It is time to resume our public discussion (this is the content of a recent report Vika and I sent to NEH): now that the first phase of intensive encoding is completed, over the next few weeks we will concentrate our effort on designing the home page and, along with it, the second major component of the VHL, the Virtual Seminar Room. There will be a single “door” into the VHL, with links to its various components, including links to the Catasto/Tratte databases, which won’t be “integrated” into the VHL but will be part of the research and pedagogical activities made accessible through the VSR. The VSR will be the entry point for a number of collaborative activities based on our encoded texts (including the Decameron) and made possible by the capabilities of the VHL platform (annotation engine, etc.). The idea, as you know, is to combine traditional study and research activities (the reading and interpretation of texts), distance collaboration and the digital editing process, broadly conceived: participating scholars and students will be able to contribute annotations, participate in the process of semantic encoding etc., as well as interact and communicate with each other in this virtual space (we are planning to add a chat room). The VSR will thus provide a testing and training ground for those collaborative practices made possible by the digital platform. Feedback from scholars and students will also determine the evolution of the VHL, the additional tools, texts and functions that we will include in the future.

Of course, during this brainstorming phase, feedback from every member of the teams (Esposizioni, Pico and Villani) is vital: help Vika and I conceptualize the kind of activities that your specific scholarly community (literary scholars, historians, philologists and philosophers etc.) may find most useful and compelling; and help us envision how these activities could become part of graduate or ug. seminars and courses focused on, or including one or more of our texts.

Upon thinking further about the search functions for the Villani text, I was wondering/hoping that our encoding would allow us to make searches that involve those encoded elements. Simple string searches may currently be done on the OVI database, where Villani’s Nuova Cronica also resides. Will we be able to search for occurrences of Person/gender/female within the same paragraph with the word “Francia”, or examples of Person / position / priore in close proximity to Theme / econ? Will this open up a world of complexity that will cause the program to crash, or is this exactly what we are aiming for?

As we look more closely at the Villani encoding, we are looking to clean up and simplify as much as possible. Some series of attributes are naturally infinite, such as the names of persons or names of places. On the other hand, we have one set of attributes “natura”, under type=”forza”, under the element “theme”. They are as follows:

Giustizia, forca, Condanna, Multa confinamento, Prigione, Guerra, Guerra esilio, ribellione, attacco, Assalto, Confiscazione, Mutilazione, Incendio, Assassinio, presa citta, Guerra presa citta, Congiuro, Arresto tortura, Assissinio mutilamento, Assassinio, Mutilamento cannabalismo, Esilio, Guerra presacastello, Assassinio, Blocco, Guerra, Assedio, Ostaggio, Assedio, Ribellione Guerra, Condanna esilio, Scomunica, Tortura forca, Mutilamento, Confinamento, Furto, Tortura multa, ribellione, Guerra cattura, Presacitta, Presacastello, Giustizia, Confina, Forca cattura, forca condanna

Would it be helpful or not helpful to combine these into groupings? Groupings like punizione civile, punizione ecclesiastico, guerra, for example. Would that add unneccesary complexity? Or would it be best to leave these forms of force in their original wordings as we did with the person: “name” attributes, and give up trying to classify them.

I imagine a sort of grim catalogue of violence, and index of evils that could be drawn from this encoding. Villani has his spicy side too. Any suggestions welcome.

I’ve been working on Boccaccio’s citations of other literary sources in the Esposizioni, and was wondering how we were going to incorporate them into the online text. How about if I add the references to sources in as annotations to the section of the text that we have up now (Esp., Acc., I-IV)? I thought I would simply add the references as they appear in Padoan’s notes to the Mondadori edition (Milan, 1965), and then others more erudite can come after me and make further comments or corrections. Or would it be better to wait and encode them instead?

I agree with Mike that the Balzac example is an interesting one, although it clearly applies to a corpus (oeuvre) by a single author. From this point of view, let’s remember that VHL is not a “single author” project - I find more affinities with the WWP or the EEBO. Of course, the search engine is a valuable tool for annotating. However, what our search engine should be able to do, eventually, is to maximize the possibilities embedded in our “differentiated” encoding. For example: crossreferencing names, places, dates, visualizing text strings and paragraphs etc., but also allowing to perform more sophisticated searches for authorial, thematic, semantic/rhetorical structures as we identify and encode them in the various texts (what fields would be appropriate for these other tasks?). Our goal is to enable a comparative and explorative approach to texts that belong to the same cultural context but also to different typologies of writing and rhetorical genres (we have chosen these texts precisely because of the wide spectrum they represent). How does the search engine help us reach that goal? Another question raised by Mike: keeping commentary and text separated is ok, but isn’t encoding a form of embedded commentary? Does Mike mean annotations? Will we be able to search annotations as well - in relation to text - once we have a significant amount of annotations? I suppose we can proceed by stages and add functionality and power to our engine as we progress in the encoding and annotating process. However, in designing it, one of the fundamental prerequisites we should keep in mind is its “expandibility” - to keep it open to the possibilities that lie ahead of us, including potential applications in the seminar room.

