Ghost of Christmas past …

“Working sheepdogs” flickr photo by gwyn_bard shared under a Creative Commons (BY-NC) license

Whilst some hashtags are short-lived, perhaps responding to an event in the educational news (#onboardwithGonski), others are ongoing and in continuous use (#mfltwitterati). Hashtag chats like #ukedchat do the work of both marshalling regular events and serving as a point of contact for those wishing to provide resources or news targeted at a specific community of interest. In a previous post, I wrote briefly about #12daystwitter, a hashtag that appears intermittently, but which recently enjoyed its fourth birthday. This post by Mickie Mueller, the founder of #12daystwitter provides the background and history, but essentially those who join in are set a simple, daily challenge which they respond to through Twitter. The intention is that they learn about participating on Twitter and perhaps have some fun into the bargain. Here I want to delve a little more deeply into what the hashtag shepherding this apparently simple activity actually does.Read More »

Silent majority

When @MartinaEmke kindly retweeted my tweet promoting my last post, she followed it up with this provocation:

Although the people associated with the tweets was beyond the scope of that post, I must confess, it’s a topic to which I’ve already devoted some thought. How might we account for those who remain invisible because they don’t (inter)act? Approaching this with a sociomaterial sensibility, it would be too easy to claim that since there is no action, they cannot be deemed to be actors; they cannot be ‘followed’ and there is therefore little to say about them. Studies of this phenomenon, often called ‘lurking,’ have emerged which frame this behaviour in positive terms (Nonnecke & Preece, 2003; Walker et al, 2013). Crawford (2011) suggests using a metaphor of ‘listening’ as a way to conceptualise lurking. This then redefines the activity from being “vacant and empty figurations to being active and receptive processes.” However, Martina’s question seemed to require a more methodological framing; how does actor-network theory deal with something it can’t ‘see?’Read More »

Deconstructing a tweet

“The Note!” flickr photo by Smitten with Kittens shared under a Creative Commons (BY-NC) license

Maybe it’s the brevity of being restricted to only 140 characters that causes us to take a tweet for granted. Or perhaps the rate at which we read and produce them has rendered them into little more than message carriers; digital versions of the scribbled notes we might once have passed under the desk in school. Of course they are message carriers, but I’d like to suggest that, thanks to the different components we include, and to Twitter’s algorithms, they actually do much more than that.Read More »

Untangling, then becoming entangled with #NAT5HRUAE

Hashtag. Hash … tag. A symbol and a few characters.

I was pretty sure when I wrote this post about hashtags and how they were used, that it was unfinished business. When the following tweet popped up in my timeline, I knew it was time to pay a return visit:

An initial inspection of Malcolm’s tweet reveals it to be a quote tweet (QT), in which the original tweet is embedded in full (although not shown above), together with Malcolm’s comments. (As a separate issue, perhaps the QT is one way of sidestepping the 140 character limit whilst performing interesting additional work, and is probably worthy of a post in its own right?) In the embedded tweet, we see the original hashtag to which Malcolm was referring, plus two additional hashtags that he used in his own tweet. Apart from their structural difference, are they also performing different work? Before I begin to unpick that, let me first say a little about the exchange which unfolded when I asked Malcolm whether he knew anything more about the hashtag. Now that Twitter threads an exchange of tweets, you’ll be able to follow the whole thing for yourself by clicking through to the above tweet, but let me summarise.Read More »

To do is to be?

flickr photo by benwerd shared under a Creative Commons (BY-NC) license

Next Saturday is the 2016 Sheffield Institute of Education Doctoral Conference; I’m both co-organiser and  presenting a seminar. With my pilot study completed, and following a successful Confirmation of PhD seminar, I had a lot of potential topics from which to choose. In a weaker moment, I thought I’d talk about my preliminary findings, as revealed by the sociomaterial sensibility that Actor-Network Theory (ANT) is enabling me to bring. The tricky part is that I’ve been wrestling with conceptual approach all year. I guess that’s why I chose to use it to frame my talk; at some point I have to lay out my understanding to scrutiny so that any weaknesses are exposed and I can begin to do something about them. Unfortunately I only have 30 minutes in which to discuss my findings, AFTER having introduced a perhaps unfamiliar audience to ANT, using my only limited (current) understanding. Here then, with only the space afforded by a brief blog post, I’ll attempt to summarise what I intend to cover.

Read More »

Tootling along

When you approach your research with an actor-network sensibility, the one thing that you’re pretty much guaranteed to have absorbed through your reading, is to ‘follow the actors’. The principles in virtual or digital ethnographies similarly encourage you to follow connections and flows; an arguably much easier proposition in the online hyperlinked world than in the offline. It was these approaches that led me to #TootlingTuesday.

