I’m proud to say that clement greenbot, a Markov chain-based textual analysis of the art critic Clement Greenberg, is live and tweeting. Every few hours, it tweets randomly constructed sentences that, in terms of the probability of one word following another, are probabilistically similar to Clement Greenberg’s writing (specifically “Avant-Garde and Kitsch,” “Collage,” “The Cross-Breeding of Modern Sulpture,” and “Modernist Painting”).

Some things I learned:

Random isn’t always random. Wrapping your head around Hans Arp’s process for a work like the one above, Untitled (Collages with Squares Arranged according to the Laws of Chance) (1917), it becomes apparent quickly that Hans Richter’s account of that work’s genesis–that Arp took a sheet of paper, laid it on his studio floor, and then took another sheet, regarded it for a moment, and “finally [he] tore it up, and let the pieces flutter to the floor of his studio”–seems unlikely to be the whole story. Did he select only those objects that made slight modifications to a gridded composition? Did the air currents flutter just so on that day? And so on. This was true for clement greenbot, too: walking through Greenberg’s corpus randomly produced some interesting results, but it also produced so much noise that it wasn’t clear Greenberg’s work hid in there at all.

Really, it’s not random. These essays have a few distinctive phrases that mean if clembot walks into the their first words (“ineluctable”), the next is sure to follow (“flatness”). If you watch, “ineluctable flatness” comes up with a striking frequency. While this makes the text distinctly Greenbergian, it also made for some weird moments where I thought I had truly made my bot brilliant, only to discover that it was just good ‘ol Clement doing that. Likewise, when clembot walks into a list of artists or poets (common especially in Greenberg’s writing on modernist painting), it usually finishes the whole thing.

No, it really isn’t random. Finally there is the problem of the simple legibility of the sentences generated. They needed a little cleaning of their punctuation and of how long they were. While always having the sentences end with a period is a fine result for my purposes, it does point out the limits of pure chance as a way to analyze a text.

Anyway, it was fun to build (many thanks to Shabda Raaj, whose Markov example chain code is baked into clembot) and sort of fun to read. As a method of analysis of humanistic data, submitting a text to chance still seems to me like a useful approach, but I’m not entirely sure if the Markov chain itself is the best way to achieve this. It may be, for example, that simple chance, or a differently-weighted probablistic approach would be more insightful for a given set of data.

Source: Clement Greenberg, Arranged According to the Laws of Chance

Tomorrow is the final day of “Building a Digital Portfolio,” and with it comes an opportunity for reflection. When I applied for the workshop, I carried several vaguely-defined projects in my head that sounded interesting, but I had no idea of how to pursue them. I attended Digital Humanities events on my campus with great enthusiasm, but wasn’t sure how to engage in meaningful conversations on the projects discussed.

At the end of the two weeks (and a lot of reading!), I feel confident about my working knowledge of the field and relevant vocabularies. As for my projects? It’s a somewhat more complicated story that is best explained by framing them as short, long-term, and intermittent projects.

Short Term: 

In previous posts, I discussed the possibility of creating a map related to a dissertation chapter, and evaluated different platforms that might be able to support the project. Earlier this week, our session dedicated to mapping platforms helped me to identify what exactly I need my map to do, and how it might function within my larger project. This, however, opened a pandora’s box of questions with regard to the role of this digital tool in my dissertation. I began to wonder if every other chapter needed a corresponding tool, and how to remain consistent without becoming redundant.

As these questions were mounting, we read Paige Morgan’s essay “How to Get a Digital Humanities Project Off the Ground,” which advises against deciding hastily to incorporate a digital component into your dissertation. As a result, I’ve tabled the plan to incorporate the map into the bound version of the dissertation. However, there is still value in creating the map, which consists of a modest number of data points, and can help me to prove the chapter’s overarching argument. When I return home, I plan to begin locating historic maps of Los Angeles that I can use to overlay the generic map that CartoDB offers so as to more accurately represent the changes to the landscape of Watts between 1965-1967.

Long Term:

The idea that compelled me to apply to the workshop initially is a desire to build a database of Los Angeles artworks sited in public spaces (that don’t necessarily fall into the categories of “public” art or “street” art). This is a monumental task, and could constitute a dissertation project in its own right.

