engaging experience

[WUD] World Usability Day: Seventh Session

World Usability Day 2006
Landmark College, Putney Vermont

Conference notes by Meg Houston Maker

UDI Lab: Measures of Satisfaction

Steve Fadden, Sarah Horton, moderators

This session comprises heuristic evaluations by students and experts of existing sites. Some attendees indicate their interest in methods for evaluating the qualitative aspects of a website.

There are tools that let you evaluate the usability of sites, e.g., Techsmith.com's UserVue or Bobby. The University of Illinois offers tools within their accessibility support services area that can be used by others.

At Landmark, the courseware system is WebCT, which is not an accessible tool, which makes it challenging.

Sarah Horton asks, why are there so few programs in interface design? Taxis answers that there are a lack of researchers (Ph.D.s) to teach HCI. HCI has always been on the periphery of academic computer scientists.

Sarah says it's hard to think about design checkers, because it's a methodology, it's not something that you can actually check for, fully. The process of using a checker can be educational as long as you extrapolate up from the errors you are given. But that's somewhat like learning to do web design by looking at websites that suck.

We go around the room introducing ourselves. Some issues emerge:

- Some of us must work with vendor systems that have low usability, and we emphasize the importance of client communities sharing feedback from usability studies.

- Content must be reworked for the web environment. This is an issue for pedagogy, in which those running media services for student courseware must rework professors' content to make it usable and accessible for students. This requires the courseware developer to confirm that the material has been faithfully translated prior to posting. The web is really a different medium for writing.

- Usability labs that are video-based must be transcribed in order for the information to be accessible for research. A necessary burden.

- Usability labs also place extra cognitive burden on the subject because you're asking them to verbalize while they're also performing the task. In this case, eyetracking studies and video of usability studies can remove some of that user burden.

- The strong AI claim says that we can emulate human cognition to the extent that we could use algorithms to model human response to products. Being an expert in an area means you've seen more examples of any given phenomenon. This is possible with computing algorithms, too.

[WUD] World Usability Day: Sixth Session

World Usability Day 2006
Landmark College, Putney Vermont

Conference notes by Meg Houston Maker

Can a Picture be Worth 1,000 Words? Using Eye Movements and Video to Understand the User Experience

Steve Fadden, Landmark College

A user is anyone who interacts with a service, product, or system, for any purpose, in any environment. Usability is "the extent to which a product can be used by specified users to achieve specified goals with effectiveness, efficiency, and satisfaction in a specified context of use." So you have to specify the requirements! Usability is about effectiveness, efficiency, and satisfaction. Effectiveness and efficiency are quantitatively measurable, whereas satisfaction requires more qualitative reports.

But we're missing some facets of overall user experience: emotional components, impression of the company (brand experience), durability of the impression, overall value. Affect-aware systems are being designed to be sensitive to a users emotional responses to the system. (See, e.g. Battocchi, Pianse, and Gorn-Bar (2005) in ACM proceedings, and Sindells, MacLean, Booth, and Meitne (2006) in the Canadian Human-Computer Communications Society.) Researchers are developing a database of kinetic facial expressions called DaFEx. This information is being built into software systems and and work to train them to respond to a user's facial expression and present a different version of the UI.

Facial expressions are transitory, and the user's emotional response changes through time as they use a product and make discoveries.

Steve shows video of eyetracking scan paths over a webpage as a user's saccades (movement) and fixations are diagrammed through time as the user attempts to solve a task. The scan paths demonstrate different information seeking behavior, from jumpy and random to a laborious linear process of walking the page searching for the answer.

[WUD] World Usability Day: Fourth Session

World Usability Day 2006
Landmark College, Putney Vermont

Conference notes by Meg Houston Maker

How Google Works (And Why You Should Care)

Takis Metaxas, Wellesley College

...And not just Google, but all search engines—MSN, Yahoo!, etc.

We already have the web hooked to cell phones, PDAs, even toasters. But it's actually really hard to make a program like Dragon learn to recognize human speech better—that's a really hard problem. The web is over 10 billion static public pages now. This doesn't include the deep web of login-based pages that can't be indexed; this makes the web about three times larger. You can't even think about your life without the web. If you include dynamically generated pages, the web is, effectively infinite.

Email has spam, the web has spam, too—millions of pages offering information ungrounded in truth. These are pages modified to influence search engine results.