Using NVivo, I was working through my first coding pass of a corpus of tweets when a particular tweet caught my eye. A single click on the url of that tweet took me out of NVivo and into my browser so I had a better chance to see it in context. The tweeter’s bio suggested this might be someone I could benefit from following, so, following my usual algorithm, I did a quick check of their last few tweets to confirm that they tweeted interesting material. In their stream I spotted a reference to #TootlingTuesday which further piqued my interest. This was a hashtag I’d not come across before, so I clicked on it to initiate the Twitter search page. A scan through the returned tweets revealed them to mainly be celebrating or praising what other’s had done or said or shared. But I was keen to know more and see whether my interpretation was correct, so #TootlingTuesday next migrated into a Google search. Although the search results didn’t provide much background, one image which was returned helped a little:

(If you know the origin of this image, please let me know in the comments)

Different search engines were even less helpful, so unfortunately on the basis of the ten minutes I spent, somewhat ironicaIly, I’m therefore unable to credit the originator … or even the designer of the image. If I desperately needed to know, my next step would be to follow it up with some of the folks who’ve been using the hashtag.

When I reflected back, what was interesting was the way in which my actions had been influenced by the materiality within the environments. Initially a tweet appropriated my interests which took me to a person’s Twitter account, where I sought out the standard elements I always draw on; in this case the bio and the twitterstream. From a tweet within there, the #TootlingTuesday hashtag mobilised me into further action to seek its origins. I now needed to employ several search engines. Most of these acted only as intermediaries, briefly taking my inputs, but failing to transform them into anything more meaningful. Google images however became a mediator, serving up further information which transformed my knowledge and understanding of the hashtag – I was changed as a result of the output of the Google search. Are the #TootlingTuesday hashtag and I now part of each other’s actor-networks?

I find myself speculating on each of the transition points where that sequence of activity might have broken down after seeing the original tweet. If the person’s bio, or subsequently their twitterstream had not satisfied my criteria for sustaining interest (perhaps I ought to lay them out at some point?), or if I had not scrolled down sufficiently far, then I would not had seen the tweet containing the hashtag. If it had not been a Wednesday (i.e. just after Tuesday) then the tweet or one similar might have been too far back in the temporal flow of the stream. If the hashtag had not been of interest, or not a hyperlink through which I could immediately access Twitter’s search page and thereby instantly form an impression. If at least one search engine had been unable to provide a significant piece of the puzzle. Is it coincidence that these elements all lined up? Or serendipity? I wondered too about the ways in which other people are enrolled by the #TootlingTuesday hashtag and different paths they take and outcomes which result. Perhaps that’s all part of the richness and variety of learning experiences on Twitter … or anywhere else?

Finally in a more methodological reflection, one might assume that when dealing with a tweet corpus, you’ve left the field and are back in ‘the office’ analysing the data. In one sense that’s of course true, but in digital ethnography, you’re never more than a click away from being back in the field.

Why We Post – Facebooks?

An actor-network theory interpretation of Week 3’s activities on the FutureLearn ‘Why We Post’ course, prompted by a few brief comments from research participants in an all too short video.


Week 3 opened as preceding weeks have done by introducing new field sites; this week was the turn of Turkey and South India:

Here we hear the views of a small sample of the people who live there, edited together to illustrate some of the findings from the study. On this occasion, one of the major platforms is allowed to take centre stage as the participants briefly describe how they used social media. In the accompanying course discussion, several fellow members remarked how dismissive some of those appearing in the video had been of how other people used Facebook – ‘they don’t get it’ remarked a young Indian man. My peers observed how strange it was that someone might think there was a right or wrong way to use social media. Perhaps this reflects their wider experience, being aware that different people use it in different ways.

As the course has unfolded, it’s become increasingly clear how risky it is to draw from one’s own cultural hinterland when interpreting the actions or views of people from other cultures. Although there appears to be some common uses of social media across cultures, we also see heterogeneity too. Rather than imagine this Indian man having a naive(?) view of social media, I wondered whether his notion that there is a specific way in which Facebook is used might be right? Or rather, that there might actually be different Facebooks, each individual user’s practice bringing their own Facebook into being? An example of the multiple realities I discussed previously?

Here then I’ll attempt to interpret three examples from the videos, illustrating three different Facebooks – the Three C’s (because all models these days have to start with the same letter don’t they?). To do that, I’m going to explore different Facebooks, by adopting an after-actor-network theory (ANT) sensibility, tracing associations and revealing the sociomaterial. Providing a detailed description of ANT is beyond the scope of this article (and possibly beyond me!), however, the aspects on which I’ll draw here include generalized symmetry (the idea that human and non-human actors should be afforded equal status), associations (the means by which actors become entangled and continue as such), and mediators and intermediaries (the former being actors which ‘transform, translate, distort, and modify the meaning of the elements they carry’ and the latter which ‘transport meaning without transforming it’ (Latour, 2005)). I’m grateful to Leonardi et al (2012) for helping me begin to understand sociomateriality by first breaking it into materiality (‘those constituent features of a technology that are (in theory) available to all users in the same way’), then reforming it by reminding us that materiality and its effects are actively shaped by and shape social practices.