I’ve determined to call the task of building this database my “second book project,” which is a catch-all term I generally use to describe everything I’m interested in that can’t fit in my dissertation. However, after hearing Michelle Greet discuss the Transatlantic Encounters database and how it evolved in relation to her second book project, such an undertaking seems quite possible. Although I can’t dedicate a year to building the database right now, I know that I can start collecting data while researching my dissertation, and possibly even build relationships with some of the arts organizations I am writing about that have coordinated relevant projects.

Intermittent:

In the midst of strategizing on the short and long-term, this year, I will be teaching two courses: a 70-student class on Chicana/o Art, and a 20-person course on U.S. Art and Visual Culture. In both of these classes, I plan to incorporate digital tools in various ways, and have begun reading about digital pedagogy.

For the Chicana/o Art class, I recently learned about El Museo Eduardo Carrillo, an online archive of works by an Eduardo Carrillo, an important Chicano artist who lived and worked in Santa Cruz and its related rotating exhibitions that features essays by key scholars in the field. In addition to creating an assignment that will invite my students to engage with these materials, I am also considering incorporating an image annotation assignment, as well as a “Wikipedia-editing” assignment that would be directed toward the lackluster “Chicano Art” entry after reading Adeline Koh’s piece featuring suggestions for introducing undergraduates to the digital humanities.

My U.S. Art and Visual Culture class initially planned to use Scalar as a platform for showcasing collaborative student projects, but am still considering what I want my students to take away from the course, and whether Scalar is the right platform for those objectives. Throughout both courses, however, I plan to blog about the process of introducing digital projects into my courses, so stay tuned!

All of this is to say that the workshop has left me with not only a new enthusiasm for my research and teaching, but the world of collaboration and conversation that exists within the digital humanities and digital art history. I not only feel as if I have found a community of scholars whose work resonates with my own, but that within this community, it is possible to contribute to the conversation in a variety of ways.

Source: A Template for a Digital Portfolio

It took nine days—or 54 classroom hours—of digital humanities immersion to finally take the preliminary steps in a digital component of my research project. The acquisition of skills (or really, tool know-how) paradoxically led me to shed most of my paradigm about working digitally. Over the last two weeks, I became so full of information and ideas that my only choice yesterday was to take one tiny step. So, around 2:45 yesterday afternoon I opened an Excel document, placed my cursor in the top left cell and typed “Last Name.” Yes, I need to build a data set. I realize now that I was not at a loss for where to start or what question to ask, I was in denial. I thought that compiling a basic data set with artists’ biographical information somehow did not count as research or was not essential. I realized that all along (this week and in the grander paradigm-constructing scheme), I had tacitly based my decisions about what digital methods and tools to implement on a vision or expectation of particular end products. After spending just one hour compiling tidy data about artists in my purview I found gaps in my research and questions I need to answer before moving forward. I realized I needed to get to “zero” before anything else could happen. And so yes, I had the “a-ha!” moment: it’s really not about the end, it is about the means. I knew this to be true for “traditional” research, but had not yet experienced it while testing any of the tools in the workshop. Data-entry is the tip of the digital iceberg. But inputting data brought clarity to my work. My motivation to continue exploring the range of methods has shifted from an abstract enthusiasm to a steadfast pursuit of what is necessary for me to accomplish the most sound scholarship.

Thank you CHNM. Onward!

Source: One datum at a time.

The Andean landscape is striking and breathtaking for the dramatic scale and range of its geographic features. In the north-central plateau of present-day Bolivia, within the span of hours one can travel from vast Lake Titicaca’s shores, which edge onto eastern Andean mountains peaking at over 20,000 feet, and descend some fifteen thousand feet into subtropical forests and valleys. Both in its natural formations and in its shaping by human activity, the landscape has long imprinted itself on the identities of native communities and practices.

My research into Andean visual culture looks at the Andean landscape as a site of resonant meaning during the early colonial period, when new claims to land were reshaping identities, and examines the woven textile as a representation of ties to landscape.  I have many questions rather than firm positions on how to understand textiles in relation to land but my hope is that visualizations of data regarding textiles will help bring my questions into a clearer orbit.