The web is a directed graph of notes and arcs. Each page is a nodes are pages, and links between them are arcs. Since it's a graph, it can be explored and indexed by algorithmic methods. These engines created an inverted index of the docs it has crawled, mapping words to the docs they appear in.

Search Engine History:

• First generation - 1994 (AltaVista, Excite, Infoseek): ranking based on content (pure information retrieval). Content similarity ranking treats docs with more rare words as more closely related. In this paradigm, docs are treated as simple "bags of words." Similarity is measured by vector angles, and quiery results are ranked by sorting the angles between query and documents. It's easy to figure out this algorithm so spammers can exploit this through keyword stuffing, usually in the META tags.

• Second generation - 1996 (Lycos): ranking based on content and structure. In this phase, search engines added popularity rankings, in which a hyperlink from a page in sitie A to some page in site B is considered a popularity vote. Docs are then ranked including popularity. Spammers then created link farms that have sites that are heavily connected to one another to increase rankings.

• Third generation - 1998 (Google Teoma, Yahoo): ranking based on content, structure, and value or reputation. Google introduced a reputation pagerank algorithm, in which the page rank of P(i) is the sum of a fraction of the reputations of all pages P(j) that point to P(i). This idea is similar to co-citations in academia. There's some beautiful math behind this. It can be spammed through "mutual admiration societies" in which a site may have high reputation in some field points to another site in a different field. If everyone does this, it means everyone's points go up. This is the premise of contemporary SEO strategies.

• In the works today: ranking based on "the need behind the query;" which relies on the system knowing something about the user and his or her needs.
  

Spammers early figured out how to game the system, which forced the search engines to evolve.

Societal Trust is also a weighted directed graph of notes and weighted arcs. The nodes are societal entities (people and ideas), the arcs are trusted recommendations from one entity to another, and the arc weight is the degree of trust. Lessons from propaganda theory influence this. Card stacking, in which you take a word unrelated to the endpoint and link it to something unrelated, you can make the search engines turn up the endpoint in searches for that query. You actually only need 30 of these; this is how 27 instances of "miserable failure" linked to George Bush can turn up in search engines.

We need to use our critical thinking skills, and to create trust engines and cyber-social structures that mimic societal structures, to know whey to trust or distrust information sources. Personalization of search results will help this, and use of back-propagation mechanisms to train the network who is trusted. In this way you can create your own "neighborhoods" of trusted entities.

[WUD] World Usability Day: Third Session, Student Panel

World Usability Day 2006
Landmark College, Putney Vermont

Conference notes by Meg Houston Maker

Student Panel: Discovering Usability Issues in our Classwork

Students: Isabel Arathoon, Nick Braley, Landon Gottlieb, Shaw Bates All from Landmark College

Landon opens describing a class project in which he produced a paper on the song "Los Angeles is Burning" and used the usability lab to document his process of using PowerPoint to create the final presentation. Landon has AD/HD, and is not a fan of PowerPoint because the application itself is distracting and makes the process of creating the presentation "so much harder.... PowerPoint should just be thrown away!" he says. Landon played a tape from the usability lab showing his process of struggling to create the presentation PowerPoint file. The main pane of the film is the computer screen, with PowerPoint loaded. There is also an inset of Landon's face, as from a camera mounted on top of the display. The film is riveting. It so clearly demonstrates how impossible the product is to use in the creative process. Landon blames his AD/HD for his difficulties, and the "jumping around" effect of using the tools, but it seems clear this is mostly about the product's usability.

Shaw did a similar project, and shows his tape from the lab. Shaw's tape shows his struggle to work with and within the PowerPoint tools. At one point he gets lost in the app by hitting a button by mistake. Toward the end of the tape it's clear his presentation is coming along. "This is how the learning process works for me," he says, "you start with not knowing much, and then learn a lot."

Nick Braley discuses his use of Dragon and Kurzweil. Dragon allows him to talk his papers and have them be typed for him. He went from straight Fs to As and Bs. Kurzweil, a text reader, reads his research materials to him, removing the process steps from his learning, and allowing him to focus on the material to be learned. These two tools have given him, he says, a new way of expressing himself.

Isabel uses Kurzweil and Inspiration and couldn't live without them. The act of reading was, previously, an act of decoding. Kurzweil lets her focus on extracting meaning and information, rather than simply decoding. For writing, she uses Inspiration to create concept maps to organize her ideas and ensure that her paper topics are well organized. These dimensional topic maps then convert into outlines that she can use for writing. Additionally, she uses color coding because to reinforce her auditory and visual learning approaches.