General principles

Despite there being several Facebooks, there are nevertheless some universals which we’ll interrogate by starting our journey with an individual user who we’ll call Alex. For Alex to ‘get onto’ Facebook, a mediator is required; a device of some kind. That might be a laptop, desktop computer or tablet, or as is often the case, a smartphone. Alone and isolated however, our device (and Alex) remains unconnected. Some means through which the device can become part of the Internet is required – network connectivity. This may be through a fixed, hard-wired cable system; through a local wireless network or through a digital cellular (mobile) network. The device<->connection is significant in what it allows Alex to do and what behaviour it influences. A smartphone with 3g connection is (usually) always on and (usually) always connected. Alex doesn’t have to go to the device and switch it on; it’s ready and waiting … but it’s often impatient. This state of on-and-connectedness invites another actor onto the scene. Push notifications from Facebook (and elsewhere) change Alex’s behaviour over time; the insistent ‘bing’ causing an immediate Pavlovian response to check the phone. A different push notification, delivered whilst at a computer, also clamours for attention by popping up in front of Alex’s forefronted activity, but is rendered latent when you step away from the computer to brew a coffee.

Having leapt too far ahead, let’s retrace our steps to pre-Facebook. Until signing up for an account, Alex and Facebook were apart. Was it an intermediary that briefly came into Alex’s life, initiated the liaison, then moved on? Or a friend with whom Alex subsequently connected and who continues to influence and be influenced by her? The signup process itself might be viewed as an obligatory passage point through which all pre-Facebookers must pass, but as Light and McGrath (2010) observed, there is much to be learned from probing further. The information which is sought and the options offered present different alternatives and begin to set in train the potential for different Facebooks. What details did Alex share and how private did she choose to make them? Will others be able to find her and what she posts later? A few checkmarks (or absences) at this stage will determine the Facebook she starts out with, but also hints at a more fluid conception – a mutable Facebook.

Account created, Alex now has a profile to manage, a profile which might influence other actors in different ways. Also fluid, the profile can change as Alex’s attitudes change and consequently influence the way her network develops. Her profile is only one actor amongst the many now available to translate. The search tool can be used to find other actors, be they people or different forms of content; some will be mediators, some intermediaries. While Alex is reaching out to others, others will be reaching out to her through friend requests, likes, highlights, status updates and ads which arrive by email, push notifications or pop-ups. What will Alex choose to share? A simple textual status update, a YouTube video she found, a re-shared meme from a friend? And what will the outcome be? What fresh associations will those sharings forge and will they be fleeting or long lasting? Is a ‘like’ button an actor before it is pressed? There is much to consider and many traces to follow, even from vanilla Facebook, but let us now return to our Why We Post friends.

The first example I’ll be considering is Connecting Facebook which links you with friends and allows you to communicate with them; would the word ‘traditional’ be appropriate for something only a decade old? Next is Community Facebook where people gather around a shared interest and finally Commerical Facebook which supports business enterprise. I hope to show how different practices, in Mol’s (2002) terms, generates its own material reality; three separate actor-networks. Three Facebooks.

Connecting Facebook

But Facebook is more about connecting with your old friends.. after school your friends may be abroad or you aren’t able to meet or speak even if they are residing locally then you can chat with them on Facebook…that’s what Facebook is about

This Facebook is one where communicating and connecting with friends is paramount. Status updates, messaging and the video call assemble with friends to establish those connections, and continue to do so as network associations continue to evolve through new friend requests, both outgoing and incoming. The friend request as a feature is an important actor here, but perhaps one to be distinguished from individual friend requests (denoted by default colour), which once made and fulfilled, have played their part. It’s like the agency that sticky notes confer or the activity they encourage, versus the one-off outcome of a single sticky note. It’s the (never ending) sticky note pad, versus each note stripped off, used, then forgotten

It would be helpful here to have more data than the brief video sequence available, but instead let us imagine Ajay walking to work. A bing from his pocket signifies that a friend across the city updated his status. This status update prompted the Facebook app on Ajay’s phone (and those of other friends) to enlist the phone speaker to make a sound and provoke Ajay into an action. His entanglement with other actors like status updates, messages and video calls renews his associations. Each anticipates a particular input and encourages specific and different forms of communication; one can be conducted asynchronously and extends temporality, the others demand synchronous participation; a shorter, choppier form and existing of the moment. All are fleeting associations brought into being, then are gone or are relegated by the timeline, yet nevertheless act to maintain the networks. These are the individual status updates, as opposed the feature, status update; specific messages as opposed to messaging. Status updates are a permanent constituent of Connecting Facebook; status updates are transient. Both cause other actors to do things.

Community Facebook

I created a group there … It is called Dağıskal Network Photography. There, I post the photos I have taken and the ones I edited. Or interesting things in nature… I don’t actually have a purpose, I just want to share in order to get likes.

The usual Facebook actor suspects are to be found acting here, but other actors gain significance. A group has been and is being formed (and reformed), enrolled by another (non-Facebook) actor we’ll call ‘shared interest’ (photography). While other human actors continue to be associated with shared interest, they also remain associated with group. However people may have shared interest, but not yet be members of group. A friend request, a photograph, a status update may translate them into Community Facebook, or search may have helped them seek out group.

Photographs are actors perhaps brought into being by the smartphones which are likely also significant in enacting Community Facebook. The metadata baked into digital photos may have used GIS to imprint the location of the subject of the photograph, each perhaps acting in a different way on members of group. “Where is that temple, so that I can go there and produce my own photos?” “What setting was used to produce such an interesting effect?” The subject of the photo ‘out there’ is brought into Community Facebook, present only in this artefact, but able to act through it.

