Words’ End

Oh, Friday! Friday was a big day for electronic literature and the digital humanities (see this earlier post). A great session to go to was “New Reading Interfaces,” presided over by Rita Raley who knows how to get a discussion going. Here are some cool projects and topics discussed in this session.

Jeremy Douglass talked about tag clouds as an aesthetic medium. They are web browsing interfaces, and despite their name they’re usually organized alphabetically or by popularity. Douglass takes the idea of cloud and runs with it, exploring them as a creative medium: the example he gave of this was the Flickr Fiesta 2005 invite he received by email. Algorithmically sophisticated literary renderings like TextArc, where terms have geographic meaning may look like tag clouds, but the latter are much simpler, Douglass said; plus, TextArc isn’t searchable, whereas tag clouds are. Later in the session, he brought home the broad(er) point that the tag cloud isn’t just a utilitarian interface; it can be portraiture, for example when some blogs replace their mastheads with tag clouds.

Then Joseph Tabbi talked about the semantic literary web, mostly in the context of the ELO Archive-It MediaWiki, a joint project with the Library of Congress. How do you preserve something, Tabbi asked? Well, you can tag it, which is limited but useful as a field-building (as opposed to literary) activity. OK, so what counts as a literary interface? Clouds are interesting as conceptual art, but their literariness (found through reading) is limited. Tabbi talked about Electronic Book Review (ebr) as an example of experiments in literary interfaces: the ebr website gets completely overhauled every couple of years, sometimes with sub-optimal for readability results. The key, for Tabbi, is to find conceptual connections while reading, and cross-link, cross-categorize – both to writing within and outside electronicbookreview.com.

Elizabeth Swanstorm talked about the interface in Jeffrey Shaw’s installation piece The Legible City. This is one I would travel overseas to play with, given more time and financial resources. Here’s how Shaw himself describes it:

In The Legible City the visitor is able to ride a stationary bicycle through a simulated representation of a city that is constituted by computer-generated three-dimensional letters that form words and sentences along the sides of the streets. Using the ground plans of actual cities – Manhattan, Amsterdam and Karlsruhe – the existing architecture of these cities is completely replaced by textual formations written and compiled by Dirk Groeneveld. Travelling through these cities of words is consequently a journey of reading; choosing the path one takes is a choice of texts as well as their spontaneous juxtapositions and conjunctions of meaning.

Better: the latest installation is multiplayer! If people are using more than one stationary bike, they may encounter each other’s avatars in the virtual world. So each rider is a node in a distributed networked system; their actions influence others’ virtual world; and their physical surroundings, irrelevant, fall away. So what kind of interactor, Swanstorm asked, does The Legible City produce – readers, riders, writers? Her eventual thesis was that this project highlights textual analysis as something one does by actively interacting with the text. No kidding; imagine giving undergraduate students of literary writing and/or criticism the visceral experience of this installation. They’d have a different relationship with literature forevermore.

Finally, Victoria Szabo talked about teaching, reading and creating scholarly works in 3D environments. Specifically, she talked about how they (Information Science + Information Studies at Duke) use Second Life in teaching. Students create objects, hold events and collaborate on criticism virtually… oh, just look at the ISIS site I just linked to. Szabo’s overarching point was this architectural metaphor: building and thinking are closely related. They put this into successful ongoing practice over at ISIS, encouraging students to combine creative and critical acts in their use of 3D virtual worlds.

Been home for 24 hours now, and I realize that I didn’t finish writing up the exciting stuff I saw at MLA on Thursday. So:

1. NINES, “a networked infrastructure for nineteenth-century electronic scholarship,” continues to impress with its impact and exemplary use of the net for collaboration. It arose, Laura Mandell said in her talk, in reaction to the prejudice against electronic publishing among tenure review, faculty search and other profession-influencing committees. The NINES editorial board not only aims to separate high-quality electronic scholarship from the chaff, but also do so in a sustainable manner. To that end, from what I understand they review sites and projects but leave things like copy-editing to authors themselves, ideally aided by their own institutions.

Laura’s point that the digital resources don’t, and can’t, disguise the human agency that creates them is worth repeating every once in a while. One of the ways in which electronic scholarship has been good for the humanities is that computation forces us to admit we’re constantly making choices, and some of these choices are arbitrary in that equally valid options exist for many editorial decisions. Objectivity as an aim falls away when you’re working computationally, and what’s left is a need to clearly explain your decisions. As we know from so many spheres of life, transparency is key communication. Scholarly communication is no exception from that.