Questions and Answers

Q: For Shaw and Landon: What did you learn from watching the process of creating your presentations, by looking at the usability lab tape? A: Shaw: I learned when to ask for help, ask the professors for help. Landon: I learned I need to slow down and use the tool's capabilities.

Q: How do you use Kurzweil to take notes?
A: Isabel: While it's reading, you can pause the reading and open a toolbar to attach a note to a section or words.

Q: If you had a mic that let you talk to the developers of the software that you love or hate, what would you say? What one thing?
A: Nick: in Dragon, the training process takes a long time, so he couldn't use it in the first semester. So he would change the training process and accuracy, because his version is about 95% accurate.
Isabel: She tried to train Dragon, but her first language is Spanish and she has an accent, so she was not able to use Dragon.
Landon: He uses Dragon a lot, and agrees with both of the above comments.
Shaw: If someone were to invent a new technology, it would be something that hooks up to your head so that when you think, it types for you!

Q: Have you used the templates in Inspiration?
A: Isabel: sometimes, but they're not always applicable.

[WUD] World Usability Day: Second Session

World Usability Day 2006
Landmark College, Putney Vermont

Conference notes by Meg Houston Maker

Concept Mapping

Michelle Harper-Sciarini, University of Central Florida

Concept Mapping is a technique used to elicit knowledge of a domain. It's a visual representation of the relationship between key concepts—a structural relational depiction. It is context-dependent, usually representing different aspects of a larger domain.

Concept mapping provides a medium through which a subject can depict their knowledge structurally. It can be used to tap into a subject's full understanding of a topic outside traditional assessment or testing methods. It can thus be used for individuals who, e.g., don't test well, but do understand the material. You can use concept mapping to draw out this information.

This technique is based on the "meaningful learning" theory, which states that when we learn something new, we attach it to what we already know. This can be used for product design, instructional design, or assessment, eliciting mental models of how a system should be, look, and be used. Students can use concept mapping in their learning process to actively link concepts while they're learning and seat them in their mind.

Components of a concept map: Concepts, links, and labels. Concepts are the declarative knowledge about an object, situation, event, place, person, etc., represented by a word or phrase. For example, key concepts that explain Santa Claus include beard, Christmas, children, toys, chimney, reindeer, etc. In a lot of cases, concepts are printed on cards and presented to the subject, who will start making associations or links between items, creating a map. These linkages are then labeled to indicate the relationship between concepts.

Concepts, links, and labels form what's called a proposition, which defines an aspect of the domain you're trying to depict. A facilitator can be present during the process of creating this map to bring out the thought process of the person creating the map.

Often, the user or research subject should be the one to generate the concepts, but sometimes, when, e.g., you want to do quantitative analysis, it's useful to present existing concepts for subjects to push around.

Some assert concept maps should be hierarchical, but in a lot of systems, knowledge does not fit neatly into a hierarchical structure, or the subject doesn't think of the information as structured in a hierarchy, forcing them to use a hierarchy might mean you lose some of their input.

To create a good concept map, first identify a user who is extremely familiar with the domain. Then create a context in which the content map should be made by asking a focus question to get the user thinking about the topic, e.g., if you're making a concept map of a holiday, ask the user about the history of the holiday. Then, identify key concepts about what the user is talking about. Here it's useful to compare input from various experts, keeping concepts that appear often.

Using concept maps for assessment might involve allowing the rater or scorer to assess how well the subject identify both label and links. Her research, though, focuses on examining how well the subject identifies links between concepts, and rating the subject's results against experts' linkages. It's a practical method of applying concept mapping to assessment, because the scoring is a bit less involved.

[WUD] World Usability Day: First Session

World Usability Day 2006
Landmark College, Putney Vermont

Conference notes by Meg Houston Maker

"Usability of Prescription Pharmaceuticals"

John Urquhart, UCSF

How does one get accurate data on patients' use of prescription pharmaceuticals? One can take behavioral approaches, such as looking at patients' diaries, questionnaires, interviews, histories, etc. But only about 7% of the diary entries are done at the time of actually taking the medicine. You can do ad hoc methods, such as counting returned or untaken tablets, monitoring refills, or looking for chemical markers. Or you can monitor the medication events, meaning the entry into and exit from the package, which can be electronically time-stamped. This last is an indirect method, but it's highly reliable in most situations.