Textiles in indigenous Andean cultures are as layered with meaning as is the Andean landscape. In its attributes, a textile can reference gender, status, wealth, community ties, agricultural productivity, local topographies, as well as belief system.  Partly, what I want to do in creating visualizations of textiles according to sites of origin and attributes is to begin to understand the corollaries between certain design features and specific regions or communities. Can we trace the way a textile pattern of a lake, for example, corresponds to local features of the landscape and how far do such associations between textile representation and landscape feature reach within a region? Where does one landscape feature take prominence over another in a textile representation and for what possible reasons? Where and why did certain features of landscape construct identity, to be seen in the textile as it was worn on the body as a second skin? An offshoot of these types of questions is to see to what extent it is possible to trace the movement of textiles between communities, storehouses and centers of power in the Inka Empire and then in the colonial period.

During the Inka period textiles were strongly associated with their sites of production but they were also highly mobile objects. Not only were textiles moved and exchanged like currency within the tribute system but they were also active as objects worn on moving bodies, and can be understood as representations of landscape that moved through landscape.

In certain cases and as a facet of Inka imperial strategy, certain communities would be moved across the landscape but would have to maintain dress that identified them with their place of origin. In the colonial period the Spanish would become very involved in the production and movement of textiles as well and would exploit indigenous weaving while simultaneously harnessing notions of landscape to a European understanding.  Within this scope the questions then revolve around how the Spanish reshaped the landscape and how this came to be represented, perhaps in contrast to the indigenous textile with its representational significance. Can I use spatial visualizations to pose these questions in ways that lead to deeper investigations of the visual culture of the colonial period?

Source: Tracing landscape in textiles of the colonial Andes

Although today’s theme at Building a Digital Portfolio was Project Planning & Grant Proposals, what I truly received was a lesson in the power of generosity.

Generosity in academia is a peculiar problem. Certainly, academics need to be conscientious of their intellectual property and protect it. But this does not preclude an attitude of scholarly friendship and openness.

Scholarly friendship can be enacted through providing access to the data that you collect for your own research. Today, Michele Greet visited and presented on the process of creating Transatlantic Encounters: Latin American Artists in Interwar Paris (http://chnm.gmu.edu/transatlanticencounters/). Her project is an example of generosity—it shares the data she meticulously collected. For me, the most notable remark she made was that the digital project, instead of hampering her ability to publish and present her research, has provided opportunities to write articles and lecture.

In the afternoon, we had free time to work on our projects. I have been collecting a very detailed list of works by the artist I am researching. In the midst of discussions about different platforms for project planning, I ended up on a website I use occasionally to find information on lesser-known objects—www.wga.hu. Staring at the site I realized how much I wanted an excel file of the data they collected. Much to my surprise, I found that they had provided an excel file of their data for download. What generosity!

Today was an affirmation that generosity is important in academia. Being fearful of opening your own data to others is unproductive. Instead, embracing sharing and generosity as the core underpinnings of your academic life aid in forming a rich and supportive scholarly community.

Source: Generosity and Academia

My weight shifts from the heel of my left foot to the ball of my right. From taking the weight of my body, my right big toe begins to strain; the back left part of my ribcage starts to slowly shift toward the left to compensate. Then my left knee bends slightly and holds my bodyweight so that my right foot can slide forward one centimeter.

Heat is building deep in my abdomen and expands the space inside my torso. The pressure of expansion causes my breastplate to vibrate with quick, low-frequency flutters. The muscles in my chest, neck, and shoulders activate and begin to draw energy from what is now a current of warm electricity circling through my torso. Then, abruptly, the top of my right shoulder surges with energy. This same wave then shoots to my right elbow and fizzles at my wrist. Then again–an energy surge–but this time it travels down through my legs, to my knees, ankles, and feet. It escapes through my metatarsals into my shoes and dissipates as it reaches the floor. Now the floor is electrified beneath my feet and the current is grounded. The once local current now travels through my arms, my chest, my back, my legs, and the floor. My face is active; the muscles are engaged and directed to look, but they are continually receiving secondary and variable waves of radiant energy.

As the current travels through each extremity it reaches a boundary: a web of bound muscle. Just as my facial muscles are directed to do the activity of “looking,” the muscles in my body are directed to stay still. To unbind would be to cease directing my muscles and would release the current. As the current moved it would bring with it the muscles it touched and they in turn would bring the bones, with them. My entire body would be in motion.