2. In the same session, Robert Blake talked about the UC Language Consortium, which totally blew me away even if their site has been down for a few days now. They’re developing online resources for the teaching of foreign languages, starting with impressive projects in Filipino and Arabic. The consortium solicits proposals for development of these resources, and gives out small ($5,000-20,000) grants. The courses for which these resources are developed proceed to be open – for credit and all – to all students within the UC system, and the online materials are open to anyone to look at. Now that’s open courseware. And their next big project is Punjabi Without Walls! Apparently the Punjabi communities in the U.S. (and presumably elsewhere) are excited about this, since they want to keep their language alive and these materials will make that easier.

On to MLA Friday in the next post.

Highlights from Thursday:

The first session I went to was “The Challenge of a Million Books.” The title refers to computational mining of huge amounts of text at a time, in an attempt to discover bird’s-eye-view-level things we’d have trouble seeing with the naked eye. I discovered at this session that text mining is also called knowledge discovery. The latter is a term a bit too generic, I think: my encoded Roland excerpts also permit, even encourage, knowledge discovery, but what I’ve done with manual encoding and a simple interface is a far cry from sophisticated algorithms and machine learning.

Sara Steger presented on her research of sentimentality in nineteenth-century literature. This doctoral dissertation work is one of the test cases for the MONK project (Metadata Offer New Knowledge), one of the coolest collaborative endeavors currently out there. Simply put by the project creators themselves, MONK “is a digital environment designed to help humanities scholars discover and analyze patterns in the texts they study.” Sara took a bunch of mid-19th-century English texts, designated some chapters as sentimental (she brought up Little Nell’s death scene from Dickens’ The Old Curiosity Shop as an example), and other chapters as unsentimental. She used these as the training set for the MONK algorithm, “asking” it to figure out more or less on its own what makes a text sentimental or unsentimental, and then having new chapters automatically classified. Statistical analysis then revealed some interesting things: some words are clearly associated with sentimentality (having to do with the female gender, or children, or death, or love), while others are just the opposite (including titles such as Mr./Mrs., and business- and law-related words). Sara’s theory is that this means sentimentality is not just there when we “feel it.” It’s at least in part a formula, used by 19th-century writers to political ends. Her research is still in progress, but is already producing quite cool results.

The other cool URL I gathered from the session is SEASR (pronounced Caesar), Software Environment for the Advancement of Scholarly Research. This project works in tandem with MONK, and seems to aim for “construct[ing] data services that access and normalize unstructured information.” It looks as though the final product will be available not only to large projects but to individual scholars as well; exciting.

Later that evening John Unsworth spoke on “Cyberinfrastructure and Open Standards, Methods, and Communities.” As usual with Unsworth’s dense and whirlwind talks, I quickly gave up on taking notes, Luckily, the entire talk is online, albeit a bit difficult to read without margins. But copy-paste, print it out even, read it: this powerhouse of digital humanities always impresses with his ability to synthesize large, important topics in an accessible way.

Chicago-town has strange weather. I got in on Thursday to a dry, near-freezing city very similar to Boston – but yesterday there was a wet-snow storm that was gorgeous, diagonal and swirling, out the huge hotel windows but left almost no snow on the ground. That’s lake effect for you.

I’m here for the annual convention of the Modern Language Association. MLA is an odd beast. With about ten thousand attendees a year, I’m pretty sure it’s the largest humanities conference in North America. (I’d be curious to find out that I’m wrong! If you know of a larger one, tell me.) It’s of necessity impersonal, and filled with stressed-out people interviewing for jobs, sitting in one committee meeting after another, taking every advantage of being in the same town as far-away colleagues to cram in as much geeking-out about their favorite geeky topics as they can, losing sleep in the process.

OK, that last part is true of any academic conference. But still, MLA isn’t generally thought of as an exactly enjoyable event.

This year, though, the organizers seem to have gone all out in promoting digital humanities sessions. The poster/demo session I was in, “Electronic Literature: Reading, Writing, Navigating,” was mentioned in the Winter 2007 newsletter – a big deal, considering the thing goes out to 30K members. The result was a rush: the hour-and-fifteen-minute session was packed with people, and I didn’t get to see my colleagues’ work until the very end because pesky people were coming up and being all interested in RolandHT (poster, 1MB, and teaching modules, 31K, both PDF files) .

I loved every minute, of course. The whole thing left me flyin’, feeling much like I do at Digital Humanities conferences. This was both unusual in the context of MLA, and a welcome respite from the past few months’ job search both in and out of academe. So, if you’re reading this and were there: thank you! If you have any further thoughts on the project, please comment here or email me, username vika at this domain.

I’ll post a few session notes later on. For now, breakfast.

This coming Tuesday, September 18th, come to MIT for the first in the Purple Blurb digital reading series. “The readings will start at 6pm at MIT in 14N–233 (second floor of building 14, in the wing that is across the courtyard from the Hayden Library),” says organizer Nick Montfort in the announcement.

