Aligning digital (computers) with humanities

Editor's Notes

• #2: I put a sub title in because I wanted to make a comparison between cultural heritage computers systems and the second most advanced computer in the galaxy Deep Thought, the first computer to be given the ultimate question of… life universe and everything – which is sort of what we Museums deal with.In Douglas Adam’s book the hitchhikers guide to the galaxy, a “pan dimensional, hyper-intelligent species of beings” (i.e. computer nerds) create “Deep Thought” the greatest computer of all time. Thismassiveartificially intelligent machine is the given the job of answering the ultimate question. The ultimate research question! After millions of years Deep Thought concludes that the question was not well defined or understood and that in order to answer properly you could need a vastly different approach, including the modelling and simulation of data. Which is what the digital humanities was originally about. This is the point we have reached with the digital humanities. We have hit a wall in which the level and sophistication of our questions and answers have been limited by the way in which we understand and interact with computers. The adage ‘garbage in garbage’ out has a particular resonance for the humanities because data that properly reflects what we do is complicated and dependent upon context. If we don’t supply computers with this context then we cannot expect to get the answers to any really big questions.
• #3: Knowledge perhaps also not the right word.Humanities still finds it very difficult to interact with computers.
• #4: 42 Rubbish in and Rubbish out!!!In Shakespeare's Romeo and Juliet, Friar Laurence gives Juliet a potion that allows for her to be in a death-like coma for "two and forty hours".A single Big Mac contains 42 per cent of the recommended daily intake of salt.Elvis Presley died at the age of 42.Cricket has 42 laws.Off by 1What actually then happens is that Deep Thought realizes that what is really required is a new type of computer one that can simulate and model and collect the contextual data necessary to answer the ultimate question. The new computer is the Earth, us. It is the earth that creates the necessary conditions to examine human life in the context of space and time. It needs to use real world data.
• #5: Unity of Understanding These are both Wunderkammers or Cabinet’s of Curiosity or early forms of museum.On the left is Ole Worm’s Wormianum. Worm was a Danish philosopher and collector during the 17th century.His Wunderkammer is a representation of nature containing natural and artificial objects (that are nevertheless reflections of nature).There is an underlying knowledge system or ontology of the day that allows resemblances (for example patterns) and meaning (common concepts) to be identified between these objects.In this way a unity or harmony is found across the chaos of nature. The Agsburg cabinet (displayed at the Getty Institute) on the right, although different in appearance, works on the same principles.These two cabinets are part of the same network of meanings. They are compatible
• #6: The Enlightenment During the Enlightenment more sophisticated questions were asked about the development of humankind. The collections, rather than being representative are far more comprehensive.However, questions about the world still involve connections between natural and artificial objects. As you can see from the picture of the Enlightenment gallery at the British Museum.The greater sophistication and breadth of the questions is matched by more sophisticated and scientific classification systems.This in turn, leads to the splitting up these large collections into specialist museums and libraries.The expansion of knowledge is achieved at the expense of the unity found in the world of the Wunderkammer.
• #7: DisunityAn example of this split is demonstrated in parliamentary proceedings in the 19th century.The reason why the British Museum’s collection was so varied is soon forgotten by its own curators.Oldfield quotePanizzi quote…when challenged by the National Gallery that classical art should be part of an art gallery
• #8: YOU CAN APPLY THIS CORPUS TO THE BRITISH MUSEUMTHIS SHOWS AT THE TOP THE ENTIRE CORPUS OF LITERATURE AND BELOW JUST FICTION. YOU CAN SEE WHEN THE OTHER NATIONALS WERE CREATED AND THE GROWTH OF THE BRITISH MUSEUMTHROUGH THE GOLDERN AGE OF EXPEDITIION AND THE EXPANSION OF THE BRITISH MUSEUM ITSELF THROUGH 1800IN BOTH CORPORA A DECLINE STARTES AROUND THE 1970S. THIS MAY CONTAIN IMPORTAND INFORMATION ABOUT STRATEGY.
• #10: But while we have understood and documented are collections better applying more sophisticated and specialist ways of classifying. We have also further removed our collections from the real world. Yes we do a lot of work behind our collections but essentially with apply a new context we re-contextual and in some ways market our collections in artificial ways. The British Museum’s collection being associated with the chocolate drinking Hans Sloane a darling of the Enlightenment. On our literature we use the phrase enlightenment museum a lot. But not everything about the Enlightenment is necessarily good or help in understanding the objects within the collection.