One early implementation of monitoring medication events was an eyedropper bottle (the medicine was for glaucoma) from 1976, which had a battery and magnet that would record, by flipping a switch, the removal of the cap and the tipping over of the bottle. Now, that doesn't mean the eye drop went into the eye, but the data from this unit was, at the time, breathtaking to the medical community. The bottle cost about $500 and weighed a lot, but by 1986 the electronics had been redesigned to be easier to use, smaller, lighter, and a bit cheaper at about $400.

The principle was later applied to oral dosage forms, including the Ortho dial pack for oral contraceptives. These were about $300. By 2000 there was an $80 bottle that monitors dosage. The only hitch is that with this, the patient has to bring the package back. Now they're developing units that have cell phones that let the data be sent back to a central system.

All of this needs a robust taxonomy. The process includes prescription, acceptance by the patient, execution of the drug regimen, and discontinuation (for whatever reason). So basically the process has a beginning, middle, and end, and the parameter "persistence" is how much they adhere to the regimen, and the parameter "quality" is how well they perform the regimen.

Persistence, over time, initially drops off precipitously as you lose those patients who don't accept. And then persistence gradually dwindles over time, so that by the time you're out one year, you've lost about 40% of patients, even for medicines that are for chronic conditions. On day 100, about 80% are still engaged in the regimen, but only about 60-70% of patients took the dosage properly. That gap is the gap in execution, and this can lead to other, larger problems, especially drug-resistant strains for drugs that create selection pressure.

These averages give you some idea of the herd, but doctors treat individuals, so you have to look at individual patients in order to perform good medical care. Urquhart shows individual patient data revealing certain patterns—from very steady and persistent, to weekend irregularities, to a "wobble" that may reveal a patient is about to discontinue.

Are variations in adherence a practical problem? We have examples from several major public health issues, from TB to oral contraceptives to HIV/AIDS. TB reached an apex of cases in 1992, and the only thing researchers could think to do was bring the patients in four times a week and watch them take their medicine. This worked, but that's a brute force solution not useful for, e.g., HIV care, because the doses are so toxic. The Norplant implant for oral contraception had a much better success rate than daily pills: 0.05% failure versus 0.1% for perfect use of oral pills or 5.0% failure for "wobbly" usage of oral pills. The same is visible in use of protease inhibitors for HIV.

The strong focus on the dosing process appears to reinforce patients' grasp of the importance of treatment, with longer persistence with treatment. This falls under the rubric of measurement-guided medication management. Showing the patient how they're doing, and intervening to show them what is optimal during an intervention and discussion session, will make adjustments. Doctors have to keep at them, though, to keep encouraging them to take their doses consistently.

In conclusion, usability of prescription pharmaceuticals is disappointingly low, but can be improved by a simple management technique, grounded in objective data on the patient's record of use. The issues are treatment costs, risk, and costs of failed treatment. Ideally the doctors would not administer this MGMM process—it would be managed by the pharmacists, nurse, or paramedicals. This nascent discipline is called pharmionics.

[WUD] World Usability Day: Keynote

World Usability Day 2006
Landmark College, Putney Vermont

Conference notes by Meg Houston Maker

Introduction

Dr. Brent Beatty, Landmark College

All Landmark all applicants have a diagnosis of learning disability or ADHD. One student remarked that he didn't have a learning disability—he had a teaching disability. Meaning that educators need to ensure that pedagogy fits the learner. This is where universal design fits in.

There are 218 events scheduled to occur today for WUD in dozens of countries. And here we are in Putney, celebrating and learning how to apply these principles.

Keynote: "Access Great and Small"

Sarah Horton, Dartmouth College

Show of hands: faculty, web developers, and librarians are about equally represented here in the room. (There are about 60 of us.)

"The person who is doing is the person who is learning."
—Chris Jernstedt

Jernstedt was last year's keynote at WUD. Sarah tells us to expect some interaction! Sarah says she's a web designer. Her original goal was to make attractive sites, but then she got excited about making sites in which the information is accessible even without the "Ah" feeling of beautiful design.

"I would rather that my sties attract attention because they are widely useful and usable than because they are pretty."
—Sarah Horton

How can we do this? Well, there are ADA guidelines, but, basically, you don't just want to build to code, because that can be the bare minimum. For web design, we have W3C and other web accessibility guidelines, but you wouldn't necessarily want to use then. There are a lot of accommodations we can do that will make our content accessible to people with disabilities, but Sarah's goal is to make sites that are wonderful and accessible.

