From concrete to fluid, but not yet to social

Lectures are often static objects. The size of the audience, time constraints, and the need for conference organisers to keep some editorial control make sessions mostly about transmission (“this is going to be just me speaking now”) rather than engagement (“I’ll guide a discussion through which you’ll learn”).

Although the audience size and format are often quoted, the traditional static nature of lectures was also connected, and in no small measure, to the way visual material was prepared. Slides (of the old kind, not digital ones) were expensive things you’d have to plan very carefully. Object needed to be photographed (expensive and slow), and slides developed and mounted (ditto). If you wanted text on slides, you’d need to prepare the text separately and shoot the slide on a rostrum (yet more cost and delay). Once lectures were delivered, they would either remain in the memories and notes of attendees, or be published as pamphlets or transcripts, usually omitting the images. The situation for lectures that are preserved as podcasts is similar.

My oldest lecture with PDF slides is from early 1999. I still remember the elation of being liberated from film. Scanners and Acrobat made planning of versions for different lengths and audiences orders of magnitude easier, and eliminated many costs. And working with text in slides became trivial. Yet, while the speed and flexibility of building lectures improved dramatically, the format of the lectures changed very little. From the point of view of the audience, the only difference is that transitions between slides were much faster and smoother, and that it was possible to linger on a single slide for many minutes, since doing this with a transparency risked burning the slide. Regardless, the transition to PDFs did not change the structure of the lecture from a somewhat rigid narrative punctuated by images.

The gradual adoption of presentation software like Keynote and PowerPoint for public lectures (because teaching environments are a different case; another blog post)  precipitated a shift to lectures being structured as sequences of images with annotations attached to them. The ease with which presentation apps allowed tree-style outlines to be built or imported strengthened this trend as a way to compose a lecture.

 

two lecture structures
From a single script with reminders to change the slide, to a series of slides with annotations.

For the speaker this means a greater number of slides, since each point in the narrative needs a slide simply to exist, let alone be elaborated. For the audience this is a Good Thing, since it adds a visual dimension to explanations that would otherwise be left to words alone. This is pretty much where we are today; tools like Prezi do not shift from this model. (Sliderocket offers collaborative functions and tracking elements, but these are intended for internal teams, not public engagement.)

But whereas lecture composition and delivery has (sort-of, if you’re in a generous mood) kept up with developments in content authoring, it has not progressed much in adding value to a lecture after it has been captured. We have acceptable records of what was shown and said at the time of delivery, mostly by sticking a camera in the aisle so that both the speaker and the slides are in the frame, or by splicing a talking head in a frame that is mostly taken over by the slide, or by trying to switch between the two. From the speaker’s point of view, the best you can hope for is a separate feed for the audio from the  microphone, instead of the camera’s own.

GL at ALUO talk, Ljubljana 2013
Close but no cigar, v.1: Less that half of the frame is important, images are skewed, and details may be lost. (From http://www.youtube.com/watch?v=oTMop3DUvs8)

Both these options exist in silos on YouTube or Vimeo, mostly. Although comments on the video’s page are possible, these stick on the page of the video, and by default refer to the whole: comments cannot link explicitly to a point in the stream.

River-Valley.TV slide
Close but no cigar v.2: both speaker and slides are visible all the time, but the interaction between the two is lost. (From http://river-valley.tv/e-book-trends-an-industry-analysts-viewpoint/)

Depending on the how the speaker uses the lecture slides, posting the deck on Speakerdeck or Slideshare may be anything from very useful to utterly confusing. Some speakers use their slides to illustrate points and punctuate their talk: they conceive the narrative as a combination of verbal and visual content in sync. But these decks tend to make little sense on their own, since the speakers’ explanations and bridging sentences are missing. (A “bridging sentence” spans two slides, and is used to join the transition to a new visual message with the verbal narrative.)

