational values. It is for this reason that we find and use here basic logical associations: connection, conjunction (and, and), and disjunction (and or).

Related: I think the different types of relations described above might be useful to efforts such as the synaptic web. Combined with subjective relations formed around user relations — to other users, to their content, to activities and practices — we might map out some of the emergent subjective web’s future possibilities.



  • Adrian,

    Don't you think that relations described in your post could be substituted by implicit combinations of a few general ones? After all, everything is content on the web. Photos, text, people, everything.

    Here's an example. “Driving” is a relation between person and car. Now suppose we have a photo of dad driving his car. Obviously “dad” and “car” are parts of the photo. “Driving” is a generalization of the photo's content, and “person” is a generalization of “dad”. With two general relations (part of and generalization) I can re-define “driving” as: A is “driving” B if A is “person” and B is “car” and they compose a piece of content that features “driving”.

    By doing this I shifted the variety of connections to the already existing richness of content. Moreover, if these general relations would be defined as part of a collective action, implicit relations and hence entire ontologies could “form” on their own along the process.


  • interesting, but i'm not sure i grasp how you move from web content as
    form of content (photos, text, etc) to content as semantic content
    (dad, driving).

  • Through connections. The theory rests on the presumption that semantic attributes that we use in content conceptualization are already out there as content, too. We only have to make sure they get connected. This is where I see the brilliance of the synaptic web: in it, the semantic web could manifest as an additional layer as opposed to embedding metadata right into the content (e.g. as RDFa).

Leave a Reply