It’s good to hear that our annotations can include links to other web pages. Might they also include images? I’m annotating the war between the English and French. As you might imagine, it would be useful to provide campaign maps and some of the many images I’ve found.

I have begun work on the annotation of Book Thirteen, choosing to gloss first those chapters that deal with Queen Joan of Naples, the spicy intrigues that occurred around the murder of her husband Andrea of Hungary (exciting also because they cause Villani to explode into unprecedented fits of moral outrage). More generally, I will annotate the succeeding chapters dealing with the episode, and Villani’s treatment of the reaction of the Hungarian royalty to these events.

I am looking forward to highlighting the connections between Villani’s portrayal of these events and Boccaccio’s representation of Queen Joan in his De Mulieribus Claris of 1362. Joan has a prominent place in this group of illustrious women, being the end point in the series of 104 biographies.

One question, and this for the technical staff…can the annotations include weblinks to other documents? I hope the answer will be yes.

In the meantime - - and this is for everyone - - I am trying to locate the e-text of De Mulieribus Claris in English, Italian and Latin. If anyone happens to know reliable etexts and where they reside, please let me know. - - Rala

During the last week of encoding, I have been dealing with a bunch of “double names” (or even “triple names”). The typical example is a character that had (fairly accurate) equivalents in the ancient Greek and Roman worlds, and then sometimes a “third” name in medieval Christianity. A fex examples will make this point clearer: Medusa-Gorgone, Furie-Eumenidi, Parche-Fate-Fato etc. In these cases, for the glosses, we employ the tool seealso=”", which could be useful for names as well.
Roberto Bacci

We have reached a point in our biennial NEH grant period when we can make a general assessment of our progress and look carefully at upcoming challenges. I would like to publicly praise our project director for her tireless efforts in keeping all of us committed and getting things done. Thanks to Vika and Paul, great progress has been made in the overall design infrastructure of VHL. One of the major challenges we have encountered so far is the relative gap in the advancement of the various projects that compose VHL (Esposizioni, Conclusiones and Cronica). The encoding of Esposizioni is proceeding at good pace (thanks in particular to the excellent work done by Roberto, this semester, but also thanks to feedback from all members of the team, Mike in particular, and Guyda). The Pico Project is now ready to enter a new, exciting phase, that of active, collaborative annotation, involving some 25-30 scholars over three continents (our thanks to Francesco Borghesi for all his painstaking work of coordination, over the past few months). Pico and Esposizioni are more advanced than the Villani project, and a genuine effort will have to be made now for the latter to meet the goals we assigned ourselves for the second year of our grant. While the different stages of development of the various projects do not represent a problem per se, it is of the utmost importance that we try to bring them all to a level in which the common framework and interface provided by the VHL can be effectively exploited. This is particularly true for the Boccaccio–Dante-Villani component, as the possibility of cross-referencing these texts is greater and also depends on our encoding choices. First of all, it is crucial that the XML encoding schemes for both Esposizioni and Villani continue to be publicly discussed and tested, hoping for a feedback from the scholarly community at large. Documenting our encoding practice, as we go, is therefore essential. The blog is an excellent tool for this purpose. The more we blog about our work-in-progress, including our doubts, problems and difficult choices, the more feedback we can expect from our team members as well as scholars elsewhere. Moreover, we can now begin to at least think forward to further developments of VHL. As you remember, its two fundamental components are what we call the Editing House and the Seminar Room. Both together make up our Laboratory (or Workshop). Both share the same platform and a suite of tools to be developed as we go. Let’s think of these two components as both virtual meeting and working spaces and a set of scholarly and pedagogical practices. While the Editing House is taking shape, we should also begin to envision what the Seminar Room may be like. Searchable and interactively annotated editions of our texts provide also a platform for a number of learning activities. When we look at encoding, we should always keep this (its potential application in the “virtual classroom”) in mind. On the annotation side, the Pico Project may serve as a prototype. The experiment in collaborative annotation about to begin can also be a test for interfacing research and pedagogical activities: annotating can be conceived as an ongoing seminar in which participants learn from each other (the complexity of Pico’s text imposes more than allows this). I expect our weblog to be also a venue where methodological discussion about the annotation process and its procedures takes place. Next semester, the Decameron Web will also be “reactivated” for the course I’ll be teaching again for the first time in four years: I hope this will offer us another opportunity for a general updating of its content as well as for thinking about its potential improvement within the framework of VHL.