Tribute to Adam
Close but no cigar v. 3: “What’s this guy doing here? What’s the speaker’s point? By the way, nice jacket.” (My slide, from a TypoLondon 2012 talk)

On the other hand, speakers that use their slides as a record of the argument trade a less engaging presentation for a more useful record of the talk’s key points. This category of decks spans anything from a few sentences on a slide, like this:

Slide from a W3C workshop
Close but no cigar v. 4: “Why do I need to hear the speaker say these things, if I can read them already?” (My slide, from a W3C workshop panel, 2013)

… to semantic soups that make your head spin and scream “FFS, what where you thinking?!”

NSA PRISM slide
Close but no cigar v. 5: It really is impressive that any eavesdropping happens at all if they use these for training.

Both video capture and deck publishing are undeniably useful. But they are closed objects, with very limited scope for interaction and cross-referencing. Especially in non-academic circles, where a talk is not an exposition of a scholarly paper, the video or slide deck may be the only “text”. Speakers may transcribe their points in blog posts, but then the text in the blog post encapsulates the ideas, not the talk itself.

It is also possible to take a deck as a starting point, and annotate it in a way that it becomes a more-or-less self-contained text. I tried this with my latest talk on the relationship of tools and innovation, delivered in Warsaw a week ago. The slides went from 67 to 93, and the word count from 590 to 1,330. This is an experiment to compare the reach of this deck with other decks that were uploaded within minutes of delivery, warts n’ all.

Warsaw Design Debate uploaded slide
A slide that was projected during the lecture.

This was the slide that I added immediately after, in the uploaded deck:

Warsaw Design Debate uploaded slide
This slide summarises the explanations of the four points in the previous slide.

In some slides, I added text on the original slides:

Warsaw Design Debate uploaded slide
The text in white was presented in the lecture; the darker text was spoken, and added in the uploaded version.

So far so good?

But a good lecture generates commentary, both during its delivery and after it has been published. While a lecture is being delivered, things are happening: people are reportingcommenting, expanding, and even making old-style notes:

speaker evaluation

Services like Eventifier or Storify can build a partial record of an event after the fact, but they are not optimised for the smaller scale of a single lecture. And they primarily compile what’s already out there, without the functionality to edit the results or comment on specific parts. Even so, these results are not linked back to the lectures themselves, let alone the moment the tweets were posted or the images taken.

Worse, if someone writes a coherent and engaged response to a talk (like John D. Berry did for my Ampersand talk) this is isolated from the source, whether it exists on video or slide deck. Or any other part of the discussion the talk might have generated, for that matter.

Not very “social”, then. Events that are, in essence, starting points for discussions and catalysts for ideas, become fragmented, flat sets of disconnected objects.

So, what then?

A good lecture is a story with convincing arguments. A great lecture will leave the audience with new ideas, and set off ripples of discussions and further “texts”. Ideally, all these things are connected, and become part of a collaborative document. This is what citations do in the academic world, and what links do online. It seems paradoxical that we have easy ways to connect verbal hiccups, but do not have an easy, robust, and open way to link within lectures. Considering the effort that a good lecture encapsulates, this is pretty wasteful.

I don’t know if this platform exists, but here’s my back-of-an-envelope model for a slide deck viewer; obviously only one slide (and the discussion it generates) are viewable at a time:

 

Model of a slidedeck viewer
A vertical timeline, with author content on one side (slides and annotations) and social content on the other (comments, tweets, links) arranged alongside a specific slide, or a span of slides.

For this to work every slide would need to have its own URL, but that should be really easy. (So, my slides above could have addresses like
speakerdeck.com/gerryleonidas/2013/11/23/tti/slide28
and a comment
speakerdeck.com/gerryleonidas/2013/11/23/tti/slide28#http://twitter.com/username/status/01234567890123456789

For a video talk, something like this:

Model of a video viewer
Author annotations (if supplied) below the image, appearing according to timestamps. Below, a timeline with a liveblog-style scroll of tweets, in sync with the timestamp they relate to. And, next to them, a column with links and external references.

If there’s an easy way to link to a specific time point in a video stream from within a comment or a tweet, and collect all that together, I’ve missed it. But I’d like to be able to link to
vimeo.com/clearleft/2013/06/28/ampersand/gerryleonidas/20-45/.

