Thanks for a great Tutorial. I need to distill this down perhaps, but it just felt as if there was so much to write about at this point.
We find ourselves in a bit of a potential algorithmic ‘loop’ here, interrogating an algorithm(s) which purports to represent our Tweetorial upon this very topic. We must in some sense perhaps have to ‘reach into ourselves’ in a significant way, also therefore as a part of this reflective act. A starting point for my reflection upon the Tweetorial of last week then, is to consider what is perhaps not captured in the representations of the Tweet Archivist and Keyhole . Ben Williamson (2014) phrases this (in the slide from his lecture above) as algorithm as synecdoche, as an abbreviation for socio technical assemblage, technical tricks, human hands and public discourse. This sums up nicely how I experienced and visualize our Tweetorial as an interlocking pattern of events. I am less certain that the visualization represented in the Tweet Archivist (TA) and Keyhole (KH) reports capture these nuances, much of what is important lies in what is not visible in this type of representation.
This point relates perhaps to the actual nature, or sense of what Williamson describes as the ‘public discourse’ (2014) taking place itself within the Tweetorial, and in some important respects therefore our ‘human hands’ (or intentionalities; ‘hands’ actually seems an odd metaphor here), and this is not something which I can see in the straightforward quantitative representations of the menus of TA and KH: Top users’, ‘Top words’, Top URLs’, Source of Tweet’ and Top influencers.
My point is this, what is the story behind these statistics? What led me to do what I did, and what were my intentions and indeed the intentions of others? I have already given some specific detail on this aspect of things in my previous post.
For the present then the observation is not that the algorithmic visual representation of the TA and KH are not relevant, they are, but like points on a map, they don’t tell us necessarily what journeys have been undertaken (between points), what precise routes were taken between which points and why, or what was experienced, seen, learnt and understood along the way. This would not be, unless a Tweetorial protagonist posts a blow by blow Tweetorial narrated ‘livestream’ (not difficult to capture of course, as an aspect the browsing history and automate its posting, at even a simple level IFTTT would do this perhaps, and there are much more powerful algorithmic tools out there which time precisely for example how much time you spend on a page, and which search results you access. However, do even these capture some of the qualitative points made with respect to learning above?), which could provide more insight: a live blog per say.
We were discussing of course algorithms, and their affective play within any such journey, but specifically it seemed the aim was the interrogation of algorithms, and so it seems fair to note (simply from my standpoint) that to accept the surface dimension of the TA and KH reports is to acquiesce to what Williamson, in describing algorithms: are the accepted procedure: privilleging quantification and automation in human endeavours (2014). The ultimate outcome of this is to succumb to the psychological impulses, as I inevitably did on first viewing the reports, and to seek any potential significances in the construction of what Pasquale describes as the ‘algorithmic self’ in ‘The Algorithmic Self’, (2015). He has this important insight to bring to bear:
‘Gary Shteyngart in his 2010 novel Super Sad True Love Story, In an anomic world where social mores are adrift, the characters in the novel scramble to “find their place” in the social pecking order by desperately comparing themselves with each other. No one dwells on what these matrices signify or how they are calculated; they just want high ones.…. Black-box rankings become a source of identity, the last “objective” store of value in a world where instability and short attention spans undermine more complex sources of self-worth…’
Hence in looking at the TA and KH report I first breathed a sigh of relief, I was mid ranking for number of posts and lower third for engagement (not a complete disgrace). As Sian noted wryly, at the outset of this task the TA shows us ‘who was present and who was absent’. There is an odd divergence in ‘Top influencers’, I’m apparently top on the KH algorithm (unlike the TA), defined by the highest number of re- tweets. Firstly should I accept this (no)? Secondly, does this ‘fact’ confirm to me that there was something subtly more effective in my approach, by my design- that I have unearthed more significance, somehow in the way I approached the tutorial? Should I feel I have achieved as much as I could in terms of learning and understanding and there is an over all epistemic value of ‘my journey’. I think not, I cannot read off such ‘facts’ from such ‘facts’ or even assume in pedagogic terms there are certain insights from this statistic.
To this extent I confer with PJ, with his observation with respect to the number of Tweets recorded in the TA and KH reports, that this yields no similar necessary significance. This is not to say that the approach that I took wasn’t productive or enjoyable, but just that there are omissions in the story or explanation of why this is the case, in terms of the TA and WH representation. Turning again to Pasquale (2015), the consequences of conversely following a path of inference where one accepts the algorithmic representation of TA and WH simply in terms of their surface dimensions might therefore reduce our ‘sense making’ in terms of learning to a self- readjusting as a result of algorithmic interventions (2015) which are designed and have the effect of funnelling behaviour and thought into ever-narrower channels’ (2015).
Jeremy Knox (2014) in Abstracting Learning Analytics seems to express something of what I mean in talking of a myopia in the kind of algorithmic representation found in TA and KH (he uses Course Signals as an exemplar for an algorithmic representational LA scheme), if we consider it to be an analytic tool for expressing or measuring learning, and if we understand as what I describe as ‘my journey’ during the Tweetorial as processes, and crucially not just as computational ones: Knox 2014:
What are the broader societal and economic factors that produce an educational concern for retention over that of enjoyment, for example, and how is the image of the traffic light amplifying this concern?.
What is being asked therefore is whether the LA rubric or algorithm’s processes of analysis adequately capture, reflect and synthesize these processes. Jeremy Knox (2014) also tells us:
‘I think we should focus less on the results of Learning Analytics, and whether they measure up to reality, and more on the processes that have gone into the analysis itself. Understanding these processes, I contend, is as crucially important in understanding the ‘realities’ of education in our current times.’
The interplay of algorithm and agent (learner) and learning is complex and distributed, and the processes and the relations, necessarily perhaps, abstract in the way in which they can be conceptualized and expressed, as Knox (2014) seems to intimate. I struggled in my initial post ‘Disciplined By Algorithms’ to find an adequate visual representation, or medium for my algorithmic play. I perhaps understand better now why. The image below from Williamson (2014), in talking of the future of social research, comes closest to how I visualize the relationships within the journey and processes within the Tweetorial experience, it appears to be one which is non reductive, with a multiplicity of players, human and computional (algorithmic). As Williamson notes the implications of this state of affairs is unresolved.
As a footnote, I would like to acknowledge that the article from Jeremy Knox (2014) ‘From MOOCs to Learning Analytics: Scratching the surface of the ‘visual” was the source of my ‘Top Influencer’ ranking (such as it should be understood as such at all), so surely the content of the Tweet, as a reference (and therefore its author) should be considered the key agent of influence here. It is interesting to reflect that Williamson (2014) notes that there is a specific correlation between top tweeted articles and top cited articles, as illustrated in his slide below, although of course this is just glimpsing in itself:
