Today we were gifted with the enormously vast pleasure of having Scalar‘s Information Design Director Craig Dietrich join us virtually in Duke’s PhD Lab in Digital Knowledge for a consult and conversation.

Florian and I were joined today in the lab by our colleague Dwayne Dixon, whose digital Scalar dissertation was recently featured on HASTAC.

The three of us were greedy: we asked questions about Scalar’s past and future; we asked about personalized design possibilities and about specifically-tailored functionality; we asked about assurances as to Scalar’s longevity and about our options for sustaining our digital publication; and we asked for tips about how we might introduce Scalar as a digital dissertation platform to Duke U deans and administrators.

Our greed was well-rewarded. Craig enthusiastically addressed all of our questions in stride and offered not only comprehensive but surprisingly satisfying answers. It is indeed rare to find a digital publication platform that fills both needs, wants, and pie-in-the-sky desires. Our faith in – and enthusiasm for – the Scalar publication platform was undoubtedly confirmed today.

In addressing Scalar’s future, Craig responded by way of covering various angles:

1) Craig said the recent media server crash “opened their eyes” and they’ve made quick progress to not only remedy the issue but also to prevent its re-occurrence;

2) Scalar is actively working to create new features, designs, and functions that promise to be thrilling for publishers, authors, archivists and artists (we were treated to an example of this and are excited for its forthcoming release);

3) The Scalar folks want, and indeed insist, that Scalar live on for decades. Craig believes that if Scalar did somehow fail or fold or was in any way ‘taken down’, it would not just be a loss for Scalar authors and publishers but would also be a “catastrophic” blow to digital scholarship writ large (I thought this a profound and important statement);

4) Scalar evolves with automatic updates and innovative new features;

5) Scalar actively supports projects that solve, or at least speak to, the issues that drove the need for Scalar to be created.

 

In addressing issues about digital scholarship, digital dissertations, and publication persistence, Craig reminded us the following:

1) the Scalar software can be downloaded and hosted on a user’s own server. In this way, it can be a persistent environment. What if universities like Duke downloaded Scalar, tweaked it so as to suit a standard set of required ‘digital dissertation’ specifications (mimicking current dissertation submission regulations but acknowledging the affordances and the limitations of the digital format and thus satisfying the need for dissertation submission consistency), and hosted student-designed dissertation publications?

2) Using a Scalar API, with one call, an author can download the entire archive of his/her Scalar book in an RDF (Resource Description Framework) format.

3) Each Scalar content element is a page itself and has a URL. Each content piece then – from text and media element to annotation and public comment – can have its own unique identifier thus making archiving an interesting but not an impossible operation.

When we asked about the community of practitioners, Craig pointed us to  several impressive projects and people and I wanted to share those here as points of inspiration that we’ll be following.

Media Ecology Project Mark Williams & John Bell

Hypothes.is

Jentery Sayers

Duke Press’s collaboration with Scalar

We were honored by Craig’s generosity and expertise and we look forward to many (and many more) future conversations and collaborations.

 

asg

After clarifying our ideas and goals for the publication on Monday, and thinking about the role of keywords in the publication and explaining the reasoning behind Scalar as platform of choice on Tuesday it is time to think about the design of the Network[ed] Ecology publication.

Design of a digital publication is not only about the visual look and feel but also about the interaction with the content. To not invent the wheel anew we asked

  • What existing infrastructure and readymade interfaces can be reused to create a digital publication easily?
  • What features and functions are missing and needs to be created, that allows easy scholarly publishing?

Scalar as a platform already takes away a lot of the decisions on the side of the backend of the project and offers a standard reading interface, where the author can decide on a default view for the page, which are distinguished in an emphasis on text or on media, the positioning of the media content affiliated to a page as well as a range of data visualizations, that visualize the content / network structure as well as tags or metadata. The default view can in the “reading” process be changed by the reader. This standard interface can be adapted like most website-theme by for example setting custom background images and sound, custom CSS styles, custom JavaScript or JQuery functions to enhance the behavior. But besides the default interface provided by Scalar the system offers the possibility to use the data edited and structured in its backend in a custom interface by accessing the data through the Scalar API.

We used two starting points to think about the design.

  1. design research on electronic publications and digital storytelling – both produced employing Scalar and other backend technologies
  2. our data corpus and its specific needs for mediation.

1) Design Research

Already before the meeting we started to look into diverse practices of digital publication and digital storytelling, and focused on practices which would transform “reading” into exploration of a content network. In a second step we looked at the showcase on the Scalar website, in order to see best practices and get a feel of how other scholars employed Scalar for their publication. We will leave the analysis and a deeper dive into the mechanics, aesthetics and interface grammar of digital publications for a later post. But at least some projects, we discussed today should be mentioned here.

Visually we were absolutely struck by the narrative structure and aesthetics of projects like the Highrise Project, which is an audiovisual interactive narrative with various vantage-points to dive deeper into more detailed information layers branching off the main narrative. Where the narrative structure is interesting we also realized that this particular format would not fit our data. A much better role model for our project is Performing Archive: Edward S. Curtis + ‘the vanishing race. It acts as a “meta-archive” and interpretive layer to archive material through essays from several scholars (see project description on Scalar showcase) and does not only make use of bold visuals, good readability and making the network character of the project explicit through using a visualization of the content network but also has an open ended character through being open for future contributions.

 

2) Starting from the data corpus

Each data corpus has its own needs in terms of how to make it explorable and how to mediate it. There is no “generic” design if you are aiming at a sophisticatedly edited publication which needs to bring different streams of content together – as our publication does. Decisions need to be taken on how the user can interact with the content; how to structure the different data streams to make them explorable, to form meaningful connections and slipping points while leaving enough “gaps” to be explored.

 

Design decisions taken:

We decided to re-use a experimental new reader interface for Scalar, that is also employed in the project “Performing Archive: Edward S. Curtis + ‘the vanishing race” and adapt it to our needs. It already offers a modern and clean design, bold use of images, a clear and straightforward navigation, a better readability than the standard interface and a navigation feature called “pin-wheel”, which uses a visualization of the content-network as for traversing the data-network of the Scalar-publication. Tomorrow we will clarify with Scalar, how to work with this new interface technically.

We decided on three entry points into the data:

  1. topical paths, which make data around a specific topic accessible
  2. 2 sections: essays and symposium
  3. Keywords and annotations as “slipping points” for the transition between different content layers.

Moreover we decided to use the Twitter stream of the symposium as annotation and additional layer over the video material captured at the Network_Ecologies Symposium. That way they act as enhancement and contextualization of the audiovisual material and on the other hand offer vantage points to other layers of content.

 

ToDo for tomorrow, Thursday:

We will Skype with Scalar representative Craig Dietrich to discuss several questions with regard to the production of this particular publication but also about the future of Scalar as a publishing platform.

Why Scalar? Well, here’s why…

1. Because it gives the reader multiple reading interfaces: the reader is permitted to traverse the network of content as s/he chooses

2. Because of its vast annotation possibilities -, annotations can provide slipping points

  • Our annotations will work both as traditional annotations – as footnotes, comments from fellow scholars, as captions, as subtitles – AND as networked and networking paths linking our pieces of content.
  • Also allow commenting. Might we consider continuing our editorial grab by graduating some of the public comments into the document? Or do we want to publish the project as-is and give it to the public as a finished-beginning. The living-content.
  • Idea: Present an incentive by saying that comments for the first year will be considered for in-text publication. After the first year, we end our editorial control and the piece lives on (hopefully) through the public commenting)

3. Because it is a network of data with architectures built atop and within that permit one to ‘read’ the piece as a network(ed) publication

4. Because we can author passages wherein a reader can intentionally slip between layers of content.

5. Because we can author passages wherein a reader can unintentionally slip between layers of content – this lends an element of emergent discovery. Through unintentional slips (that the reader can always escape if s/he chooses) the reader can fall into an unexpected path

6. Because we can do both 4 AND 5 AND at the same time also author a prescribed reading path that acts as do the more traditional publication navigation markers and systems

7. Because pages are paths and paths are pages. Pages can have specific roles – paths, comment, annotations, tag, etc

8. Because it permits public commenting

asg

Amanda’s Tuesday Morning Keyword Manifesto:

1. This is an authored and edited collection. The project began from a set of keywords – network(s), networked, ecology, ecologies – and sought interdisciplinary understandings of each. These keywords are at the center of the project- they were indeed the very guiding ethos of the project – and I believe they must remain there.

2. To these originary keywords, we’ll add others that emerged throughout the project. These are:

  • time: temporality, dates, times: future, past, history, synchronized, simultaneous, repetition, continually returns, reiteration, immediate, persists, current etc
  • digital humanities
  • architecture/infrastructure
  • data, metadata
  • database
  • information
  • connectedness/connectivism, relation(s)/relationships (associations, attachement/detachment), intersect(ion), interaction(s)
  • protocol(s), policies
  • communicate/communication
  • emergent, evolving
  • code/encode
  • programming
  • uncertainty, ambiguity, abstract
  • platform
  • nonlinear
  • interdisciplinary: across disciplines, spanning disciplines, etc
  • archive
  • [digital] scholarship
  • co-creative, co-generative, co-emergent
  • culture/cultural
  • art
  • body/bodies
  • sense(s)/sensation(s)
  • scale
  • medium/media: letters, text, videogames, painting, books, notes, etc
  • experience
  • memory (computer and human)

3. We will then put these into dialogue with computer-generated keywords that reveal themselves by way of computer-driven text analysis. Because our publication is a born-digital publication, we want to acknowledge the computer’s perspective vis-a-vis keywords but we do not want to privilege these keywords over and above those created in & for steps one and two.

We will edit these keywords then and curate navigation paths, slipping points, access entrances into and through the content of the digital Scalar publication.

 

asg

What we did today, Monday April 28, 2014

We addressed our Main Goals as were yesterday outlined for today’s session:

1. Write short concept paragraph describing the publication purpose and design

**Our short paragraph turned into a rather long and detailed document, which is here in its evolving entirety.

2. Begin by doing text analysis of data and do some sample keywording on one specific case so as to think about how we might use the digitality of the medium to digitally analyze our data. We want to integrate digitality at all stages.

1. We’ve published and opened our collaboratively-authored Network[ed] Ecologies Digital Publication Concept Document to allow for further public collaboration. We invite you to view our document and to join our project by adding comments, questions, and critiques.

Excerpts:

When thinking about how we might manifest our Network[ed] Ecologies publication, we thought about the project on multiple (networked) levels: the idea was not only to create a digital publication on the topic of networks but at the same time to rethink the possibilities and potentials of digital and digitally-networked publication.

…a digital publication about networks, which lives in a networked environment, can only be properly presented as a network in itself. Thus this digitally-born publication does not only reside in the network environment of the internet, where any context is only a click on a hyperlink or a search away, but has to be thought of as a network of data in itself – as a networkED publication.

We want Network[ed] Ecologies – A Living Publication to be an open publication and a living conversation. This digital publication, that will use the Scalar publication platform, will house the born-digital content created during Stage One of the Ecology of Networks project – an online exchange that merged original content from an interdisciplinary group of invited professionals with born-digital media content to form an online responsive dialogue – and Stage Two – a live-blogged, live-tweeted, and video-captured in-person symposium at Duke University that featured both our online scholars and new invitees.

Network[ed] Ecologies – A Living Publication will use the affordances of the digital to remediate all of our content – from video and text to tweets and images – into a living but persistent and carefully curated publication, and will  present a multimodal, multiply-interactive, poly-interfaced ‘reading’ experience.

Many parties from author to reader to publisher  have an interest and a stake in the challenge to create an open publication that is still professionally composed, editorially rigorous,  thoughtfully – and provocatively – authored and dynamically maintained. This is our challenge.

2. For our second goal, we scraped the text from our initial essays and conversations and processed those through several open-source digital text analysis tools. Stay tuned for more details about the computer-generated keywords these tests revealed.

