Don’t look at me, I didn’t make the word up, I’m just sharing the pain. But yes, I was at a conference, as you may recall, one of a series called Electronic information: the Visual Arts and beyond, or EVA, and this one entitled Digital Cultural Heritage – Essential for Tourism. And from this and the title of this post it may not shock you to learn that though the conference was in English, this was mainly for my benefit, as it seemed, because bar me and a couple of Italians everyone presenting could have stuck to German of some flavour and understood each other.1 And although this has little to do with the tenth century, as even I was presenting in my professional capacity, I know that there are museum-goers, heritage professionals and networking geeks reading this (at least one of each!), and besides it was interesting so I’m gonna blog it. There will later be a picture-heavy tourist post, but I’m in a different place from the images as I write so this will come first.
This conference was dominated by the Pattern Recognition and Image Processing group in the Technische Universität Wien, and consequently responded to rather different expectations than the humanities ones I usually attend. You can tell this because I have the proceedings in front of me; in theory, all papers were double-blind reviewed before acceptance and the proceedings were in the free bag they gave me on registration.2 (I’m very glad someone else was paying for this; it was more expensive than Leeds, but this sort of spending is why.) Actually, the proceedings are a lie, especially when it comes to the Coin Recognition Workshop of which my paper was part, which came together too late. Only one of the papers that was given is in them, and another with it was not in fact given. In how many ways this is like a charter, that records a public reference version of something that may not really have come out that way because the charter was written apart from it, I shall try and avoid saying again. Anyway, the Coins Recognition stuff is of interest to few and discussed elsewhere, so here I shall only say that the software is impressive, arguably not only does what it was supposed to do but could be the root of even more, and that my paper had little of this about it but since it will not be published, I see no harm in sticking it up for download anyway for anyone that might be interested.
Repeated themes of the technology and applications being developed or demonstrated here were, firstly, assembling recognition libaries of cultural content, either specialised as with coins or, in the most ambitious case, the whole of human archaeology maaan, so that you could find something in the ground and have a decent guess at what it is from the web. We seem to be a long way from that, but the technological pace towards it is on the runway, even if not yet taken off. There are two aspects to this, firstly the image recognition, in which we are having to do a lot of crunching still, and then the actually meaningfully saying what the image is of, which involves massive international multi-lingual thesauri. Weirdly, that bit is going faster, perhaps because words very rarely get turned round so that you have to recognise them at an angle.3
Another theme was digital technology for improving your tourist experience. Istanbul have a piece of kit under work which will set you up maps and routes on a PDA-alike to guide you through the rather confusing city streets and find you the monuments and period stuff that interests you particularly.4 Somewhere in the middle is similar tech for your PDA in Florence and Winchester, and at the smallest end is something very much like the e-Guide we’ve just replaced at my place of work. The one being described here had been trialled in a museum of energy in Recklinghausen in Germany and seems to be better than the old one we had and perhaps not as hot as the new one, but working on a much better model of locating the visitor.5
A tile of a new project to map the Byzantium of c. 800 C. E.
The appalling neologism of the title came from a paper by one Elena Bonini, who was applying a great deal of theory about knowledge uptake and learning processes to the question, basically, of how to interest people in museums. The theory predictably annoyed me, especially with words like that in it (“edutainment” should also be taken out and shot, which would be both educational and entertaining) but the paper, despite some weak points, was very interesting, and it won a prize for best student presentation out of the nine present quite deservedly. She was very much for interactive content and learning by doing, which seems fair enough. Her user surveys however suffered from a generation gap, in which those above forty or fifty don’t expect to learn like that and just want to be told something by an expert, whereas among the under-20s virtual rooms, puzzle games, conversational learning in text and so on scored much more highly. I thought that was what her figures showed, and that that was interesting and worth making something of, but she was more inclined to just brush the unpopularity of her methods among a big part of her userbase under the carpet. That was one of the weak points, but I was impressed with much else. All the same, the traditional panel with a label on it has some way to go before it need fear retirement.6
Another theme that had a session to itself was OCR work on manuscripts. Two of the three papers focussed on a single manuscript, a Slavonic missal which came from St Catherine’s on Mt Sinai, which has produced so many incredibly rare and ancient manuscripts and apparently keeps doing so.7 This battered book is the oldest manuscript known in Glagolitic script, the forerunner of Cyrillic that was actually invented by Cyril, or Constantine as he usually called himself, and Methodius, apostles of the Slavs. It was in use in Croatia until the nineteenth century but it isn’t very well understood in as early a form as this. Breaking down that script to reproducible characters has therefore not only been challenging, but extraordinarily valuable for researchers working on fragments elsewhere.8 The script is incredibly detailed, I don’t know how anyone thought it useful, but perhaps that was the point, it was sacred writing.