Commercial Facebook

You cannot have your make up done or your hair done and not have your nails done because this is the number one thing people look at, your nails. The kind of nails people upload is well, you have Instagram now, the ones who have me on Instagram or Facebook once I do the nails they’ll say ‘can you take a pic please’ and they will upload and they will say done by Giselle …

Commercial Facebook may be one of multiple Facebooks, but itself is also multiple. We have learned that some users use Facebook pages to construct storefronts which sell their wares or promote their business. Here however, the photos others post, their status updates, enacting their own Facebooks bring into being Giselle’s widely dispersed and diverse Commercial Facebook. The materials and design Giselle uses to adorn a client’s nails are enrolled by the same client’s smartphone into photographic form and through Facebook have greater reach. The Trinidadian need (as we learned from the findings) to cultivate one’s appearance and be seen to be doing so also contribute to Commercial Facebook. The photo of the nails may translate others to become new clients for Giselle, and who will then further enact and extend this particular actor-network.


Another helpful actor-network term is black-boxing where associations between actors become stabilised to such an extent, they no longer needed to be considered as individuals. They are now black boxed as a single actor within the wider actor-network. I wondered whether to black box the device<->connection, but thought better of it. Change the device or interrupt the connection (highly likely with any form of wireless connection) and the associated elements within the actor-network are invoked differently. So no black-boxing here.

As my first ANT interpretation, I have to confess how much I struggled choosing words. This stemmed from three sources:

  • My inexperience with ANT
  • That there have been different phases of ANT. Undertaking a largely after-ANT view here, is it legitimate to use vocabulary and concepts associated with earlier versions?
  • I’m not sure yet how explicit it is necessary to be when describing the actions of actors. Do they have to be names as such?
    My worry now however is that I’m beginning to see multiplicity in everything.

Writing this has been a tortuous process for me, but one I needed to undertake to continue my journey of becoming more familiar with ANT. I felt clumsy and know this carried through into the writing, so if I have blundered do please point out my errors by adding a comment. Many thanks.


LATOUR, Bruno (2005). Reassembling the social. London: Oxford, .
LEONARDI, Paul M., NARDI, Bonnie A., and KALLININKOS, Jannis (2012) . Materiality and organizing: Social interaction in a technological world. Oxford University Press on Demand.
LIGHT, Ben and MCGRATH, Kathy (2010). Ethics and social networking sites: a disclosive analysis of Facebook. [online]. Information technology & people, 23 (4), 290-311.

Multiple worlds?

flickr photo by Swamibu shared under a Creative Commons (BY-NC) license

Just before then end of 2015 I took part in the What have the ANTs ever done for us? Packing your cases to follow the actors…discussion on the Networked Learning Conference 2016 forum. A couple of people (Steve Wright and Chris Bigum) were kind enough to point out the shortcomings in my understanding of actor-network theory, but also generous enough to provide pointers to sources which might help me “reposition my thinking.” Since then I’ve also explored a few other sources, so just wanted to test my realigned understanding by summarising some of my thoughts.

One book which is often referenced as an influential example of the application of science and technology studies (STS) is ‘Of Bicycles, Bakelite, and Bulbs: Toward a Theory of Sociotechnical Change by Wiebe Bijker. It constructs cases for three technological developments in which the social and technological elements are inextricably intertwined. Although the development of Bakelite appealed to me as a materials scientist and the development of the fluorescent tube as a physicist, it was the historical development of the safety bicycle that really captured my attention. It was also that study which provided my first conceptual breakthrough. The modern safety bicycle was only arrived at following an evolutionary path which discarded a number of earlier iterations. One of these had been the Ordinary bicycle, which we often refer to as the Penny Farthing. Bijker advocates an approach involving interpretative flexibility which asks us to consider the views of different social groups, but of two in particular. For the ‘users of the Ordinary,’ this was a difficult bicycle to master, but consequently portrayed them as riders as daring and allowed them to display their athletic skills and prowess. For them, the Ordinary was the Macho Bicycle. For ‘non-users of the Ordinary’ however, how difficult and risky it was to ride made it an Unsafe Bicycle or a non-working machine. Essentially then, the single bicycle, the Ordinary, has been deconstructed into two different artefacts, the Unsafe and the Macho. These are socially constructed artefacts with different properties as defined by different social groups, rather than being intrinsic to the machine. The Ordinary is no longer a single artefact, but multiple.  This was new for me; the two social groups didn’t have different perspectives of the same device, they actually constructed two completely different realities!

flickr photo by worldwaterweek shared under a Creative Commons (BY) license

The second source which nudged me in a new direction was The Zimbabwe Bush Pump by de Laet and Mol. The eponymous pump has enjoyed considerable success in bringing easy access to sources of water in remote African communities. It was a device that reduced the labour involved in fulfilling that most basic of needs; collecting drinking water. It was also a device which could easily be reconfigured, either in response to local needs, or when repair became necessary. A secondary effect was found to be that communities enjoyed improved health, since the water source was free from contamination, unlike the open sources previously used. To health workers then, the bush pump was a source of clean water. As a reliable and adaptable device, designed and built in Zimbabwe and championed by the government, it was also, for the state, a national asset; an emblem they could trumpet as a success. De Laet and Mol describe this as a ‘fluid’ device – ‘a device installed by the community, a health promoter and a nation-building apparatus.’ It is a different device when awaiting despatch from the manufacturers to when it is assembled and functioning in a rural village. Once again then we see multiplicity.