Sarah likes Ben Shneiderman's definition of universal usability: "Enabling all citizens to succeed in using information and communication technologies to support their tasks." Sarah loves the word "succeed" here, because it's not just about use, it's about success.

Equitable Use: The design is useful land marketable to people with diverse abilities. This is the fundamental goal—to provide the same means of use for all users: identical whenever possible; equivalent when not.

This isn't always easy to do. Think of physical spaces, like restrooms. We have a larger stall for people in wheelchairs, and non-wheelchair users can use them, but wheelchair users can't use the other stalls. Sarah shows pictures she's taken in restrooms, including design choices about where to put the paper towel dispensers, where to put the soap, where the mirrors go, etc. Many in the audience chime in with their own experiences with access.

Physical spaces have constraints that mean we can't always accomplish universal usability. But the web is flexible, and it's modifiable. It's user-defined, so the user can make changes to the display of the information. Sarah shows the Wikipedia entry on universal access with different stylesheets to illustrate the ways this info can be presented.

Access
Someone has to decide what content to provide, and then make it accessible to everyone who needs it or would be interested in it. For example, at Sarah's son's middle school, the learning center makes the week's homework schedule available to students with learning challenges. But if you didn't have a diagnosis, you wouldn't know it was there. On the other hand, in the eighth grade, the material is just online for everyone who wants it. There's a low threshold of effort.

Examples from Dartmouth College
Sarah shows the Blackboard courseware system at Dartmouth. Students want syllabi, which some faculty have a hard time putting online, because it's an extra step, the technology is mystifying, they're concerned about IP, the material changes and professors don't want to commit to the syllabus from previous classes, etc. On the other hand, one professor is recording all his lectures and putting them on iTunes, essentially podcasting them. In this instance, there are no transcripts, but this is what Sarah means about creating access first, and then making sure that the material is usable to everyone.

Sarah writes books, and puts the full content of her books online to make the material accessible to everyone. Because after all, she writes them, she says, because she wants people to read them. Sarah challenges us all to think about what information we have that might be useful to other people, and then to provide the material to everyone.

[Mediacy of New Media] Keynote "Forget New Media, The Ship is on Fire!"

Conference Notes by Meg Houston Maker

Allucquere Rosanne "Sandy" Stone (University of Texas at Austin)

She takes a poll of the audience: What mode of speaker would we prefer? The "wise kind elder," or the "fire breathing barn burner?" The audience votes for the barn burner [MHM: I voted for wise kind elder, one of only two of us to vote this way. I'm always in the minority.]

She shows examples of her work, some of which concerns affordances and anti-affordance behavior, such as a sugar bowl that flinches away from the hand as it reaches for it, and a send-up of the film "Winged Migration" in which cleaning product bottles stand in for the migratory birds. [MHM: it's amusing, but it seems like it's supposed to be subversive, yet doesn't have that impact.]

New Media: a term first used in the 1920s, and referred to lantern slides combined with gramophone records, which was novel at the time. There is a separation between mentation and making. What is the conception of human existence that permeates our work and gives it meaning?

The arc of death-by-naming: something emerges as an oppositional (culturally oppositional), and you write about it, then other people start writing about it, and there's an efflorescence on the topic. People who study it meet at conferences, so it's a nomadic discipline. Then someone invents a jargon for it, so they can get a job, and suddenly, there are books in the bookstore with the jargon or name on the cover. So if you owant to keep your discipline alive, you have to focus on the framework, the metaphysics of the pedagogy.

The codeswitching umbrella -- at UTexas, the ActLab she runs is a messy, creative, unintelligible space, and she's responsible for codeswitching it into products, language, and structures that are digestible to the institution. The prime directive of the Actlab is to "Make Stuff." It should feel like a space of possibility. When students enter the space, they should feel like they've crossed the "limin," the threshold, into another realm, a realm of exploratory behavior.

Her program is interested in "knowledge in the body." She says, "Knowledge is in the senses meeting the world and feeding the imagination so... Engage the senses with purposeful physical activity." They do all the "fun tech stuff" but at the bottom of what they do in New Media work is about the metaphysics of healing.

[Mediacy of New Media] Panel 2: Text"The New and the Old in New Media"

Conference Notes by Meg Houston Maker

Jay David Bolter (Georgia Institute of Technology)

What constitutes a media in our culture is linked to our notions of what is old and what is new, and are constituted in their relationship to earlier forms. They depend on earlier forms for their cultural significance.