You get the picture.

Any takers, internet?

DeL 2012 paper: Distance learning in archives-rich environments

Proposal for the paper delivered at the Designs on eLearning 2012 conference in the University of the Arts, London, talking about the g MA (Res) TD programme. The description below is in academic-speak, but is useful for those interested in how we’re approaching this.

 

Developing a New Model for Distance-Learning in an Archives-Rich Discipline

This paper describes the challenges in the development of a new programme targeting distance learners in a domain where conventional literature is not easily available, and engagement with original artefacts is essential for the research skills.

 

Context

Typeface design is a design field that has experienced considerable growth in the last decade. Central to this growth have been the strategy of OEM suppliers to support global markets without localising instances of their products, and the shift to region- or worldwide branding by major companies. The specialised skills required for high quality multi-script typefaces exclude autodidacts, and underline the need for structured education in multi-script typeface design. The University of Reading has pioneered teaching in this area through a very successful full-time residential MA programme in the Department of Typography & Graphic communication, whose graduates occupy dominant positions in the industry. The programme has inspired similar initiatives at postgraduate level, most notably in Argentina, Mexico, and Switzerland. Teaching relies heavily on the use of artefacts from the Department’s Collections & Archives, and particularly the Non-Latin Collection. The Collection comprises around 10,000 drawings of letterforms, commercial correspondence, and material relating to the technology of typesetting non-Latin typefaces. The artefacts are unique and irreplaceable, and generally sensitive to repeated handling. Student work on the MA is split evenly between practical and academic work. The main academic output takes the form of a rigorous dissertation based on original research. The better examples are of publication quality, and contribute to the nascent scholarship in the field.

 

Objective, and a Challenge

We surveyed the field and identified a community of practitioners transitioning to teaching careers, and educators seeking to gain higher qualifications in a research-intensive environment. Unlike early-career designers, this community does not require practical skills building, but is characterised by a lack of engagement with the literature in the field, and a lack of understanding in specialist areas, most notably working with archival material, documenting artefact-based research, and integrating artefact-based research into practice. We have also identified a broader lack of academic writing skills. Seeking to capture this audience, we designed a new MA programme: we expanded the academic elements to occupy the full credit weighting, and strengthened particularly the research methods elements. However, our target community is international in location, and limited in mobility: professionals cannot interrupt their practice, and educators cannot easily take out a full twelve months. This represented significant challenges for three reasons: firstly, because the print literature in typeface design is not generally present in university libraries, even if these institutions run graphic design programmes. Secondly, because our methodology for building research skills is founded on intimate engagement with original artefacts. And, thirdly, because we place considerable expectations on group-based learning and peer engagement.

 

Programme Development

In response to the limitations to student mobility, and the three challenges we identified, we developed a hybrid mode of study. Our model combines a part-time, distance-learning mode for the majority of the 24-month registration, with three full-time residential periods of two weeks each. The aim is to combine self-directed learning through guided study, discourse development through engagement with an online community of peers, face-to-face feedback on presentations and discussion, and hands-on experience with sensitive artefacts. The programme follows a three term per academic year structure. Students will start the course in October of Year 1, with the first residential period towards the end of the the first term (late November – early December). The second residential period will take place in the summer of the first year, at a time that coincides with the vacation period of most HEIs. The third residential period will take place in the autumn of Year 2. Currently we intend  to recruit only one cohort every two years.

 

Online Presence

We audited the literature we intended  students to have access to, and identified only partial coverage by our institutional provision, especially for a worldwide cohort. We are addressing this by making available online ex-copyright material in an environment that allows shared use and annotation, and working with our institution to enable global access to copyrighted print resources, in electronic form. We will be employing collaborative tools for asynchronous seminars, and building a knowledge base around the core texts of the programme. For the second residential period we will run parallel student-led blogs on predetermined areas of study. All material will be shared amongst the whole cohort and staff, and final states of texts made available more publicly.