The final piece which helped my mini-breakthrough (much more preferable than a breakdown!) was in a text recommended by both my supervisors and the contributors to the forum discussion – After Methods by John Law. One chapter in particular helped; that which discussed Mol’s account of atherosclerosis in The Body Multiple. It is proposed that this condition has to be brought into being, and only when doctor and patient come together. Prior to this, the patient has leg pain when walking, or cold feet for example. It is only when that patient comes before a doctor, who, with their hinterland of education, skills and experiences, is able to interpret the symptoms as poor circulation in the leg arteries, which in turn is caused by atherosclerosis. In the pathology lab however, under the microscope, atherosclerosis is thickening of the intima; a layer of cells which line blood vessels (like a furred up pipe). Atherosclerosis may be produced in the doctor’s clinic and the pathology lab, but it is enacted elsewhere. The radiography lab for example, where X-rays and dyes can be used to show blood flows easily or not. Or for medical sonographers who use ultrasound and the Doppler Effect to detect differences in the speed of flow. As Law has it – “each of these method assemblages is producing its own version of atherosclerosis: that there are multiple atheroscleroses.” Methods, or perhaps more accurately, methods assemblages shape reality.

This has required me to shift from my single-world viewed from different perspectives, to a multiply produced world where different social and material relations produce different realities. As Law advises, one consequence is that if the methods we choose to use are capable of producing different realities, then there are potentially profound implications of how we deal with our research findings.

Affordances … just one little word!

flickr photo by edoardocosta shared under a Creative Commons (BY) license

At the last supervisory meeting whilst going over my RF1 submission, my supervisors asked me about one of the words I used – ‘Establish the features and affordances within Twitter that enable professional learning’ – wondering whether I realised that it was such a contested term. I didn’t! Nor had I appreciated how easily one could be undone by such a (notionally) simple term. One word, in one objective -> can of worms!

I was advised to contact someone in the University who had recently wrestled with the complexities of affordances as part of their research. I duly did and they were kind enough to both spare me a little time and to suggest a seminal paper, crucial in helping one to come to terms with this issue. In ‘The Problem with Affordances,’ Oliver takes us through the history and evolution of the term, critically examining the trajectories of development and ultimately suggesting that dropping the word might avoid the confusion and misuse which has arisen during its lifetime. This seemed somewhat defeatist; by all means be critical, but at least suggest alternatives or better options. Yet, in a case of the pot calling the kettle black, I too adopted Oliver’s dismissal approach and dropped the term by re-jigging that particular objective to allow me to submit my RF1. However, for me, that’s not the end of the story and I’d like to better understand it in order to make a more informed decision. I knew what affordance meant to me; I simply hadn’t appreciated that it meant a variety of things to different people, and as such had become problematic across a range of contexts.

A little history

Let me attempt to briefly summarise the story so far. James Gibson (1977, 1979) gave birth to the term ‘affordances’ whilst conducting work on animals’ visual perception. “The affordances of the environment are what it offers the animal, what it provides or furnishes, for good or ill.” (Gibson, 1979; 115). His view was rather positivist, seeing affordances as pre-existing within environments, independent of whether an individual perceives them, but nevertheless still related to the capabilities of the individual.

The term was appropriated by Donald Norman (1988), working in the field of design, who shifted the emphasis. He adopted a largely interpretivist standpoint in which “the term affordance refers to the perceived and actual properties of the thing.” Here we see the locus shifting from the object to the subject and how they perceive an object based on their past knowledge and experiences. In exploring affordances of technologies, Oliver contends that we have either the Gibsonian view of ‘technology as part of the environment’ and we abandon incorporating people’s minds, or we choose the ‘analytically ineffectual’ route of Norman which recognises ‘the cultural, constructive nature of learning.’

With those two streams of potential set in motion, McGrenere and Ho (2000) found that some people followed or attempted to build on the Gibsonian notion, others the Norman, some have tried to pull the two together and others used neither or failed to even define the term. They were working specifically the HCI community and in thinking about the design of objects attempted to clarify and extend the Gibson view by introducing and distinguishing between usability and usefulness “The usefulness of a design is determined by what the design affords.”

Also building from Gibson, Gaver (1991) proposed that ““Affordances, then, are properties of the world defined with respect to people’s interaction with it.” (my emphasis, to which I’ll return later). He develops the idea that  affordance and perception are independent of one another by offering different forms of affordance:

Adapted from: GAVER, William W. (1991). Technology affordances. In: Proceedings of the SIGCHI conference on human factors in computing systems, ACM, 79-84.