Technical problems all fixed…now I can blog!

A quick update on the Villani Project lest anyone presume that it has fallen into oblivion. As of the middle of last week Rala and I posted our complete encoding of Book 13 of Villani’s Cronica to the server.

When we began this project we encoded a handful of themes and a little bit of information about persons named in the Cronica. Over the course of the year we simultaneously narrowed and simplified our method of encoding themes, broadened our treatment of persons, and encoded new information about textual objects (places, buildings, offices, peoples, family names) and textual structure (citations of various sorts, Villani’s “moralizing” interventions in the narrative, etc).
Undoubtedly we will be “tweaking” our encoding and, in time, adding still new categories so we would really appreciate any comments participants might have on the encoding as it stands now–especially on the encoding of “themes”.

Now the bulk of our effort shifts to the annotation and, to a lesser extent, the translation of the text. Rala and I will be “tele-meeting” this week to divide up the annotation work.

That’s all for now!

I believe, like Pertusi (Leonzio Pilato tra Petrarca e Boccaccio, 109), that Boccaccio got most all of IV.lit.110 from the Chronicon of Eusebius. On p. 66a of Helm’s ed of the Chronicon, the name that Boccaccio cites as Aristotene (Aristothenes) actually corresponds to Eratosthenes. Boccaccio makes a mistake of transcription here. On line 17, the affirmation that Homer flourished a hundred years after the fall of Troy is attributed to Eratosthenes (a word whose only variants, Helm records, are “Eratostenes” and “Erathosthenes.”). Padoan, in fact, reaches a similar conclusion in his notes. How should we encode this?

Last week I attended the Summit on Digital Tools for the Humanities at the University of Virginia. It was a great time, and in many ways a rather unusual conference. For one, this humanist event was in part sponsored by the National Science Foundation (!!), thanks to the tireless work of Anita Jones and others. Most notably, it was more of a workshop than a conference environment. We were specifically asked to talk not so much about our own projects as about the state of the field and where we, as tool builders, would like to take it. I knew about half the people there from past contact, and was glad to meet so many new (to me) scholars all of whom had interesting things to say.

On Wednesday evening there was a welcoming reception and a keynote by Brian Cantwell Smith, a well-spoken computer scientist at the University of Toronto. Here are some fragmentary notes from his talk.

Digital is a trendy word, he said, second only to like on college campuses.

Descartes was a smart guy. He separated the work, or process, of understanding the world from the world thereby understood.

Around the turn of the 20th century, we discovered that we could fuse meaning and mechanism. An example of this would be us. This idea eventually gave birth to computers.

Computers aren’t anything special, and computer scientists aren’t studying anything special. Or maybe anything in particular. This is liberating: instead of a restricted domain, they have a sort of monopoly on the universe.

We (computing humanists) shouldn’t be party to propagating the dualism between the ostensible “abstract” and the concrete. A server going down loses not the representation of mail, but actual mail. (That is, as I understand it: email, web pages and other electronic artifacts have become so ubiquitous, and so central to everyday life, that we can no longer think of them as fakes, substitutes for physical things. They are just as real.)

Descartes said that we should have clear and distinct ideas. But this isn’t the way the world actually works.

Maybe the tools we build are digital at the level of the bits, but what matters about them is humanistic.

Computers are a historical moment (a long one, which started in the mid-1800s and is still going) in which we are getting past Descartes.

Matter is both a noun and a verb. Material comes from matter.

Computing is allowing us to get past the temporary, 300-year divorce between matter-noun and matter-verb.

Our commitment to what it means to be human shouldn’t be ideological (”if it’s human, it’s good”).

People can be special as in worthy of study and careful consideration, not special as in this is where inquiry stops because there’s nothing more to say.

…..

