Digital Humanities 2010

Live blogging from the HGIS panel on Saturday morning. The first paper in the panel is Natasha Smith on 'Unfolding History with the Help of GIS Technology'. This is from the library point of view so is based upon the experience of serving researchers and users at a university library. The users of the library at UNC began more and more asking for digitized maps. The speaker is starting with a lot of detail about how they digitized maps - 'purchased equipment'. The speaker thinks that cultural institutions are still struggling with how to make the digitized maps useful for specific types of users. One of the things they did with digital resources that couldn't be done with the physical objects was to knit individual map sheets together into larger digital images - specifically using Sanborn Fire Insurance maps to reconstruct historical cities in North Carolina. There are number of interesting looking projects in this talk, many of which can be found at http://docsouth.unc.edu/browse/collections.html. A related project called 'Main Street, Carolina' has also created a toolkit with an API so organizations and scholars can create their own pages using these maps.

Next up is Ian Gregory on 'GIS, Texts and Images: New Approaches to landscape appreciation in the Lake District'. He starts with a quick history of how HGIS developed, but the approach in this paper is about applying GIS to qualitative types of research, especially texts. When you start using GIS to analyse literary texts you have to take different approaches, one he proposes is Franco Moretti's concept of 'distant readings' from his 2005 book Graphs, Maps and Trees: Abstract Models for a Literary History. The initial project which he got involved in was to do with creating a literary GIS map of the Lake District looking at Thomas Gray and Coleridge. The project was about looking at what mapping on GIS had to offer for literary analysis, and Gregory explains the process of GISing the material. The research shows that although the two writers are 'lake district writers' their place names have very little overlap, so they're talking about different places. The project then mapped the altitudes of the places where the authors visited, and also looked at population densities. One aspect of this work is that it abstracts the texts, so they were keen to try to use the analysis to enrich the text - and so they created a Google Earth mashup which links the text to contemporary maps and aerial photography, so the user can see the place names mapped along side the text. They also did some work incorporating photos from Flickr, in order to think about how modern tourists experience landscapes. They did a lot of varied and interesting work in the attempt to think about how you use GIS qualitatively, such as working with corpus linguistics techniques for automated extraction of place names etc. He argues that this pilot work shows that we can do fruitful analytical work on texts with GIS, do some some automated geo-referencing, and all of this will allow researchers to integrate a vast amount of disparate data sources.

The final talk is on Elton Barker and Leif Isaksen on 'Mapping the World of an Ancient Greek Historian: The HESTIA project'. This project is about interpreting Heroditus's world through the use of GIS and the digital text, so they have looked at approaches to place in Heroditus's writing. Their research tries to put the narrative sense back into the mapping of Heroditus's world. Some technical stuff in the middle here describing the architecture. A lot of this is about taking Heroditus and running it through GIS techniques, some well established but also some innovative work to try to map Heroditus's world, moving back and forth between mapping and the narrative in the way Gregory talked about in the last paper. The speaker finishes off with the intellectual payoff which has two facets. The first is to do some qualitative analysis about the 'textualization of space', and the second is creating network maps which give you a sense of the real life topography of the narrative.

It’s 11.00 o’clock at DH2010 London and I am attending the Professional Reflection panel. Topic of the moment is: “Pointless Babble or Enabled Backchannel: Conference Use of Twitter by Digital Humanists”.

As a Twitter user, probably more for personal use than as a Digital Humanist, however, I am quite interested to see what these folks have got say about it.

The speaker begins by saying: “Twitter is absolutely everywhere!”. And this is so true. But the question is: Is it useful? Is it distracting? Irritating, even?

Twitter has increasingly being used at conferences, in fact people are tweeting right now, while I blog, and tweet, I confess!

For those of you who don’t know Twitter much, these are its main features:
* Constrained to 140 characters. (and this is a challenge, I add! It forces, and challenges you to be direct and clear!)
* RT = to re-tweet. Some ask: what’s the point? I personally use it to express my agreement with what that particular tweet has to say. Is that right? Is this how you use this feature? It is interesting to see from the results from this DH conference tweets that most of the activity is actually original, not so much re-tweeting.
* links = external links
* @ = indication of conversation

So Twitter is everywhere and lots of people are using it. However, this year DH statistics show that only a quite small amount of people are tweeting, it’s the usual thing: “90% lurk, 1% really, actively participate!”. Are you still sceptical? Is that why you are not tweeting?