The first reader will be Robert Kendall, and I’m very sorry to miss it due to a prior obligation: Rob’s words tend to transport me somewhere familiar I’ve never been before. At the next event on October 18th, I’ll be reading from Roland^HT and talking a bit about narrative threads running through it. The other two readings this semester will take place on November 13th (Barbara Barry) and December 4th (Andrew Plotkin).

For a good time, call on Purple Blurb.

The Maryland Institute for Technology in the Humanities, which hosts a series of informal Digital Dialogues, has put up podcasts of the talks that have already taken place this semester. One of them’s my November talk about the Virtual Humanities Lab.

Whew, that was grand. Just one thing about the closing panel discussion, while it’s fresh in my mind.

This year’s CaSTA was billed as “a joint computer science and humanities computing conference.” And it was! And [we saw that] it was good. Of the five keynote speakers, three were humanists and two – computer scientists. The final discussion was called “Humanities Computing Science??”.

William Arms, in his remarks during the panel, said that during the conference a word was frequently used that isn’t generally used in his usual [computer-science] circles. That word – knowledge. He, and just about everyone at the panel, said that what they primarily want from the “other side” is dialogue.

In light of that, what I’d like to see in this continuing dialogue is a bit of discussion of the word science. As it’s been used lately (in the last, what, 200 years?), it implies “HARD.” Humanities implies “soft.” That’s a major point of contention.

But given that “science” pretty much means “knowledge,” should we revisit our use of the word?

Ray Siemens is a computing humanist and Renaissance scholar working at the University of Victoria. The full title of his paper is “Knowledge management and textual cultures? Work toward the Renaissance English Knowledgebase (REKn, pron. “reckon”) and its professional reading environment (PReE).”

REKn seems to be aiming to amalgamate and integrate knowledge in its area. Its implementation is based in the study of disciplinary activity of, and professional interaction among, those in the humanities. It’s founded in concepts of knowledge representation and modeling. A short description of the project can be found here.

Knowledge representation: draws on the field of AI and seeks to produce models of human understanding that are tractable to computation. Modeling: REKn/PReE model data, intellectual processes, and beyond.

Key elements of REKn’s model:

- representation of archival materials

- analysis/critical inquiry originating in those materials

- the communication of the results of these tasks (the dissemination of primary and secondary materials)

REKn’s assumptions: all of the above are interrelated and inseparable, and electronically representable.

They’ve collected primary and secondary sources, and have built tools for working with them (the tool-building process seems to have been multi-stage: many tools built and discarded as inadequate). They’re looking to long-term partnerships with Renaissance materials providers in the future. Right now REKn has about 13,000 primary sources and over 80,000 secondary sources. About 1500 of these resources are currently available for public use, but the majority are not open-access.

So that’s REKn, the text base. What about PReE, the reading environment? It’s a rudimentary document viewer, and analysis and communication facilitator. Currently the UI is a “down-and-dirty prototype,” as they’ve been concentrating on making things work in the back end. [vz: he's showing PReE in Windows; I wonder what it's written in.] They’ve made several analytical tools for the encoded texts. Primarily, though, analysis will be carried out using TAPoR tools.

Communication facilitated electronically: they’re attempting to provide a system by which people can manage their professional interaction.

Short-term goals:

- integrate better with TAPoR and the Public Knowledge Project reading tools

- conduct usability studies

- consult with “contextual” stakeholders, including acad. publishers

- move prototype to a web environment

- scale up!

And that’s Ray’s talk, the last talk of this conference. Next is the panel discussion, titled “Humanities computing science?”. The panel will consist of the five keynote speakers. I’m not sure whether I’ll be taking notes on this; it’ll be video recorded, and there’s little probability that I’d do it justice. Again, I’ll update when the webcasts are up.

Eugene Lyman is an Early English scholar at Boston University. The full title of his paper is “Presenting an electronic critical edition of Piers Plowman B.” The B refers to one set in a multitude of manuscripts of this Middle English alliteraative poem, comprising only one of its versions.

EL frames his talk with the following two quotes:

“Computers compute, of course, but computers today, vfrom most users’ points of view, are not so much engines of computation as venues for representation.” –Matthew Kirschenbaum

Understanding the poetics and principles of electronic scholarly editing means understanding that the primary goal of this activity is not to dictate what can be seen but rather to open up ways of seeing.” –Martha Nell Smith

Lyman has created software that allows you to look at the existing manuscript pages from ten different manuscripts, enlarge portions of those pages, read the transcriptions, look at erasures that tell us interesting things about how people might’ve edited in medieval England. You can search within the dataset, visualize the text in various ways, view the underlying XML markup… yesterday (or was it the day before?) EL actually gave some of us an informal demo of this thing, and it is sweet.

