Saturday, November 17, 2018

#el30: Prepositions on the Edge of Identity

Last week, Stephen Downes assigned an identity graph for those participating in #el30. Like Jenny Mackness and Mathias Melcher, I was initially perplexed that the graph “should not contain a self-referential node titled ‘me’ or ‘self’ or anything similar”. Surely, I thought, any picture of my connections should include me, right? Then I had a light-bulb moment and realized that the web of connections, the graph, was me, and that this view of identity is in keeping with Downes' connectivism theory which says, among other things, that meaning emerges within the network of relationships (edges) among nodes rather than in a single node itself. I subscribe to this belief, but old mental habits are difficult to break. I still want to see me as … well, just me, a single, individual node. So building an identity graph could be a therapeutic exercise for me.

I examined the graphs built by others in #el30 for some clues about how to go about this—I always like a model to use even if I intend to violate the model. Melcher based his identity graph on his Twitter and library interests. Mackness used a variety of life events, roles, and locations. Roland Legrand based his map on his spiritual/philosophical beliefs and life roles. I found them all to be wonderful insights into the people who created them, but none of them clicked for me—not wrong, mind you, just no click.

For one thing, I was troubled by the edges, the links, between the nodes. The nodes at least have labels, but the links are nothing more than a thin line from one node to another. This strikes me as a serious oversight. If the meaning is in the relationships, then the links ought to mean something. In most of the graphs I've examined and the tools I've tried for generating graphs, the links are just skinny little lines. At best, they might have an arrow to indicate directional flow. That did not satisfy me.

Then, I teach writing, and I write. Writing seems to be a solid chunk of reality out of which to build an identity graph, and the chunk is definitely related to how I identify myself. Moreover, writing includes those built-in links (prepositions, conjunctions, commas, and other linking devices) that can add texture and color to the edges. Of course, most language scholars (both poetic and rhetoric) tend to favor the nodes—the nouns and verbs, or actors and actions—of writing and ignore the little words. We don't capitalize them in titles, for instance (Gone with the Wind); yet, it's the little words that connect the big words to each other and create much of the meaning, as I discussed in a handful of posts as part of Rhizo14 four years ago. Prepositions were on my mind because of some remarks by Michel Serres. In his book Conversations on Science, Culture, and Time (1995) with Bruno Latour, Serres suggests that prepositions mean almost nothing or almost anything, which turns out to be about the same thing, but that they do the critical work of arranging and connecting the actors, actions, and settings. It seemed rich at the time, but I did not pursue the ideas very far.

In his earlier essay "Platonic Dialogue" in Hermes: Literature, Science, Philosophy (1982), Serres says, "writing is first and foremost a drawing, an ideogram, or a conventional graph" (65). I do not think that Serres is speaking of graphs as we have this week in #el30—he almost certainly means something like a mark or picture—but I want to play with this connection between writing and graphs. My intent is to build an identity graph using the #el30 posts that I have written thus far. The four posts result in a fairly short 3,147 word document when the text is aggregated. I'm using Voyant tools to analyze the #el30 text, which I also used back in Rhizo14, and you can see my Voyant dashboard here. I also used a dashboard that distinguished each post here. This dashboard has some interesting data about my posts as posts within my blog, but I will not use this dashboard in this post.

Unfortunately, Voyant by default uses a stopword list to eliminate prepositions, conjunctions, and other classes of small words from its tools, deeming those words as irrelevant and mostly meaningless. The documentation for Voyant says it this way: "Typically stopword lists contain so-called function words that don’t carry as much meaning, such as determiners and prepositions (in, to, from, etc.)." I'm dismayed but not surprised. Prepositions get no respect from writers, rhetors, and grammarians. However, I intend to use prepositions as edges in my identity graph. I suspect that the prepositions and other connectors will give the links texture, color, spin, and direction that will enrich the meaning of the local connection and the network of connections.

You can see a word cloud of my posts here:

We are all familiar with word clouds, but I'm thinking now that they are proto-graphs with all the nodes and none of the edges. Thus, they are limited in what they reveal. I do, however, like the different sizes and colors of the nodes, and I think I want a graph tool that keeps the different sizes and colors of the nodes and includes the different edges. Voyant suite of tools does not quite do that—or at least, I have not found the tool that does.

So I'm following Jenny Mackness' lead and also using Matthias Melcher’s think tool – Thought Condensr. I thought I would map the top 5 words in my posts, but I managed to do just one: data, the most common major word in my four posts. The graph looks like this:

My writing over the past month reveals a preoccupation with data (the most common of the big words in my posts at 39 occurrences), and the identity graph above expresses my particular orchestration of nodes and edges that identifies me like a thumbprint. Everyone in #el30 is interested in and thinks about data, but I daresay that none have a print like mine. Yes, they have similar prints, perhaps, but not exactly this one. Just as all fingerprints have lines, arches, loops, and whorls, none have them arranged in the same way. That graph above identifies Keith Hamon—or at least a bit of him at a certain scale. This graph orchestrates drowning in data from backyard (in general, read the node/edge clusters from the blue node on the left, through a green connector, then to the red data node, and on to another green connector and a yellow node) with the other clusters of nodes and links to create a unique yet still recognizable fractal image.

Unfortunately, I cannot identify a given cluster of nodes and edges, so you can actually create new ones by following left into data and then out to any other node. You can also read from right to left to create even more clusters that generate different meanings. I think these are limitations of the graphing tools, my skills with the tools, or both. I need a graphing tool that will allow me to identify both nodes and edges and the resulting clusters and to view them in a 3-D or 4-D space. 2-D is too limiting.