On Thursday and Friday, we talked. And talked, and talked. Instead of trying to edit my aphoristic scribblings to recap all of what we talked about, I’ll point you to Geoffrey Rockwell’s (and, in small part, my) comprehensive summit notes on the TADA wiki. (TADA is the Text Analysis Developers Alliance whose motto is, endearingly, Real Humanists Make Tools.)

The main thing I came away with is this. There are many people currently building digital tools for the humanities. The tools themselves may be differently specialized, but there are many similarities in what motivates us and what we hope the near future will bring. On our way out of the summit many of us put our names down on one or more of the six (?) major topics discussed at the summit. I hope the conversations, and collaboration, will continue.

My question here to everyone is about encoding stars and constellations as “places.”

Although I know that we have to chose one or the other (at least for now), I personally think that Jupiter, Venus, Taurus etc. would be better listed as “people” insofar as they are active influences rather than possible destinations. What’s the opinion out there?

M

As a follow-up on the encoding of personalities of deities, I’d like to add “Giove Liberatore.”

If anyone has any ideas about what to do for “Campi Ciri” or the “Rostri Aguglia,” I’d be very grateful to know about it! All bibliographical suggestions welcome.

Find it here. Please feel free to start working on it, per instructions contained on the page linked above as well as in the appropriate forum.

Just to let you know, the inaugural issue of the Digital Medievalist is up, featuring our article: ‘Towards the electronic Esposizioni: the challenges of the online commentary’. A full list of the other articles is here (I’m having my usual browser troubles, so this is via the MEDTEXT-L archive). Enjoy!

Vika’s detailed critique of Hillesund’s article deserves praise and a comment or two. I thought the article was interesting precisely for its shortcomings (or what Vika sees as its shortcomings). However, I do not want to turn this in a debate about the merits or demerits of Hillesund, but I see it as an opportunity for us to clarify a few (important) aspects of our own work.

Since in her comment to my original posting Vika asks for a clarification of my short and intentionally “open” comment appended to the two quotes I originally selected from the article, I’ll go back to that in this new posting. If you are interested in debating Hillesund’s positions, you should probably post a comment to Vika’s posting (ah, the intricate, branching nature of blog dialogues!). As far as I’m concerned, I’ll keep this new post as short as possible.

I’ll rephrase my original comment as a question: Is our work in the VHL to be seen as a contribution to the “workflow” of a “digital cycle”? Or is it simply a translation and re-production onto a digital platform of reading practices, born and developed within a non-digital cycle (dominated by printing)? My conviction (or wishful thinking) clearly is that we are indeed contributing to the emancipation of our intellectual “workflow” from the constraints inherited from the printing age, with its highly but “transparently” codified “objects” and the institutionalized ways we “exchange” them. At the same time, we are trying to preserve and adapt to the digital cycle practices and methodologies that were developed in trading with written/printed texts.

True, as Vika says, “any text (in his semiotic sense) that is represented digitally is “written” in ones and zeroes.” This is the fundamental “ontology” of digital “text” (any digital text). However, I am not interested in discussing a general theory of digital text, but rather the specific types of written/printed text we are working on. Or better, what I’m truly interested in is the possible “expansion” or “extension” of those texts from their original representation in written/printed form (inherited from our technological past) into new forms, specific to the “digital cycle,” our technological present and future. This, I believe, is “where our research emphasis lies.”

In other words, I do not think we are using digital tools in order to produce better, or different, printed editions of our texts (though printed editions may be one outcome of some of our projects, as is the case of Pico’s Oratio). Yet, is this clear to all of us?

The “expansion” or “extension” of existing texts directly affects (and in turn is generated by) the way we read them, where “reading” is conceived as the “last” (and “first”) stage in the “workflow” of any textual cycle (digital or non-digital, aimed at printing or not). The question is: how does reading and its possibilities change in the digital cycle? How and how much are we allowed to expand or extend our reading of texts that we inherit from our technological past?

As the acronym says, XML is conceived as a tool for a flexible extension and expansion, potentially encompassing all kinds of “texts,” including new genres and typologies. If I understand correctly, Hillesund however questions the capability of XML to provide a flexible enough framework for accomodating new genres and forms, those “new innovative digital text genres” that he leaves (in this article) unspecified.

