Wednesday’s webinar and twitter chat with Hypothes.is founders Jon Udell and Jeremy Dean — masterfully MC’d by OpenLearning17′s Gardner Campbell — gave me so much food for thought. We are starting to use Hypothes.is in the graduate pedagogy class I teach and we read “Working Openly on the Web” (7 Ways to Think like a Web) during the first week of class. So getting to listen to these three in action was a huge treat.
Our jumping off point was Vannevar Bush’s “As We May Think,” published in the Atlantic in 1945 as the imperative to leverage the technological innovations of wartime to more peaceful purposes seemed especially compelling. Bush’s vision of the memex – a computerized combination of note cards, annotations and information sources that could extend the reach (capacity) of any one learner by integrating that individual’s knowledge with the sources that informed it in a durable medium that could then be used and developed by others — underpins web annotation projects such as Hypothes.is. It also supports networked learning frameworks that facilitate collaborative learning, knowledge production and reflection.
As a historian, I’ve been intrigued by Hypothes.is since it first came to my attention last year. Historians are trained to think about how knowledge is produced and organized as an essential element of the research process: What was the author of this essay, article, book trying to say? Why was this archive created? Why are the records organized the way they are? Why did they keep what they kept? What are the assumptions behind the Dewey decimal or LOC cataloguing systems? In what context was this book, manuscript, court record, ship’s manifest created?)
Once you have a handle on those questions you need to figure out how to find where the resources you want to consult are and how to get to them. There is a dialogic process to this that involves reading, searching, thinking, taking notes, making lists, thinking, reading more, going back to your bibliography, supplementing it with new things you find, reading those things, taking more notes, thinking, going back to the older notes, etc…..I realize as I’m typing that this I might be describing a pretty generic research practice for many fields….
Anyway, at some point in there, I think two conceptual maps of a project emerge that overlay each other. The first is defined by types of sources — not so much a list, like a bibliography — but more like a grid of different kinds of evidence with points of overlap as well as nodes of distinction and empty spaces that still need to be filled in. The empty spaces let you know what you need to keep looking for and what silences your work might have to address. The points of overlap provide nuance, depth and corroboration, and the points of distinction raise new questions, redirect the inquiry or foreground a significant problem that might not have been evident when considering one source in isolation.
The second structure comprises the notes and annotations that are attached to those sources but also connected to each other (in your head or on a piece of paper or in your word processor) by the interpretation you are developing about the evidence. I see Hypothes.is as a medium through which those annotations can be assembled AND shared, which is just mind-blowingly wonderful. (Hypothes.is annotations for “As We May Think” are here.) While the analog or un-networked digital version of note taking certainly allows for all kinds of remixing and re-purposing, with Hypothes.is the annotations can themselves become nodes on or elements of a new kind of crowd / collaborative / collective “source” – a distributed conversation about a particular web page. We’re used to thinking about different kinds of sources: primary, secondary, web-based, archival, print, biographical, testimonial, etc.. Maybe a set of Hypothes.is annotations on a particular article would be a Web 3.0 source? A networked source? A memex-cubed source?
Two points in the wide-ranging Twitter chat especially resonated with me. We had been talking about how Hypothes.is helped realize Bush’s vision of “associative trails” and I asked if Jon and Jeremy saw those trails as supplements to or replacements for conventional taxonomies. Jon thought they were complementary, and Jeremy cautioned that the annotations alone might not constitute “trails” — they needed to be connected or flagged somehow, perhaps by a tag. (I like the metaphor of trail blazes.)
So, annotations become associative trails when they are marked out by tags or blazes — or any durable and accessible symbolic representation of the cognitive framework that helps you knit meaning into the tapestry (or navigate the cacophony?) of information about the world. And those trails serve as jet-packed complement to the conventional taxonomies for organizing knowledge. YES!!!!!
But how to get to the trails you really want or need? I’m imagining a future when a good chunk of the web has been trailed by Hypothes.is. And I’m imagining that all trails will not be created equal. I won’t be able to read it all, and I don’t want to fall down a rabbit hole without some warning, so how am I going to know where the good stuff is? How will the high value trails get filtered forward?
And here came the second nugget moment: Jon Udell responded to a query about this by saying “Help me grok it and I’ll help you make it real.”
I’m pretty sure I haven’t groked* it myself. But here goes:
As teachers we spend a lot of time helping students learn how to find, sort through and evaluate resources. (Crane Librarian has spoken to the challenges of doing that in the library.) And as researchers our own successes (and failures) in finding the sources and communities we need depend largely on a somewhat ineffable combination of content expertise / experience, and skill — the “scaffolding” we’re always talking about providing and developing for learners. In this sense, I do feel like I have groked the research process. But the prospect of having something so powerful and potentially overwhelming as a Hypothes.ized web makes me think I’ll need to develop another kind of sensibility and that the trails and webs marked out by Hypothes.is will need some kind of context sensitive markers to help direct individual users where they want to go. At the most basic level this would be a system whereby spam and trolls (they are, I fear inevitable) could be marginalized. But even more valuable would be a marker that would flag certain kinds of annotations — and the connections between them — and also allow for the dynamic process of ongoing annotation. What would that look like? I don’t know yet. But it would be cool. And I think it’s worth thinking about. I know I’m hoping for something that would make the web more akin to Doug Dorst’s and J. J. Abram’s book S. and would not like to see a set of user-conditioned algorithms turn Hypothes.is into a colonial outpost of my Facebook feed. It also seems that the conceptualization behind sites like Jon Stewart’s Open Note Database project could be really helpful. I’m just not sure how.
So there you go. Not at all groked, I’m afraid. But maybe glimpsed as a desirable future? Thanks for encouraging me to think about this. I will continue to do so.
*my working understanding of “grok” falls closer to the flower child sense of mastery that is so intuitive it feels innate than the techie understanding of internalizing a concept so completely it feels like second nature. But grok is also the only Martian word I know, so that might be an issue.