So, why is Twitter used: for note-taking, info sharing, to make the conference more of a discussion, to comment on presentations, to be part of a community and establish online presence.

So, is it a good thing? Why don’t you try it and see...

I heard quite a bit of focused attention to sustainability. Two lectures pointed to sustainability's heaven: Jane Hunter with OpenAnnotation, and Peter Robinson and Federico Meschini with Works, Documents, Texts and Related Resources for Everyone.
Read more on my blog at DANS.

Today is the last day of DH2010 and this morning I am in the Geography session.
It’s Saturday and very hot but the room is still packed!
Ian Gregory is talking about “GIS, Texts and Images: New approaches to landscape appreciation in the Lake District”. I am very interested in this topic since at some point I would like to *bring to life* the research taken from my PhD on Latin verse inscriptions from Britannia by geo-tagging all the places where the inscriptions come from, were re-found at some point and where they are at the moment, to create a series of maps that draw and show the history of their discovery.

Ian is showing us the results of a pilot project funded by the British Academy: “Literary Mapping of the Lakes”. At the basis of the project is the question: Can we create a GIS of text? In order to do this, two texts have been selected for the Lake District area, namely two tours of it: 1) by Thomas Gray, 1769 (9,000 words) and 2) by St. Coleridge, 1802 (10,000 words). The texts are typed by hand and the place names tagged within the text. Then, all places mentioned in the text are linked to a gazetteer (all places on 1:50,000 maps).

The place names are extracted from the text. Now, can GIS be used to bring the two things together? To do this, a close reading with Google Earth is created where the text sits below the map. All place names are highlighted in the text and plotted on the map - so, while reading the map, you can see where the place is and by clicking on the place name in the map, a little window with some info about it pops up.
It’s also possible to attach pictures of the Lake District incorporated from Flickr, which I think is quite neat!

Providing good infrastructure (in the broadest possible sense) for research and teaching is very close to my heart, which is why one of the most interesting sessions for me at DH2010 had to be Understanding the 'Capacity' of the Digital Humanities: The Canadian Experience, Generalised. Unfortunately, I could not see all of the session, but there are a few key points that I want to highlight (particular thanks to Geoffrey Rockwell).

Three points in particular where made with regards to how the value of Digital Humanities could best be demonstrated. Perhaps the most important one is that DH can be described as an enabling field, in the sense that it allows other fields to do research that would otherwise not have been possible. The second one was that DH can help academics to dramatically increase their outreach, especially beyond academia (in particular in relation to crowdsourcing, social media etc. As Geoffrey Rockwell put it: We help the Humanities reach a broader public.
Last, but not least DH helps to prepare students and young researchers for the new challenges they are going to face in their careers, for instance in the media content industry, but also in many other fields.

It was also Geoffrey who formulated a list of requirements for the basic infrastructure that has to be made available by any university that is serious about supporting (digital) research:

• Social lab for projects and meetings
• Digitisation facilities and specialised hardware
• Support for utilities (lists, bogs, wikis...)
• Virtual machines for projects
• Advising and long term technical support

This morning I am at the Annotation Mark-up session.
Jane Hunter talks about “The Open Annotation Collaboration” project

The project is funded by a W. Mellon Foundation Grant, it started in Mar 2009 and is meant to end in Set 2010.

At the basis of the presentation is the concept that there is a plethora of annotation tools (desktop vs web-based, private vs public, synchronous vs. asynchronous, txt vs image/audio/video, discipline specific vs generic, stored locally or on remote server, tags vs full/free text). The difficulty, however, is that there is no consistency, interoperability and standards.

Therefore, the main objective of the project is to develop an OAC data model to enable the sharing ad interoperability of scholarly annotation across annotation clients, collections, media types, applications and architecture.