In short, my question is: what kind of “expansion” do WE have in mind when WE approach the (semantic) encoding of texts such as the Esposizioni, or Pico’s Conclusiones - texts “written” or “composed” as tools for oral delivery and discussion? What kind of “reading” do we perform as we encode them? For example, are semantic encoding and collaborative annotating (made possibile by encoding etc.) to be kept separate in the intellectual “workflow” made possible by the digital cycle? Or should they tend to eventually become one?

These are some of the immediate questions that Hillesund’s article suggested to me.

I noticed in the two-vol Mondadori ed. of the Esp. that the paragraph number for 1.all.93 is missing. If anyone has the red Opere vol, could s/he check to see where it is?

I started this as a comment on Massimo’s post about “Digital Text Cycles: From Medieval Manuscripts to Modern Markup,” an article by Terje Hillesund recently published in the Journal of Digital Information. The comment got a bit long, though, so here it is in a new post.

Interesting article. It took me several hours to read, partly because some of its claims seemed inaccurate and required some poking around the web.

Some parts are both interesting and right on. I like his definitions a lot. For example, Hillesund distinguishes between electronic publishing (”a digitization process mainly based on principles of print”) and digital publishing (”an effort to use basic digital principles in a more flexible way of producing and distributing verbal texts”). He doesn’t go on to clearly detail said digital principles, which is a shame; but the distinction is a good thought exercise.

The phases of a text cycle (writing, producing, storing, representing, distributing and reading) are this close to a Proppian completeness. For the purposes of the article, this set does nicely; but I’m compelled to substitute composing for writing and put a semiotic disclaimer on reading, and then Hillesund’s system becomes applicable to all the semantic meanings of “text”. (Or does it? Readers?)

Section 5.4, par. 3 contains an off-handed remark that goes nowhere and yet is loaded: “…the Internet and the Web make up the distribution infrastructure of a global digital text cycle.” Where are the boundaries of this text? And what’s included in it? I wish Hillesund had pursued this further; doing so would highlight a discord between this and other things he says about what a text, and a text cycle, is. Perhaps next paper.

In the very next section, a poignant phrase: “Reading is the final goal of writing and constitutes the essence of all text cycles.” Again, if reading is to be understood as the active reception of information contained in a text, this could be applicable to any text (painting, piece of music, poem, etc).

Now I will focus on the negatives. The length of exposition they merit is definitely a sign that the article is useful in generating discussion.

- The core subjects would be much more interesting if Hillesund had not limited himself to “texts produced and represented in written forms.” For that matter, he fails to acknowledge that any text (in his semiotic sense) that is represented digitally is “written” in ones and zeroes.

- He says that texts have cycles, but then separates the phases of a text’s life into digital and non-digital cycles. This seems to defeat his stated aim of studying cycles of texts, in that it gives him leave to chop up any one text’s evolution into convenient chunks and ignore the transitions between them, and the influences they may (or may not) have on each other.

Sub-point: in section 6.2, he says: “A text is part of a cycle and in different text cycles all phases are related and interconnected.” Now that’s more like it!, except that it contradicts the above. It seems an inconsistent use of terminology, a relatively minor point but useful to keep in mind when reading the article.

- Discussing oral communication, H. says that “the participants… must be in the same place at the same time to communicate. To be passed on to others the oral message must be stored inside the body of the messenger, in his memory or brain.” Both of these conditions are incorrect: think, in the first instance, of a voice mail message, and in the second, a music or spoken-word recording.

- When discussing Gutenberg, Hillesund falls into the usual trap (which did not become apparent to me until I looked into it today). Gutenberg was not the first person to invent either the printing press or movable type. The Chinese predated him by several centuries. He did bring it to Europe; but since we’re talking about text cycles here, why isn’t there any consideration of Chinese texts between the 11^th and the 15^th centuries?

- There is no discussion at all about born-digital texts (written or not) and their cycles.

- Hillesund discusses XML without once referring to its utility as a semantic encoding tool, nor good examples of semantic encoding. He writes: “When XML experts analyze existing (printed) text genres, they extract structures consisting of element types like <title>, <paragraph> and <table>. Well, yes; but this isn’t where we stop, nor really where our research emphasis lies. “The aim of XML is to overcome the shortcomings of HTML and to extend the possibilities of the Internet,” claims the article. It is my understanding that XML aims for neither of those things. It exists quite independently of any network, and is not concerned with the same problems as HTML.

- Finally, Hillesund makes predictions about “new innovative digital text genres” without considering any of a number of examples already in existence.

I have more, but these are the main points that lent an air of suspicion to my reading. Comments?