He argues that ‘false affordances’ occur where people perceive an affordance, but where none actually exists and that ‘hidden’ affordances are where an affordance is present, but not perceived. However, I struggle to see the value of these distinctions; in neither case does an affordance effect change. I’d argue that an affordance is only meaningful if it results in an actioned outcome. If someone perceives an affordance where there is none, then no outcome results … so there was no affordance. If someone cannot perceive an affordance, once more there is no outcome, so there was no affordance. (I’ll pick this up later)

With the proliferation of computers and digital technologies appearing in education as the 21st century unfolded, affordances spread out from HCI to teaching & learning with ICTs. In synthesising some of the research across areas such as learning design, classroom activity, introduction of IWBs, software and the WWW, Hammond (2010; 211) identifies a consistent thread running through these studies; the interaction between user and tool. Affordance is not solely related to the tool or the person, but the interaction between the two. Helpfully, and unlike many other authors, in the spirit of attempting an agreed definition of affordance, Hammond concludes by offering his:

Affordance is the perception of a possibility of action (in the broad sense of thought as well as physical activity) provided by properties of, in this case, the computer plus software. These possibilities are shaped by past experience and context, may be conceptually sophisticated and may need to be signposted by peers and teachers. However, they may, drawing on intuition and deduction from user accounts, be ‘perceived directly’, and perception of actions can precede internal mental ordering. Perceptions of affordances can, and do, become habitual. Affordances arise because of real physical and symbolic properties of objects. Affordances provide both opportunities and  constraints. Affordances are always relative to something and, in the context of ICT, relative to desirable goals or strategies for teaching and learning. Affordances are often sequential and nested in time.

Building on Gibson, this definition attempts to be all encompassing, drawing together user (and their perception), tool, properties and the wider context within which they are set. I like the comprehensiveness and that it is forged within an educational setting, though can’t help feeling it is rather cumbersome. Maybe that’s necessary to avoid the accusation of shallowness?

From my (limited) perspective

In preparation for the meeting with my university friend and having only quickly read Oliver, I attempted to articulate what my limited understanding of affordance was. I started (as any physicist might) with the properties of the object; these are naturally occurring, measurable characteristics, like dimensions, mass, density, temperature, conductivity, transparency, topology. They are sometimes inherent and fixed, but can sometimes be altered. A tree has certain dimensions, but cutting it down allows us to alter them. The tensile and compressive strengths however are (largely) fixed. The possibility of intentionally changing an object involves some element of design and here is where features make an appearance. Features are aspects of an object designed to serve a specific purpose. They are hard-wired for that purpose, yet that may be closed, having a single intention, or open and offer multiple possibilities. An affordance however is what the object, property or feature allows you to do. They involve action, are open to interpretation and depend on the user. A single property or feature may provide multiple affordances, whilst a single affordance may require the assembly of several properties or features. Returning now to the offending objective in my RF1 which set this in motion – “Establish the features and affordances within Twitter that enable professional learning.” At that stage I hadn’t thought about properties, but based on what I’ve just said, since Twitter is a designed environment, it is features that will matter. What I would be looking for then is elements of the design which enable professional learning; those features might be general and include the ability to share information, to connect with other people and to view what they share, to participate using mobile technologies. Or they might be more specific like the 140 character limit, the profile, the retweet, the URL shortener. Others might view these as affordances, but for me they only become so when a user perceives a use and subsequently is able to commit an action … which ultimately leads to them (or someone else) learning professionally. For example, one feature of Twitter is the capability to embed a thumbnail video in a tweet. Let’s say a physics teacher finds a video which explains Newton’s third law and she chooses to share that through a tweet. The feature became an affordance when she knew she could share the video and chose to do so. Had she not known the feature existed (‘hidden’ in Gaver’s terms) or had she chosen not to share the video, then the feature remains just that and never becomes an affordance … for her … at that time.


When our physics teacher turned the feature into an affordance, that set in train a chain of events leading to multiple potential outcomes. Furthermore, using the same feature of video sharing, others might perceive and enact rather different affordances. This then presents an interesting possibility; that a feature might provide multiple affordances – different people, different intentions … and different enactments? The single video shared by our physics teacher becomes multiple when incorporated into different learning opportunities by different people. Alex may view the video and see an opportunity to teach Newton’s’ third in a different way; Bobbie may want to explore the ideas further so uses the video to stimulate discussion in a departmental meeting with colleagues. On the other hand, an English teacher may use the video sharing feature to share a poetry reading with his class, and subsequently reflect on the difference it made to his students’ learning compared with when he used to read the poems directly from the book.

That single feature, to be able to share videos, is used with different intentions, so provides different affordances. This results in multiple realities (de Laet & Mol, 2000), since the (single) video sharing feature affords different possibilities for both those who share videos and those who consume them. Might actor-network theory help me to untangle these interwoven threads?

When I was first prompted to think more carefully about affordances, I never anticipated I might need to draw on an ANT sensibility, but the fact that the two concepts seem to have something to contribute to one another is promising. Or should I be worried? I know my views are likely to leave me challenged, as Gibson was by Oliver, as providing only illustrative examples, rather than a definitive, all-encompassing means to define affordance. I’m also only too well aware that my understanding of ANT has a long way to go, but then I am only starting out at understanding what affordances and ANT, together and separately, might or might not mean for me.