More information to be found here: www.openannotation.org

Here we go after lunch on Thursday with a stylometry panel. First up is Patrick Juola, on 'Authorship Discontinuities of El Ingenioso Hidalgo don Quijote de la Mancha' looking at the question of who wrote Don Quijote, by using digital tools to evaluate authorship discontinuities. These discontinuities, if there are any, are evidence either for a style shift or a change of author. They start by regularizing the document. One of the processes this involves is 'stripping out irrelevant or confusing style issues'. The 'mixture-of-experts' results are a way of dealing with anamolies. The speaker gives some detail about the metholodology of analysis and the actual operations that were run on the text. Ultimately the conclusions were that Volume I was written by Cervantes, for Volume II the conclusion was that Cervantes didn't write the bulk of this volume. There is no definitive author but the work does show that the text in the volumes are not consistent. It is possible that he changed style in the middle of Volume II but it is not likely that this is where a style change would happen.

Paper number two in this panel is Finding Stories in the Archive through Paragraph Alignment given by Maria Esteva. She is an archivist and also interested in exploring the text documents in an archive relfect the works of the archivists. This is a highly statistical analysis that I'm afraid isn't really my forte. It is about information management practices within archives in which she has worked, but beyond that I don't think it's worth it for me to try to summarize her talk, as there's no way I'll do it justice. It's important to know when to admit defeat.

The stylometrics talks that I have been to this year and last year at DH2009 in Maryland have been very specific and highly embedded within their discipline. Is this because they are not used to talking to people who aren't sylometrists or because for this sub-discipline DH is actually a place where they come to talk to others in their field.

Finally, we have Malgorzata Sokó on 'WW1 and WW2 on a Specialist E-forum: Applying Corpus Tools to the Study of Evaluative Language'. She starts with a definition of evaluative language which is a term for the expression of the speaker's or writer's attitude toward whatever he or she is talking about. Her linguistic analysis had the aim of investigating how contemporary Poles' attitudes to the two world wars are expressed in electronic discourse, on the example of a specialist e-forum. Her talk is very nuts and bolts, focussing on the methodology and the data. Her conclusions are more about how linguistic concerns - the role of evaluative language, and the effectiveness of the corpus tool she's using - are reflected in the objects of her study, not about the attitudes toward the world wars among young people in Poland.

Getting ready for the visual art/visualization panel now, and got here nice and early to nab one of the very few outlets in the room (my first live blog of the day - on Humanities Labs - was thwarted by the lack of power). Hopefully I won't get in trouble for endangering the health and safety of DHers. The theme this is organized around is to try to critique the use of digital tools and methods in art history and history, so the focus is on tools and resources rather than research questions.

First up is Nuria Rodriquez Ortega speaking on 'Digital Resources for Art-Historical Research: A Critical Approach'. Her main objective is to reflect critically about aspects of digital resources currently found on the web. Her focus is on freely available resources which are extensively used. She seems to be arguing that the structuring of information on websites implies certain art-historical narratives. Databases and catalogues of art are among the most used resources by art historians. Databasing of information about art objects makes them into discrete objects which are out of the context which are the key to art history. 'The art history that we will be bequething to the future will be a collection'. So, according to Rodriquez Ortega the images existing in museums' online collections are taken out of context by the nature of how information about them is transmitted on the web. Her next point is that the use of digital surrogates leads researchers to privilege the detail over the whole image. My question would be: is she blaming the media for bad scholarship? Just because it is possible to do the things she is talking about doesn't mean that a good scholar necessarily would do this. She goes on to criticise 3D reconstructions for the implicit interpretations which are made by the scholar. Here again it seems to me she is claiming that the digital interpretation determines the quality of scholarship.

Next up is Alejandro Toledo on Contexts, Narratives, and Interactive Visual Analysis of Names in the Hyohanki Diary. The problems he poses to begin with in the field of visual analytics include: the curse of dimensionality, motivation and the goal of the research. He then goes on to show in the field of visual analytics it is possible to have multiple perspectives on the same information. Toledo is speaking at fairly high level of about problems that are very specific to his field. It would be very helpful to understand the point of the research - what question are they trying to ask/answer? The project is looking at names in an historical japanese diary. His critique is that making different assumptions leads to different visualizations, and therefore it is difficult to draw clear conclusions. It seems to me that multiple perspectives on the same information is not a problem that is only specific to visualization, but rather something that all researchers have to tackle - how do my assumptions lead me to specfic conclusions.