This is what happened when I approached, then directed myself to look at Joan Mitchell’s 1960 painting, Marlin, on Saturday at the Smithsonian Museum of American Art. I stood in front of the painting for the next ten minutes (or more?) I traced the colors, brush strokes, textures of the canvas and the intersections of these elements with my eyes. As my eyes moved, the current of energy simultaneously moved through my body. My mind–trained to analyze based on what it sees–periodically interrupted the kinesthetic experience of the painting. But if I do not want what I see to become a set of judgements and determinations, I choose to quiet the mental chatter and allow the sensation of motion to dominate. In this way Marlin is, for me, as much a choreographic score as a manipulation of pigmented oil paints on a stretched canvas.

Physiology might be important. (This is the Joan Mitchell painting.)

Maybe it’s too convenient that the kinesthesia I am describing was in response to an Abstract Expressionist painting. Maybe I was just having the intended experience with the work. I know nothing about Mitchell’s intention in painting Marlin, and certainly don’t know what she expected viewers to sense when they approached the painting. I do know that my kinesthesia is not limited to this type of art. For example: I walked into the Peacock Room yesterday and felt the movements associated with the labor of creating such a space.

Interoceptive kinesthesia is only part of my physiological experience of art—particularly to those works that occupy three-dimensions. When I enter a space I immediately have a swath of information to process just based in my visceral response to external stimuli. Instead of an internal current of energy, I experience aspects of the space as sensations on my skin, by the quality of my breath, or by the movements of air. My body acts as a scanner, it picks up the dimensionality of an environment and my mind notes when there are correlations between changes in the composition of the space and changes in my physiological experience.

While I rarely make claims in my research papers that my physiological response to an object is a sound basis for scholarship, I do allow what I learn from this response to direct aspects of my work. When I observe my kinesthetic response to a painting without imposing descriptors and commentary, I can later (even within the same session) return to the painting and see aspects of the form that were not apparent at first glance. When I am in the presence of a sculpture I observe my visceral response at many distances and positions. My body learns far more quickly than the analytical part of my mind so throughout the course of researching and thinking about a sculpture, I refer to my memory of the physical experience in the presence of the object. I am sensitive not to conflate my assessment of the work in an art historical context with my experiences of an object. I merely use my body memory as a guide to take me back through aspects of the work.

I can witness such feelings and sensations because I have had the privilege in certain instances to let go of many of the hegemonic forces present in a museum or institution that we’ve identified throughout this institution. My self-disclosure here is intended to show the value of sharing personal experiences of objects and to provide an example of an alternative learning style that museums can foster by breaking down the curator-as-expert model. I am certainly the type of person who benefits from engagement strategies that invite experiential access to information about an object. To me it is a shame that when people enter a museum they often believe there is one way to experience a work of art. From my experience, opportunities for alternate modes of understanding (art or anything) are restricted when all my efforts are directed at having one prescribed experience. For viewers, including myself, this stilted experience can lead to any number of negative feelings.  Potential art lovers are then driven away from the thing you, as the scholar or curator, are trying to call attention to.

Positive or negative, one person’s physiological response to an object is unique to them. As facts become easier to find and access, the museum can no longer merely serve the viewer information, it must provide the means for the viewer to have an individualized experience with art. Your feelings in the presence of art are yours alone and cannot be found on a database.

When digital tools are used to elicit experiences beyond fact gathering they can reinforce or generate new positive associations with one’s unique response to an object. An animated map triggers a kinesthetic response, a favorited Tweet triggers our reward centers, and a change on the museum’s website because of a viewer’s complaint exudes the warmth of empathy. Used well, digital tools can offer broader understanding of an object, historical narrative, or institution by tapping into the omnipresent and extremely rich functioning of our physiology.

Source: The Feels

On the first day of the Building a Digital Portfolio institute, I told the group that my biggest anxiety about using digital art history methods to investigate my dissertation topic had to do with limited access to digitized data. I went on to complain in my first blog post that it is presently impossible to carry out a visualization project on fifteenth-century Naples because no structured data set exists for the topic. In the course of this past week, I have already begun to build my own data set to remedy this problem. This new project is one that I have contemplated for over a year now, and I have failed to get started because it seemed too large a project for me to undertake alone at this stage in my dissertation research. The idea of building a data set also made me uncomfortable because so many aspects of my dissertation topic lack firm documentary evidence—a point that was made even more clear to me by the word cloud of my dissertation I created using Voyant Tools, which includes the high-frequency words ‘probably’ and ‘likely.’

Voyant word cloud of my dissertation, with the words ‘probably’ and ‘likely’ included.