Everywhere on the web we find instances of sites that follow earlier forms: brochures, books, TV shows, etc. Borrowing from earlier forms. This borrowing happens in the early forms of any new medium. E.g. early films were like stage plays until it developed its own language. Or word processors, which borrow the assumptions and goals of typesetting.

But it's not only in the early phase that new media borrows from old media; that borrowing and refashioning happens continually. What defines a medium is its ability to be refashioned.

Marshall McLuhan (Understanding Media):
"The 'content' of any medium is always another medium. The content of writing is speech, just as the written word is the content of print, and print is the content of the telegraph." But his supposed a linear progression. But what we see is a complex media economy, with forms borrowing from each other. This is, in his term "remediation."

A culture has to recognize a technology as a medium in order for it to become a medium. Computing technology, when it was invented in the 1930s, were not conceived as medium. Our culture had to come to see them that way. Not until the 1960s did the medial nature of computers take hold. Transmission over networks facilitated that process. And adding the textual, visual, and audio forms within the computing environment. So as a medium, the computer inserted itself into our "media economy."

Those who contend that media should not borrow from each other are asserting Modernism, which claims that we must transcend the old to discover new ways of expression. That it's imperative to discard the old in order to develop the new. This is often combined with an essentialism that states that we must find the essence of the medium in order to express it.

"Repurposing is a transitional step that allows us to get a secure footing on unfamiliar terrain. Tub it isn't where we'll find the entirely new dimensions of digital worlds. We need to transcend the old to discover completely new worlds of expression." (Steven Holtzman, Digital Mosaics) Bolter completely disagrees with this stance.

Media designers and producers claim to present reality and an authentic experience. Are we in the era of the end of film? With the advent of games and other interactive media, we do see an economic threat to the film industry. But film still plays a defining role in spectacle in our culture. Film still reaches a broader audience, and the relationship of film and games is remedial, to use Bolton's term -- they are influencing each other.

Games appropriate, borrow, or transform elements within film, but layer on interactivity, giving the user control over point of view, temporal flow, and the narrative structure. The player or user can intervene to change the narrative. So the player becomes both actor and director.

So, how does film respond to this challenge, and incorporate games? Film has asserted the promise of realism and linear narrative. Filmmakers have adopted computer graphics techniques, redefining the look of the film and making for a new form of spectacle. They foreground and elide at the same time, e.g. in Jurassic Park -- we know that dinosaurs are not around to film, so this is not live action photography. We know there are computer graphics and animatronics behind these, but we're being asked, as viewers, to suspend our disbelief. [MHM: not uncommon with any media, esp. entertainment media]. This underscores the importance of transparency. The goal is to achieve a transparent reflection of reality. Meanwhile, interactive films have been largely rejected by Hollywood, which has made a commitment to narrative.

An aesthetic of multiplicity and hybridity is the counter to this. This has been expressed in the Avant Garde, from the 20s through the current digital Avant Garde. This hybridity is manifesting itself in pop culture now to an extent that hasn't been true in the past. E.g. MTV videos violate continuity editing and traditional narrative structure. Our current culture pursues hybridity and multiple images, in multiple forms -- think of the mall, the sports bar, in social computing spaces like MySpace and YouTube.

So, any dichotomy is really there to be collapsed: the transparent and the hybrid. The DVD (a movie DVD, to be specific), allows the viewer to be an engaged actor with the material, the making of the material.

The goal is to achieve authenticity of experience for the viewer.

[Mediacy of New Media] Panel 2: Text"Formal Materiality: Notes on Mediacy and Computation"

Conference Notes by Meg Houston Maker

Matt Kirschenbaum (University of Maryland)

Forensic materiality: Relies on the fact that in the material world, no two entities are exactly alike. The potential for individuation or individualization. The trace within material of the recorded information on the material (silicone, etc.)

Formal materiality: A formal environment of symbol manipulation without the imperfection of messy concrete materiality.

[What he's really talking about here is simply the expression through UI of encoded information.]

Allographic vs. autographic representation. Autographics are concerned with the exact instance of a representation, as whether a text is original, or whether it is a copy; these are distinct. Allographic representation is concerned with whether the two are ontologically identical.

So, digital computation is fundamentally allographic -- in fact ideally defined (through characters, numbers, and, at lower symbolic levels, 0s and 1s). [MHM: we're talking about levels of grain, here.] Computers are medial in presenting a premeditated material environment that is inherently immaterial.

Information is encodable, but not always decodable. Formal materiality is a normal condition of working in a computational environment [MHM: as long as its a symbolic computing environment].