Although lacking the awareness that affordances, as a term, carries baggage, at least I don’t appear to be alone. Papers like “The affordances of social media for inclusive urban communities” don’t even use the term beyond the title, let alone unpick it. Or “Social Network Sites as Networked Publics: Affordances, Dynamics, and Implications” which mentions it over forty times, but never defines it. Perhaps ‘affordances’ has indeed been rather too loosely accepted?

In addition to one of those occasional little insights, this episode also brought home to me how generous people often are in academia; keen to share their knowledge and help you develop your understanding. My friend needn’t have given up any of their precious time and could have rested at simply providing me with a link to the key paper. Instead they chose to sit down with me, listen to my naive and incoherent ramblings, yet remain patient and offer insights at appropriate times. How rewarding.


DE LAET, Marianne and MOL, Annemarie (2000). The Zimbabwe bush pump mechanics of a fluid technology. [online]. Social studies of science, 30 (2), 225-263.

GAVER, William W. (1991). Technology affordances. [online]. In: Proceedings of the SIGCHI conference on human factors in computing systems, ACM, 79-84.

GIBSON, James J. (1977). The theory of affordances. [online]. Hilldale, USA, .

GIBSON, James J. (1979). The Ecological Approach to Visual Perception: Classic Edition. [online]. Psychology Press.

HAMMOND, Michael (2010). What is an affordance and can it help us understand the use of ICT in education? [online]. Education and information technologies, 15 (3), 205-217.

MCGRENERE, Joanna and HO, Wayne (2000). Affordances: Clarifying and evolving a concept. [online]. In: Graphics interface, 179-186.

NORMAN, Donald A. (1988). The psychology of everyday things. [online]. Basic books.

OLIVER, Martin (2005). The problem with affordance. [online]. E-learning and digital media, 2 (4), 402-413.

A ‘like’ly story …


It’s likely that even those not on Twitter have become aware of one particular recent change. The small button previously used to ‘Favourite’ a tweet has changed to aLikebutton.

The favourite/’like’

Every tweet that flows past a user in their twitterstream has a number of action buttons appended to it. One of those icons took the form of star; this was the Favourite button. If you were so inclined, clicking on the star icon would indicate that you had favourited something; the star turned from grey to gold and the tweet you favourited would be stored in a list called, unsurprisingly, your Favourites. At any time in the future, clicking on your Favourites would allow you (and anyone else following you) to see a list of all the tweets you favourited. In addition, the person who’s tweet you favourited would receive some form of notification that one of their tweets had been favourited (how the notification occurred depended on their chosen settings).

The same all still applies now that favourites have become ‘likes.’ The functionality is all still the same; the only change is to the icon – from a star to a heart. Some feel the change is little more than cosmetic, although acknowledge that people might use the heart with a greater variety of intent, seeing it as providing greater flexibility than the star. Others see the heart as somewhat restrictive, constraining one’s actions to merely positive affirmation.

Introducing actor-network theory (ANT)

Wanting to test what I’ve been learning on ANT recently, this Twitter change provided an opportunity. A small step on the road to a better understanding. So I’m not looking here to produce a Grand Theory of how Twitter works from an ANT perspective, but to focus on one small aspect and see whether my interpretations are valid.

For a detailed explication of what ANT is, you need to look elsewhere; this post will be long enough! In short (if that’s even possible!), ANT encourages us to look beyond society composed of people, their culture(s) and interactions. The social is instead, associations enacted between humans and non-humans; where non-humans (objects, technology, texts, language, organisations) have parity of agency and should be discussed in the same terms – the principle of generalised symmetry (Callon, 1986). These sociomaterial associations constitute actor-networks; neither actors connected through networks, nor networks composed of connected actors. A melding, rather than a mixing; reminding me of lower school science where we discuss mixtures (easily formed, easily separated) as distinct from compounds where the components are inextricably linked. The analogy is far from perfect however, as actor-networks are in a constant state of formation and re-formation as they are continually performed.

There are many principles within ANT, but the one I want to bring to bear on our Twitter development is the distinction between two types of actors; mediators and intermediaries.

Intermediaries play a more passive role in actor-networks, maintaining the associations without being transformed, or transforming others. Mediators on the other hand “transform, translate, distort, and modify the meaning or the elements they are supposed to carry.” (Latour, 2005, p72). They actively engage others, recruit them, influence them, and seek to bring about change. Through mediators the actor-network is reinvigorated and regenerates.

ANT is often discussed in the abstract … and I don’t mean the bit at the front of an academic paper! Let me attempt a more concrete example situated in an educational context. Consider the actor-network formed when teachers are charged with delivering a new qualification; since I’m in England, let’s say a GCSE. For simplicity, we’ll leave out of our actor-network the governmental bodies, the documents they produce, the examinations boards which translate government policy into assessable curriculum and focus just on the teacher and what they work with. Originating from the National Curriculum Framework, a teacher is most likely to use a examination board ‘specification’ which sets out the curriculum to be covered and what and how that will be assessed. The specification document initially will be a mediator, charging the teacher to devise an appropriate teaching strategy for her students. This is often done by creating a scheme of work (SoW) – a document which maps out teaching approaches, resources, assessments, a timeline for delivery and so on. This becomes the working document; a mediator in it’s own right which continually influences the teacher’s actions and activities. In return however, she will also amend the SoW based on the way the activities work with her students; a little more time needed here, a different resource there. The teacher and SoW are mediators, but the specification is now relegated to the background and becomes an intermediary; neither capable of being altered, nor no longer influencing the teacher and the SoW, other than to provide the touchstone to which both will occasionally refer.