The final paper of the panel will be Paul Conway on 'Modes of Seeing: Case Studies on the Use of Digitized Photographic Archives'. His talk in on the notion of representation, which he sees as a form of communication. The digital representation implies two kinds of relationships - that between the digitizer and the user, and that between the surrogate and the original. Digital Humanities is (still) very text centric according to Paul Conway. His research project is called Beyond Image Retrieval, so he is concerned with how users (both withing and outside the academy) use the materials once they have it. His use case studies are based on large scale collections at the Library of Congress. This is very much information science focussed. He is drawing a theory of research processes from the way people talk about what they do when they research digital objects in these collections. His argument is that researchers have their own narrative with which they 'fight back' against the assumptions of the digitizer. There are three 'modes of seeing' in how his subjects do their research: 'discovering', 'storytelling', and 'landscaping'. It is at the nexus of these three things that the creative work of digital humanities lies. The researcher should be 1. discerning detail from digitized negatives, 2. building a narrative from both within and beyond the frame, and 3. putting the image in socio-political contexts over time. With digital resources the surrogate is becoming the archive, so we need to grapple with questions about what happens when images are digitized.

This last point can be extended to any digitization and can never be said too often.

A programme of events will be held in the Anatomy Theatre & Museum (ATM) on the 6th floor of the King's Building on the Strand Campus during DH2010 to showcase digital scholarship across the disciplines at King’s. Come and join us for one of the sessions there during the breaks - refreshments are available! Have a look at the programme.

Tobias Blanke and Mike Priddy present during the Digital Scholarship sessions at DH2010

This morning, we had a very interesting session on Virtual Research Environments in the Humanities. Several themes emerged from the presentations and discussions, some of which I will quickly summarise here:

• The role of libraries in the development and management of VREs was referred to several times. It was even argued that VREs could be seen as the second generation digital libraries.
• The point that addressing research questions was absolutely crucial for VREs was brought up several times.
• Focussing virtual environments on addressing a particular research question ad not overloading them with other functionality was seen as a promising route to take. Otherwise there would be a risk to scare away researchers or confuse them. VREs need to deliver on what researchers want, otherwise they will not put up with being part of experimental development projects.
• When we think about VREs as infrastructure we should be aware that infrastructure only really works when we do not notice it. Many VRE projects are still at a research and trial-and-error stage. It was suggested to look into research on how infrastructure was historically build up and then sustained (road networks as an example).
• The importance of "text" for many humanists was mentioned; Geoffrey Rockwell argued that humanists don't care about tools as such, they rather see them as lenses through which to look at text.
• Tools need to follow the rhythm of the research process; whereas teaching staff use VLEs regularly, VREs appear to be only used in bursts, short periods of intense activity - in-between those researchers can forget how to use them, so VREs need to be simple to use.
• Everyone emphasised the issue that VREs need to be seen and developed as social projects, through involving the community.

If you are interested in these and related issues, have a look at the recently published JISC VRE Landscape Study.

This afternoon I am attending a presentation on the project *CHARM: Centre for the History and Analysis of Recorded Music*

CHARM is a 5 year project funded by the AHRC that ended just last year.

We are first introduced to the website: http://www.charm.rhul.ac.uk/index.html

The presentation focusses on the sound files.

A detailed record of digital tools and methods used for, and throughout the project can be found here: http://www.arts-humanities.net/projects/centre_history_analy...

We are now listening to very valuable and no longer available records. It’s quite charming!

Our dedicated team of interviewers have just finished the first round of short interviews with conference delegates. Delegates speak about their background and interest in digital humanities. We will also interview presenters (both papers and posters).

The first audio recording just went online, more will be added as the conference goes along. We have an RSS feed for the latest audio.

Good morning, everyone. As you can see from the photos in our DH2010 image gallery, the registration for the conference has been set up and the first delegates are arriving. THATcamp still goes on (we have also uploaded a few photos) and everyone is getting ready to go here at King's. Exciting days ahead, no doubt.