I realize, however, that I've made the same error that the Voyant Tools creators did: I've put the noun (in this case, data) in the center, putting all the focus on it and building all the meaning around it. I should have put the focus on one of the prepositions—say, of with its 104 occurrences. I simply do not have time just now to graph all 104 instances of of, so I did just the first 10, and it looks like this:

Look at what a workhorse this little word of is. Consider how it connects all these nodes to create meaning at various scales, to make this particular arrangement of nodes and edges identifiably me. Consider one cluster: University of Miami. All of us in #el30 have some university node, but I may be the only one with a UM node. Even if another of us has a UM node, all the other nodes stitched together by of quickly identify me. I'm the one with a University of Miami node and a movement of energy, matter, information, and organization node. Add the 102 other of clusters, and you've pinned me to the wall. That's me.

I'm not really satisfied with these graphs, but I think they are a wonderful start to thinking about writing and how it creates an identity. I'm very happy Downes assigned this. I'm even happier that I tried it. Seems it was great therapy and substantial learning for me.

Sunday, November 4, 2018

#el30: Interpreting the Cloud

The point of the computing cloud for me has been the continued abstraction of data and services from the computing platform. I've been using computers since early 1980s (In 1982, I wrote my dissertation on the University of Miami's UNIVAC 1100), and I became a Mac user in 1987, so I am well-versed in the problems with exchanging data on one platform with users on another platform. I'm glad those wars are mostly over. I now use a PC at work, a MacBook Pro at home, an Asus Chromebook on the road, and an iPhone everywhere. The underlying hardware and operating systems are almost transparent windows to my online data, documents, and communications. I'm writing this post at home on my MacBook, but I've written posts on all my devices, including my iPhone. I also no longer ask my students what kind of device they have when I make an assignment as they all have at least a smartphone (again, I don't care which) that will let them access the class wiki and do the work. However, they do need a Google account to do most of the work.

And here is the one more platform layer that I want to remove: Google (or Facebook or Twitter). Some of the technologies that Tony Hirst and Stephen Downes discussed in their video chat (over Google, of course) seem to be taking the first steps toward separating a cloud service (say, video chat) from a monolithic platform such as Google. This continues a long progression in computing: we were freed from particular hardware, then from particular operating systems, and maybe soon from particular cloud platforms. So someday I may be able to fire up a container (made by Tony Hirst and released into the commons) on any of my half-dozen devices and hold a video chat with others peer-to-peer on their different devices and containers. I may even write my own containers for special services and release those containers into the commons where they can be used or remixed into different containers to render different services.


My understanding of complex systems is all about the movement of energy, matter, information, and organization within and among systems. As a complex system myself, I self-organize and endure only to the degree that I can sustain the flows of energy (think food) and information (think EL 3.0) through me. The cloud is primarily about flows of information, and the assumption I hear in Stephen's discussion is that I, an individual, should be able to control that flow of information rather than some other person or group (say, Facebook) and that I should be able to control the flow of information both into and out of me. I find this idea of self-control, or self-organization, problematic—mostly because it is not absolute. As far as I know, only black holes absolutely command their own spaces, taking in whatever energy and information they like and giving out nothing (well, almost nothing—seems that even black holes may not be absolute).

It helps me to walk outside for discussions such as this, so come with me into my backyard for a moment. The day is cool and sunny, so I'm soaking in lots of energy from sunlight. I've had a great breakfast, so more energy. I've read all the posts about the cloud in the #el30 feed, so I have lots of information. Of course, I'm pulling in petabytes of data from my backyard, though I'm conscious of only a small bit. Even with the bright light, I can see only a sliver of the available bandwidth. I hear only a little of what is here, and I certainly don't hear the cosmic background radiation, the echo of the big bang that is still resonating throughout the universe. I'm awash in energy and information. I always have been. Furthermore, I can absorb and process only a bit (pun intended) of the data and energy streams flowing around me, and very little of this absorption is my choice. Yes, if the Sun is too bright, I can go back inside, put on more clothing, or put on sunscreen, but really, what have I to do about the flow of energy from the Sun? And what have I to do with the house to go into, the clothing to put on, or the sunscreen. All of those things are complex systems that came to me through other complex systems (bank loans, retail stores, manufacturing factories, Amazon, and my own income streams). Most of the energy and information streams that I tap into owe little to me, not even the energy and information that I feedback.

In his post "Post-it found! the low-tech side of eLearning 3.0 ;-)", AK quotes George Siemens as saying something like "what information abundance consumes is attention", and this gets me, I hope, to a point about all this: Siemens is talking about only a tiny subset of information available to me, even though it tends to be the information that consumes most of my attention. There are other far more important streams of energy and information that I should attend to, I think.

Ahh ... maybe this is my point: even if I can avail myself of more access to more information, I'm already drowning in data. What I desperately need are better filters for selecting among the data and better models for organizing that selected data into useful, actionable knowledge. This is what my students need. Everyone in the U.S. needs better filters and models, especially with national elections on the horizon. In this sense, we are not so different from all the humans and other living creatures who have existed, except that our social systems are so much more complex and complicated than those that came before. What data do I trust, and after I've determined that, how do I arrange this data into actionable knowledge? Facebook and Google are filtering data for me now, and they are even arranging that data into actionable knowledge, but I don't think I trust them. Can the cloud help me interpret the cloud?