Partially textured 3D model of Xochicalco, derived semi-automatically from a stereopair of aerial images using CyberCity Modeler
The paper I was most impressed by however was the keynote, Dr Armin Grün talking about “Image-based 3D Modeling of Cultural and Natural Heritage Objects”.9 And when he says objects, he means everything from statues to entire cities. His group have flown a toy helicopter with a digital camera slung underneath it over deserted South American cities that predate the Incas, they have mapped and modelled huge pre-Aztec adobe complexes, and they have digitised the Nazca lines and Mount Everest, already! (The first two projects, along with some of the visuals that Dr Grün was using to illustrate their methods, are depicted and described online in this article of his.) But it’s not just recording they’re involved in, it’s reconstruction. You may have heard of the valley of Bamiyan in Afghanistan, which used to be a really important Buddhist holy site, with huge niches carved in the valley sides holding statues of Buddha. Then came the Taliban, and what age and decay hadn’t already achieved, fundamentalist explosives did. Dr Grün’s group, using satellite imagery (they also use balloons, aeroplanes, robot cars and occasionally human-held cameras) plotted the disappearance of the two largest Buddha statues, but a few years later they also got hold of some 1970 photographs taken by a travelling photogrammetrist who took them carefully enough that scanning them and working the team’s particular magic has allowed a digital reconstruction of the two Buddhas, from which actual replicas can be made. Some day when wars permit, there’ll be something at Bamiyan for pilgrims to see again. By then, he was forecasting, the group will not only taking their pictures with mobile phones, they’ll be doing the processing on them too, in the field, and coming home with 3D visualisations ready to put on your website after a day’s work. (A full list of their projects is here.) That was a good note to open on, and I shall use it as a good one to close on; this is at last moving beyond mere curiosity and showiness to something that genuinely does good in the world. Odd to think of Buddhists and IT meeting so obviously in a hall in Vienna, but it happened and I was there to see it and tell you about it.
Large Buddha at Bamiyan in 1963, from Wikipedia Turkey
1. Although one of the Austrians said he was used to Germans asking him to speak English instead of his native tongue so they could understand him…
2. The proceedings are Robert Sablatnig, James Hemsley, Paul Kammerer, Ernestine Zolda & Johann Stockinger (eds), Digital Cultural Heritage – Essential for Tourism. Proceedings of the 2nd EVA 2008 Vienna Conference, Vienna, August 25-28, 2008, books@ocg.at 238 (Vienna 2008).
3. Covered in Martin Kampel, “Computer Aided Analysis of Ancient Coins”, and Achille Felicetti & Hubert Mara, “Semantic Webs and Digital Islands: Semantic Web Technologies for the Future of Cultural Heritage Digital Libraries”, ibid. pp. 137-144 & 69-78 respectively.
4. Nadide Ebru Yazar & Can Binan, “Use of Digital Media for Cultural Tourism: Byzantine Monuments on the Web”, ibid. pp. 43-50.
5. The Florence and Winchester projects in Maurizio Megliola & Luca Barbieri, “Integrating Agent and Wireless Technologies for location-Based Services in Cultural Heritage”, and the museum one in Sebastian Gansemer, Uwe Großmann, Oliver Suttorp & Hanswalter Dobbelman, “Evaluating a location sensitive multimedia museum guide: Results from a field trial”, ibid. pp. 51-59 & 103-110 respectively.
6. Elena Bonini, “Building Virtual Cultural Heritage Environments: The Embodied Mind at the Core of the Learning Processes”, ibid. pp. 119-128.
7. For example, see Elias Avery Lowe, “An Unknown Latin Psalter on Mount Sinai” in Scriptorium Vol. 9 (Bruxelles 1955), pp. 177-199, “Two New Latin Liturgical Fragments on Mount Sinai” in Revue Bénédictine Vol. 74 (Maredsous 1964), pp. 252-283 & “Two Other Unknown Latin Liturgical Fragments on Mount Sinai” in Scriptorium Vol. 19 (1965), pp. 3-29, all reprinted in idem, Palaeographical Papers, 1907-1965, ed. Ludwig Bieler (Oxford 1972), II pp. 417-440, 520-545 & 546-573 respectively.
8. Florian Kleber, Robert Sablatnig, Melanie Grau & Heinz Miklas, “Ruling Estimation for Degraded Ancient Documents Based on Text Line Extraction” & Maria C. Vill, Gau, Miklas & Sablatnig, “Static Stroke Decomposition of Glagolitic Characters” in Sablatnig et al., Digital Cultural Heritage, pp. 79-86 & 95-102 respectively.
9. Ibid. pp. 11-35.
A Corner of Tenth-Century Europe 
One project that it would have been nice to have represented that wasn’t there seems to have hit the news at exactly the same time as this post went up, some software for modelling diving on digitized shipwrecks developed at the University of Hull. Two reports can be found at David Beard’s Archaeology in Europe.
Nice post, and thanks for the update and blog on the conference. Some really interesting material and projects.
I should now add to this that my paper is no longer at that link, because contrary to my initial expectations, it is in fact being published and should be out very early next year!
Pingback: Uniting the uniters: electronic resource corpora and competition « A Corner of Tenth-Century Europe
Pingback: Conferring in Naples, III: a full day’s talking « A Corner of Tenth-Century Europe
Pingback: Name in print IX & X | A Corner of Tenth-Century Europe