Although there is much about my dissertation that I can’t responsibly confine to a single cell in a spreadsheet, I think there is a great deal that I can learn about fifteenth-century Neapolitan art, and the patronage of King Ferrante in particular, by trying to visualize the limited documentary evidence that does exist. Fortunately for the Neapolitanists of the world, Gaetano Filangieri compiled six volumes of primary source material on art of the Kingdoms of Naples and Sicily, (Documenti per la storia, le arti e le industrie delle provincie napoletane) published 1883-1891. These volumes (now available via Hathi Trust) are invaluable, since many of the sources Filangieri compiled were destroyed in the bombing of the Neapolitan archives in 1943.

Sample page from Gaetano Filangieri’s Documenti per la storia, le arti e le industrie delle provincie napoletane (v. 5).

Rather than continuing to bemoan the lack of digital source material in my field, I’ve decided to focus my efforts on building a structured data set from this one critical source that is available. My project will focus specifically on Filangieri’s volumes 5 & 6, which provide chronological lists of archival material (often including the artist’s birthplace, trade, date and description of the commission, patron’s name, and the amount an artist was paid). Within those volumes, I’m focusing on records from the dates of King Ferrante’s reign, 1458-1494.

So, what do I hope to accomplish with this data?

I’m interested in visualizing, for example, the most active patrons and artists in the kingdom, and the networks among them. I want to examine the geographic origins of artists working in the Kingdom of Naples and to consider different patrons’ preferences for hiring artists from particular regions. There are many other questions relative to artists’ origins, trades, and earnings that I could study through this data set, too. These are quite elementary art historical questions, but because scholarship on Neapolitan art of the early modern period lags behind that of other Italian cities, these basic questions still require attention, and I think data visualization could be a useful way to address them.

I have only worked through the letters A, B, and part of the C surnames in Filangieri’s publication, and I have already made some interesting realizations. This morning I uploaded a PDF version of my data set to Voyant to examine the geographic origins of the artists working in the Kingdom. I was surprised to find that Cava dei Tirreni appears to have been a major producer of artists in this period.

Voyant Word Cloud of the origins of artists working in the Kingdom of Naples between 1458-1494.

I then drilled down a bit further in my data, this time using Tableau, to see how the different trades of artists coming out of Cava dei Tirreni compared to those from Naples. I found that Naples produced many painters, silversmiths, goldsmiths, glassmakers, tailors, and sculptors, while Cava dei Tirreni specialized in stonemasons, architects, and builders.

Tableau Bubble Plot of common trades in Naples and Cava dei Tirreni.

Of course, this data set is skewed because a couple of large families of builders from Cava dei Tirreni have surnames in the early part of the alphabet. Although I can’t make any conclusions until I have visualized the entire data set, I’m already getting excited about the power of tools like Tableau and Palladio to macroscopically examine characteristics of fifteenth-century Neapolitan art. The challenge for me at this early stage (in addition to building a tidy, error-free data set), is deciding which tools and types of visualizations are best suited to the questions I’m asking.

Source: Wading Into Data

Do art historians have data? The answer seems to be yes. Once data is defined not as not only numerical values much of what we consider can be “data.” At it’s most basic level, formal analysis generates data. We start to deconstruct an object through our descriptions and look for diagnostic characteristics. Historical and archival research is then used to contextualize these features. However, these characteristics can also be expressed in a spreadsheet, but in a concise, standardized and organized manner.

Visualizations help us deal with this data and can be a apparatus for finding and interrogating trends within our corpus. They are not independent conclusions, but one of the many tools an art historian can draw upon. The visualizations lose meaning without an accompanying narrative. Graphs and maps can be misleading, which makes it even more important to explain the variables it embodies.

Another application for visualizations is in talking to non-specialist audiences. After years of specializing in our chosen fields (and sub-fields), much of our professional life is spent dealing with non-specialist audiences, from writing grant applications to teaching. Word clouds may be imperfect tools, but they seem like a perfect entryway into historiographical discussion and a way for students to see latent biases. Annotated maps are essential; every non-Western art class begins with a map exercise, stressing the importance of geography and location for the formation of art, culture and history. Additionally, we cannot ignore the connectedness of the world and the layers of interaction and multiple networks it engendered. Digital visualizations can allow one to consider these non-linear relationships as dynamic and fluid and prove to be tool for deepening our research.