How to make a continuity of presenting a single text that exists in multiple manuscripts?

EL created the Elwood Viewer, which looks at documentary editions of texts. Its aims:

- tight coordination of text and image

- visual cueing to guide/reinforce reader’s attention

- handy tools, all within a metaphorical arm’s reach

- ease of navigation, especially at opportune moments

- parsimonious use of screen real estate

- simple, no-cost programming environment, open to change [this is all JavaScript, I think... -vz]

The software allows you to extract data for further analysis. One such analysis called into question the notion that scribes were totally random with respect to the ways in which they encoded [marked up, oh yes, for markup exists in many forms including punctuation and embellished first letters] their texts.

This is all a cricital edition: you take a group of witnesses, compare them, note the variations, and choose the variations you think were on the author’s agenda when she was writing the text. The notion of a critical edition, which effectively “leaves behind” the actual manuscripts, is pretty controversial; nevertheless, at the moment critical editions have a lot of weight in literary studies. [vz: I'm leaning toward the camp that's sceptical of critical editions, but only in the sense that I don't tend to consider them definitive, while many others do.]

EL has a prototype version of the critical edition. I haven’t found any screenshots of it on the web, but here’s the Piers Plowman Electronic Archive hosted at the University of Virginia. It seems more a worksite than a ready-made resource, but is worth poking around.

Johanna Drucker is a professor of media studies, and a founding member of UVA’s Speculative Computing Lab. The full title of her keynote is “Graphic conventions: visualizing knowledge and subjectivity.”

Can we shift from information to interpretation, and shift [back] to a more humanistic point of view within digital humanities? To do that, we have to re-introduce subjectivity, and maybe substitute the mechanistic with the probabilistic (the latter being humanists’ worldview, according to JD).

Visual information conveys information in a form that makes it very hard to analyze systematically. Two ways to create stable knowledge in a notation system: one is with [natural] language, the other – mathematical notation, said Rene Toms of the Oulipo. He never did talk about visual representation, and with good reason: it’s an unstable notation mode.

Subjectivity comes in two forms: position (structural) or inflection (semantic).

The notion of information comes from a particular set of assumptions of what knowledge is. JD not interested of getting rid of this model of knowledge; but rather to propose another model of knowledge.

Visualization is compact, the problem isn’t having enough space to represent information – it’s pinning down the exact nature of the information and our assumptions about its aspects.

Visual information (VI) can lie or misinform, like natural language can.

The silliness of chunking of processes (how authors write stories: “author thinks about a topic” –> “author sketches an outline” –> “author reviews the sketch”) is apparent, but we have to do this sort of chunking when we’re working in a computational environment, which requires discrete units. Because of schematics’ rhetorical power, we eventually come to believe them.

So what about text visualization (as opposed to data visualization above)? Interest in doing things to and with texts is very active, especially within the creative-writing communities. Like TextArc, that sort of thing. They can be silly, ugly, destructive of the original, and yet they have their uses.

Edward Tufte, the exquisite engineer according to JD: information pre-exists visualization. Visualizations can be transparent enough to get us access to information. JD disagrees: visualizations are interpretive, opaque, distortive. They create informmation.

Temporal modeling at SpecLab. Basic assumption: timelines as they are conventionally defined and designed come out of the empirical/natural sciences. Assumptions there: time is unilinear; time is homogeneous (metric is stable); time is continuous (no unbroken intervals in temporality). None of these three things hold. Temporality branches in our lives, in poetry/film/etc. Time is not homogeneous – some moments fly by, others are long (the moment before the kiss and the moment after are very different, JD says). Time is not continous, either: there are breaks/ruptures, recorded in historical accounts for example.

SpecLab constructed a grammar of inflections, of visual elements they’d use to represent time, types of events and their relations to each other. There’s a lot of information JD is giving about what SpecLab has been doing; I’ll point you to the Lab’s site instead of summarizing.

In the IVANHOE game, every action takes place from within a role. Each role has a set of assumptions that go with it. They employed it as a teaching tool at UVA, with the purpose of showing that, in fact, every action stems from a set of presuppositions. [vz: that can't be right, I've captured too simplistic a description. Go see the site for more.]

Subjective meteorology: JD’s current project. An art project, which JD says – duh, art is in the humanities. [vz: yay!] She charted and graphed and visually represented a bunch of weather patterns – lines of anxiety/anticipation, storms of anger – which look gorgeous on the slides but don’t seem to be on the web. These representations can be chained together and animated to playfully and visually represent one’s subjective perceptions of the world around.

Great discussion follows. I can’t pretend to capture it well enough; I’ll post an update when the keynote webcasts are up, and urge anyone interested to watch this one when it’s available.