A ‘Like’ly Story

Twitter has an expectation of how people should use the ‘like’/favourite feature:

‘Likes’ are represented by a small heart. They are commonly used to show appreciation for a Tweet. You can see someone’s ‘likes’ by visiting their profile, and your ‘likes’ are also visible on your profile.

Thanks to Nick Lewis, we can also see how that has changed from Twitter’s original intent:

Favorites, represented by a small star icon next to a Tweet, are most commonly used when users like a Tweet. Favoriting a Tweet can let the original poster know that you liked their Tweet, or you can save the Tweet for later.

So Twitter has in one sense attempted to refocus the purpose of the ‘like’. The different semiotic associations of a star versus a heart perhaps go some way towards supporting that. More importantly, will people’s behaviour change as a result? Will the heart change the behaviour of those who have always used the favourite feature or will it be business as usual for them? Might the heart enrol some who never used favourites in the past to start doing so?

The beating heart

To begin to unpick some of what was mentioned in the previous section, it is worth mentioning that what Twitter intends for the ‘like’, and what users enact, might be very different. That was certainly true for favourites. Chris Lake and those who replied to his post have provided a number of different ways in which they use favourites:

  • To like something; an upvote.
  • To dislike something; sarcastic use.
  • To bookmark something.
  • As a note to self.
  • To read later.
  • To read something somewhere else; on a different device.
  • To trigger something (IFTTT)
  • As a read receipt; an acknowledgement that you’ve seen something.
  • As a conversation stopper; to indicate that an exchange has run its course.
  • To doff one’s cap; well-played sir.
  • To flirt.
  • To show support.
  • To spread the word.
  • To attract more followers.
  • To build a personal brand; where tweets on a particular topic are curated to indicate something to others.
  • To reflect one’s personality or interests.
  • As a polling tool: favourite for this, retweet for that.

Rather than viewing the favourite (or ‘like’) button as an intermediary or mediator with only a single function – an indication that you like the tweet – the community of Twitter users have co-opted the button to serve their own needs. In some cases the button becomes a mediator, itself calling others to action (spreading the word, polling opinion, influencing others to follow or interact with the originator). In other cases it is an intermediary, simply holding information for future reference (as a bookmark), or in making a temporary connection (showing support).

With ANT however, we’re not yet finished. Although some people write in their bio profile that ‘favouriting (or RTing) is not necessarily an endorsement,’ thereby providing some small indication of their intent, that does not tell the whole story. We have only considered the button from the favouriter’s perspective, how is the action perceived by the originator of the tweet? What might mean something to the favouriter, might be interpreted completely differently by the recipient. Having different conceptions means that the notification one person receives may initiate one form of behaviour – “This person favourited my tweet. I’ll check them out and see if they’re worth following.” Which may mean they go on to follow the other person and thus reform their actor-networks as a result of the mediating effect of the favouriting. Alternately the recipient of the favourite may be perplexed by the favourite – “Why did they favourite that tweet?” Which may be either intriguing and lead to further action (favourite as mediator) or nothing more than a distraction and the actor-network of favourited tweet and favouriter extends no further. The favourite button in this instance acts only as an intermediary.

A 'like'd' tweet from my Twitterstream
A ‘like’d’ tweet from my Twitterstream

Can the ‘like’ button be interpreted in a similar way? That will depend on how people react to the change of icon. If one simply wanted to indicate ‘well done’ or ‘well said,’ does a heart symbol align with that sentiment? How will recipients of a heart view it? Might their action be different than if they had received a star and will that change the nature as either intermediary or mediator? What is clear is that people are currently seeking to reduce the likelihood of ambiguity or misinterpretation by ‘liking’ then following up with a clarifying tweet – ‘Hearted that because …’ Perhaps there’s an opportunity here while people are making their actions explicit, to better understand liking behaviour … at least until the dust settles.


So what did bringing an ANT perspective to the Twitter changes add? I suspect in this instance, not a great deal, and there are a couple of possible reasons. Firstly there’s my currently underdeveloped understanding and inexperience in using ANT. Hey! Everyone has to start somewhere! However, and perhaps more importantly, this situation might have had the ANT lens focused rather too narrowly on a single, tiny button. This post lacked the scope to go beyond that self-imposed limitation and consider the wider actor-networks within which the favourite/’like’ button acts. As an exercise in ANT-like thinking though, at least it got me off the start line.

Callon, M. (1986). Some elements of a sociology of translation: domestication of the scallops and the fishermen of St. Brieuc Bay. Power, action, and belief: A new sociology of knowledge, 32, 196-223.

Latour, B. (2005). Reassembling the social. London: Oxford.