Source: Visualizing Data

My favorite “digital” way of analyzing texts is to turn them into gently salted nonsense. Darius Kazemi’s “Hottest Startups” bot, which fuses the language of startups and marxist theory has probably taught me more about the relationship between those things than another collaborative monograph on the state of collectivity ever could. Nothing detourns a text like a bot’s combination of automaticity and serendipity.

One way of building bots like these is to store the texts they draw from as what are called Markov chains. The general gist is that a text is stored as words and probabilities. In a given text, a word–say, “the”–has a certain probability of being followed by another word–say, “avant-garde”–and a different probability of being followed by a different word. Stored this way, one can ask the Markov chain to generate text that seems superficially similar to the original text by chaining words randomly (but weighted by probability) through the chain. The idea is that Markov chains conveniently store a dessicated kernel of a given text, but given a single seed word it is unlikely we will get the exact same text back, but rather a markedly similar (and lightly nonsensical) text.

Some example sentences from a forthcoming bot, seeded with Greenberg’s “Avant-Garde and Kitsch”:

Culture down to their own primitive folk art as is the culture of the few are shared by the general. even then, however, the resentment.

Their audiences. it becomes difficult to assume anything. all the verities involved by religion, authority, tradition, style, are thrown.

There is no longer look toward socialism for a new culture — as inevitably as one will appear, once we do have socialism. today we to

Already (and my implementation is currently so, so brutal) there are some key (early) Greenberg terms here–socialism, folk, authority, and so on. Even ultimately Friedian terms like “thrown” show up, but detourned away from Fried’s ontological thrownness.

Anyway, hopefully this will appear in forthcoming twitter bots about art history. Someday soon, be on the lookout for askHeiny, a Heinrich Wölfflin-flavored nonsense bot. Stay tuned!

Source: askHeiny!

Let me start by saying that I’m a huge proponent of 3-D modeling, the topic of Day Four of Building a Digital Portfolio. We started our morning by debating the value of semi-immersive virtual reconstructions, and I argued in support of them 100%. I have used SketchUp in my own research, not necessarily as a means of illustrating a space to my audience, but as a way of learning more about my topic (in this case a fifteenth-century Neapolitan funerary chapel, known as the Succorpo of San Gennaro) through digitally reconstructing it. Learning through building—one of the main values of SketchUp according to my supervisor at the University of Maryland’s Michelle Smith Collaboratory Quint Gregory. I noticed things through this process that I’m confident I never would have recognized by merely looking at photographs or standing in the physical space of the chapel.

Workshop of Tommaso Malvito da Como, Succorpo of San Gennaro, Cathedral of Naples, 1497-1506. Image: Artstor.

I’m also a supporter of using 3-D models to reconstruct and preserve knowledge about rapidly deteriorating archeological sites. My first introduction to virtual reconstructions came in the summer of 2010 when I participated in the Restoring Ancient Stabiae Foundation’s archaeological dig at the Villa Arianna. Because of its recent excavation and the fact that it is not yet open to the public, the Villa Arianna (at least in 2010) was not in as poor a state as many other ancient villas on the Bay of Naples. The value and urgency of documenting every aspect of these places, as soon as humanly possible, was abundantly clear to me on my first day at Stabiae. The art and architecture of these buildings will not be around forever, and so much of the artwork has already been removed to museums that it can be extremely difficult even for the informed (non-specialist) art historian to appreciate and understand the sites they’re visiting. I think there is an enormous value in virtual reconstruction projects like: Rome Reborn or The Oplontis Project, and I imagine that the degree of material loss in one’s field influences one’s opinion on the significance of virtual reconstructions. If you’re studying lost, largely destroyed, or rapidly deteriorating artworks, the need to document and preserve those originals probably feels much more urgent than if your subject matter can be accessed and understood relatively easily by students and the general public.

That said, the conversation we had this morning about the importance of expressing uncertainty in a virtual model struck a chord with me. Participant Bethany Ferrell Rivello suggested that art historians take a conservationist approach to these types of projects, one that somehow reveals the degree of change the digital restorer has made from the original, as well as their degree of [un]certainty. Whether through color, transparency, or some other digital mediation, I would agree that the creators of digital models should make abundantly clear to their audiences which aspects of their reconstructions are based on strong documentary evidence, and which are scholarly conjecture. I think that we all win if we make honest, accurate virtual reconstructions publicly available in a way that serves the needs of multiple audiences.

Source: In Defense of the Virtual Reconstruction