• #11: In fact over the years we have been desperate to take any and all context out of our data.Here is a quote from Dominique Delouis (who sadly died this year), but which reflects the general mindset of time. This at a time when the European commission was funding a project called Remore Access to Museum Archives.The point is that once you squeeze things into a box what you have left has less value as a resource because you have altered its meaning..
• #12: In 1995 there was a realization that these projects not only couldn’t deliver the type of benefits that technologists originally promised.But that the whole approach has been wrong. However, before this could be corrected – Web publishing took over and these issues were hidden under cool multimedia electronic pages.
• #13: McCartyMore vision about what computers could do before the Web than after it
• #15: MeaninglessThis record from the Europeana portal, is from one of the UCL museums and demonstrates are lack of interest in precision.The model uses generalisations that encourage misinterpretation even for the technologists that created and manage it.In this record, reflected by many others uploaded by the UK aggregator, the meaning conveyed is that people in the 26th dynasty of Ancient Egypt were able to produce digital images.The identifiers are often not really proper identifiers and the common denominator approach uses sweeping generalisations.Yet many millions of euro have been spent to produce these records and many in our sector defend it as part of a ‘greater good’ argument.
• #16: The Linked Data movement is a initiative to do something about the problems of using a electronic publishing model when exposing data.But publishing linked data in itself (a technical format) does not in itself create understandable information.Indeed many people using linked data create artificial ontologies that has very little context and therefore its utility is limited particularly in the humanities.
• #17: What you end up with are different perspectives of the world. All of which separately are somewhat artificial and aligned with a particular way of framing or seeing things and we need to be careful that we are not the victims of our own marketing.
• #18: In Collections Council of Australia have over the years produced some interesting work on the significance of collections.The first statement seems to be in conflict. But our memories of things change, we rewrite them and we attempt to fill in the inevitable gaps sometimes ethically and sometimes unethically. The interpretation of memory can be based on valid arguments supported by other evidence. Our subjective opinions about history can be important but this ethical analysis needs to be supported by context.
• #19: The Death of Fishing This is a statue that I grew up with in Lowestoft. Its not particular great but nevertheless stood in the main town square.It’s a testament to the fishing industry to which Lowestoft has been a part of since the middle ages up until very recently.I suspect is has very little significance for many people now and has been moved away to an Asda carpark.
• #20: Enthroned upon an ancient hill it rests;Calmly it lifts its time-worn head; and firstOf all Old England’s busy towns, whispersIts orisons, and greets the rising morn.
• #25: Imaging sitting with your bucket and spade on Lowestoft beach watching 200 ships do battle.
• #27: Starting to focus on the types of vessel used in fishing
• #41: Not just the usual vocabularies like materials, techniques etcBut attribution vocabularies that can be specialist for individual institutions
• #43: StandardsThe cultural heritage sector has hundreds of different data standards making interoperability almost impossible with the resources available.
• #45: EDM versus CRMYou cant do this using the Europeana data model or other standards that use common denominator models or are too general is their semantics.The CRM has been built from the ground up and through 15 years of examining different cultural heritage data models to ensure the right level of semantic description that allow datasets based on different specialisations, archaeology, conservation, antiquiti4s, natural history, and so on to work together. It is compatible with all other cultural heritage standards and as such should be the primary model for publishing structured data.
• #46: Too Specialist Here is an example of a CIDOC CRM Model It show the production of an object through a production event and described how this was carried out by a person or group.An Object, was produced by an event, which was carried out by a person or groupThe CIDOC CRM generalisation P14_carried_ by could be specialised by the Museums associations codes and these terms could be different in different museums.The CRM’s object oriented approach allows P14_CARRIED OUT BY to be specialised by individual museums. However, this approach tends to create a large amount of terminological specialisations.
• #47: Best of Both WorldsInstead the ResearchSpace project has created constructs that can be used by any museum or archive. Instead of specialising the CRM it provides a mechanism for plugging local vocabularies into it.It does this by allowing events to be typed by local vocabularies.This makes the process of constructing a CIDOC CRM mapping far simpler and provides a more stable platform for application development.
• #49: Co-referencingThis framework becomes crucially important for generating new knowledge.Consider this example in which the details of a particular person are unknown or uncertain. The information about this person may differ from organisation to organisation.However, by building up semantic profiles of objects through the framework of the CRM using concepts of production, subjects, objects types, periods and events it will be possible to infer that a person recorded by one museum is the same person recorded in another simply by virtue of the object profiles connected with them. Equivalences can be found even when a field is blank or where the data itself is wrong. The CRM will allow new knowledge to be located and created using the power of computing and through harmonised datasets from across the cultural heritage sector.