Yesterday evening, the Digital Humanities conference kicked off with a keynote speech by Lev Manovich. Lev spoke about the analysis and visualisation of large amounts of cultural data: cultural analytics. His starting point was the observation that we are experiencing an explosive growth of cultural content on the web, partly driven by digitisation programmes of museums and libraries, but also by recent trends in social media. Lev argued that, until recently, we had to chose between a shallow analysis of large numbers of objects or a deep analysis of a smaller number of cultural objects. With recent improved computational power and new analytical means it is, as Lev argued, now no longer necessary to chose between these approaches – we can do both, if we want.

Together with his colleague Jeremy Douglass, Lev demonstrated how various visualisation tools and methods could be used to that purposes. He also cautioned the audience that there was an important difference between analysing (just) cultural outputs or also the context in which they emerged. The latter was much harder to achieve, but is obviously of great importance, especially from a (digital) humanities perspective.

Because of time-constraints, Lev did not really have a chance to demonstrate in more detail what kind of answers to research questions visualisation could give, but he gave a great overview of tools and approaches to deal with cultural data (and culture, of corse) - from Hamlet to web comics and design.

With over 300 delegates, this year's is one of the best attended DH conferences. Several of the speakers yesterday, among them host Neil Fraistat and Harold Short, mentioned that as one of many examples of the 'tremendous vitality (Neil) of digital humanities. Julia Flanders called the field very 'vibrant' and Neil even remarked: 'The Digital Humanities - this is our time.' Harold emphasised that interdisciplinary collaboration was a key to this and that the Digital Humanities approach would open up many new opportunities here.

What do you think about this? Are we mainly profiting from the fact that the world gets more digital every day (and DH more visible as a result of this)? Will digital become so normal that, as everyone does it, there is no more specific Digital Humanities? Or are we indeed a vibrant community at the forefront of new ways of researching, answering new questions that are relevant for our time?

It will be interesting to see what the conference has to say about it.

Digital preservation is a key issue, not only for Digital Humanities. The first session this morning, Preserving Virtual Worlds: Models & Community, deals with preservation issues as they relate to computer games. The presentations gave interesting case studies in the difficulties of preserving software - intellectual property rights; companies (and their documentation) going bust; modifications by the user community; preserving the code (preservation of developers) or the game/work flow (preservation for users); use of proprietary software etc. Preserving games is also interesting as the gaming community takes this very seriously, allowing collaboration between preservation specialists/archivists and the public.

Of course you may wonder why bother with preserving games at all? No matter if you like them or not, computer games have become an important part of our culture and preserving games is key to studying it. If we do not take this task seriously, as one of the presenters remarked, we risk losing an important part of our cultural heritage.

I'm also currently in the panel Torsten just commented on called Preserving Virtual World: Models & Community (maybe we should coordinate our coverage) which so far is primarily about preservation (the clue is in the title) of some very old computer games - Adventure (the first adventure game), Doom, and various titles popular with gaming enthusiasts which were released on platforms like the Commodore 64 and the Atari 2600. The panel members are mostly Information Science professionals, so are interested in the ins and outs of preservation at a highly detailed level. There has been some lip service played to the cultural relevance of these games, but the interest seems to be primarily at the level of metadata and FRBR reference models. The issues of preservation are, of course, important, and I'm not one to argue that computer games are any less important as products of culture than any other types of 'high' or 'low' culture. However, I would really love to hear more about what that value is. As this is a humanities conference, I want to know what the research questions, or potential research questions are which drive the work forward. Further presentations are discussing the relationships within the gaming community - between professional developers, fans and gamers, and the preservation experts. However, here again the emphasis is upon the issues of preservation from the standpoint of archiving and library science.

Next session: Supporting the Digital Humanities: putting the jigsaw together (Martin Wynne, Steven Krauwer, Seth Denbo, Chad Kainz, Neil Fraistat). This panel addressed the question of the infrastructure needed to support the digital humanities and arts. Presenters came from the following projects that all focus on infrastructure: Clarin, DARIAH, Project Bamboo and Centernet - the first two funded by the European Commission, the third by Mellon and Centernet as an initiative that is supported solely by its members.

The following key issues of infrastructure provision were identified:

• technology has to map to the way researchers work, not the other way around
• infrastructure needs to be planned with sustainability in mind
• infrastructure development is not just a technological task
• it is crucial to build on what is already there, to leverage existing infrastructure
• infrastructure development must be informed by lessons learned from the community
• needs to be broad, inclusive and international
• resource providers have to overcome insularity and silo-nature
• user engagement means to involve the community from the start

In his presentation on DARIAH, Seth mentioned a project currently undertaken by the Digital Curation Unit at the Athena Research Centre in Athens that tries to develop process models for arts and humanities digital research. These models will not just take into account the digital resources, tools and methods, but the interaction between the digital and the non-digital or more traditional processes researchers employ. This is obviously something I am very interested in as it links to work undertaken by CeRch (previously AHDS) to map methods used in digital research and resource creation: http://www.arts-humanities.net/ictguides/methods

I am now in a panel on computational stylistic analysis. The first paper is about the researcher's analysis of a number of Poets' stylistic evolution - including Longfellow, Edgar Allen Poe and others. The researcher has traced syntax features through these poets' productive lifespan, and plotted the texts and developed mathematical formulas for the analysis of poetry. It is possible using these statistical techniques to make large generalizations about the trends such as the poets' use of subordinate clauses. Not being a poety scholar myself I'm the person to comment on the meaning of this all, but the paper engendered some very lively discussion.

The next paper in this session is on computational stylistics of Ibsen plays, and had some interesting stuff about tools that the project is using to do the work of analysis. This two elicited a wide range of questions about the methodology and research decisions taken in the analysis. There seems to be great interest in such issues as how spaces are treated in character counts. By now it will be clear to anyone reading my comments that I'm completely out of my depth when it comes to computational stylistics. I'd welcome some comments from those more in the know!

Listening to Julia Flanders's paper on 'Dissent and Collaboration', in which she discusses how the customization, limitation, and extension of collaborative encoding standards (using the example of TEI) formalizes essential academic dissent. TEI provides an excellent mechanism for this by expressing such customization (and therefore dissent) openly and clearly in the ODD language. Will comment on final conclusions and discussion later...