Goals for tomorrow, Tuesday April 29, 2014

1. Discuss content & keywords

Do keywords represent an index? a chapter organization?
How fine-grained do we want our annotations to be?
Do we want to do annotations ourselves?
How can we automate annotation? (do we want to?)
Can we open annotation to the readers?
Role of annotations: provide access to content as navigation points
What steps do we need to go through to make the navigation points available?

  • Keywords as layers:

1) editor’s keywords that motivate the project at the beginning stages
2) keywords that emerge as content comes in (inductive annotation/coding method of text and video)
3) keywords that are computer-generated (automated? text-analysis) – computer generated analysis should be in dialog with but not replacements for the human-generated

2. Continue discussing how we will accomplish the goals outlined in our Network[ed] Ecologies Digital Publication Concept Document. This is, of course an ongoing evolving discussion.

 

asg

 

What we did today, Sunday April 27, 2014:

Florian Wiencek and I (Amanda Starling Gould) started our PhD Lab Network_Ecologies Scalar design project ‘residency’ today by inhabiting the lab and adopting it as our home base for this week’s open working meetings. This is stage three of the Network Ecologies project whereby we curate our born-digital and our in-person conference content into a digital publication using the Scalar publication platform.

Our discussions began today by talking about data, data relations, keywords, and content, and by brainstorming an architecture and infrastructure for their digital presentation.

Notes from our meeting:

Our First steps: Begin with the data

  1. We’ll investigate our data and how we’d like to make it explorable  (what layers of data do we have as basis for the publication and how should the ‘reader’ to be able to interact with the data?)
  2. We’ll investigate the relationships: relationships between the text, relationships we want to bring out, relationships we want to create
  3. We’ll brainstorm how we can use digital tools to aid our investigations and to create new data for our digital publication

Next Steps: How do we curate, present, (ab)use, open, provoke, and interact with that data?

  1. We’ll brainstorm possible architectures for creating a network from these
  2. We’ll think about how we can curate further by adding annotations (more data)
  3. We’ll think about how we mobilize the affordances of digital publication platforms – and Scalar in particular – to  integrate interaction into the ‘reading’ experience
  4. We’ll map out our possibilities for  various reading paths?

We chose Scalar precisely because it encourages various ‘reading’ paths. It allows for a reader to intentionally or unintentionally slip into another content path. We plan to build the following navigation paths:

-keywords as navigation element, as what I call a ‘slipping point’
-media – including text AND annotations as slipping points
-A & F devise a more traditional – a more “authored” reading path as scalar allows

Questions to keep in mind as we work:

  • What would be desirable as a digital reading/interaction interface for a networked publication?
  • How can we mobilize and/or co-opt digital (analysis-) tools in the production of our digital publication?
  • What is the role of the editor in this type of project? We are doing fundamentally different work here as digital publication editors. We are authoring data – and data layers (e.g. metadata), relations, infrastructure, architecture, interaction, annotation… How does this define our role?
  • Digital curation as filtering – what are our filters? What role do our filters play? And what is at stake here vis-a-vis digital publication?
  • Our data is digital and we want to remember that at every stage. How can we best do this? How does thinking digitally alter/enhance the digital publication? How does thinking digitally challenge the idea of publication writ large?

 

Goals for tomorrow, Monday April 28, 2014:

  1. Write short concept paragraph describing the publication purpose and design
  2. Begin by doing text analysis of data and do some sample keywording on one specific case
    1. We want to think about how we might use the digitality of the medium to digitally analyze our data. We want to integrate digitality at all stages.
    2. If we want to create a keyword navigation structure, we can think about keywords from the human perspective and can use digital text analysis methods so as to see the keywords a computer-perspective might generate.
  3. Think about what it means to think about human-created keywords vs computer-generated keywords
  4. Compare our human-perspective text analysis methods and those deduced by the computer-perspective

 

 

Follow us here for daily updates & at #netcologies on Twitter

asg

Stay tuned for daily posts as Florian and Amanda begin prototyping our stage-three digital digital Scalar publication. The goal of our publication is to capture the multimodal content – from video to blogs to images to tweets – from the first two stages of our project and to curate that into a living publication.

Starting Monday we will begin our week-long design sprint and will document our process along the way…