CIHA London 2000.
Thirtieth International Congress of the History of Art
Art History for the Millenium: Time.
Section 23
Digital Art History Time
London, 3-8 September 2000

Michael Greenhalgh <Michael.Greenhalgh@anu.edu.au>, Fowler Hamilton Research Fellow, Christ Church Oxford, and The Sir William Dobell Professor of Art History, Department of Art History & Visual Studies, AD Hope Building, Australian National University, Acton, Canberra ACT 0200 Australia. ArtServe.

Teaching and Learning Art History using the Web

© each author has full responsibility in owning copyright on the texts and on the images they publish on this website

Introduction

The aim of this paper is to promote and illustrate with concrete examples the advantages of the web as the only efficient way of offering course materials to a student population whose work practices differ greatly; to insist that lecturers must fully understand the technology in order to commission innovative applications and them to apply them in suitable ways; and to argue for greater co-operation (especially with image-banks) in this age of connectivity. It relates to the central theme of Digital Art Time in that it directly addresses teaching, learning and publishing; and research tangentially - since we must research ways of presenting digital learning materials more effectively if the discipline is to prosper under continuing funding and staffing restraints. In other words, it is a practical paper relating my experiences over the past four years in using image databases in many of my taught Art History units, and lecturing from web images as well (using a feed to video projectors in the lecture theatre). The paper touches on the pluses and minuses of such online units from the pedagogical point of view, and also on updating, unit management, and financial and quality issues. It concludes with a look at the likely future, with wearable and palmtop computers capable of virtual reality providing new study scenarios which, using wireless modems, may well free the user from a complicated university network infrastructure.

Rationale for using Computers in Art History

The point of using digital media and then broadcasting them using the web is to give students access to an environment richer than that of a darkened room with one/two slide projectors, where the images shown are of similar dimensions no matter what their real-world size.

Digital materials are the way forward because they:

digital images offer non-degradable, increasingly decent quality; they can be copied, amended overlaid with text, cut into useful sections, etc;
allow individual artworks to be given context;
offer through properly designed pages and database front-ends far greater quantities of information about artworks than any 35mm slide can bear; and - if we got our act together - this could be in some standard format similar to the MARC etc rules for cataloguing books. Full information - the result of network-wide cooperation? - could then be extracted via databases. Simple to say quickly - so why isn't it happening?
extend students' horizons: the aim of employing as many media as are reasonable (multiple media are possible: sound, video, constructed worlds) is to make a poor substitute to "being there" - but a better one than slides, or photographs in books, can offer; we do not throw out old ways of learning, but enrich them: image copyright problems exist only with 20thC work (although access problems exist for all periods); many out-of-copyright texts are available for reproduction, and large banks of standard "classics" are already online;
supplement the printed book and are often more convenient, especially for art-historians, because they offer multimedia possibilities, at the very least integrating text with images from databases; see my High Renaissance web-based unit;
can sometimes help re-integrate art in its context; see my VRML model of Borobudur;
in association with email and listserv-type code, offer the only sane way of administering courses, offering visual tests, returning marks, etc;
can be "layered" to address different levels of knowledge, but if necessary always using the same reservoir of images;
in contrast, slides presage problems and costs in labelling, curating, filing, breaking, losing; their colours are fugitive, and a replacement policy (unnecessary with digital images) is advisable. Again, digital images can be "posted" much more easily than physical slides; so the age of every Art History Department having a Slide Library should be nearly past, because there is no longer any need to look after objects which exist in the one location in only one copy;

What is right/wrong with the web for teaching/learning?

If the attractions offered by digital media are persuasive, may we say the same about the web? Thanks to increasingly cheap machines and storage, ever-faster networks, and 24-hour availability, this has certainly been one of the important innovations of the past few years, and there can be no university that is not using it and contemplating ways both to fulfil its mission by so doing and (vain hope?) to save money in the process. But the fact that this section is not integrated with the previous one underlines the fact that using the web is not a precondition of going digital. However, at the moment the web, in spite of the snags listed below, seems the best vehicle for learning materials, at the very least because it provides a basic lingua franca to which increasing numbers of people have access - and much of the technology is in the public domain.

The web is:

powerful but fragmented: we need better catalogues. Where, for example, should I go to find a complete, illustrated digital catalogue of the works of Raphael or JL David? We can look up the holdings of specific museums; but the web should be an agent for synergy - not fragmentation;
well-patronised but offering a lot of garbage (self-publishing means academic standards - if any - are self-imposed), all of which needs to be weeded out by sophisticated search-engines; nevertheless, such engines can only work if the basic structure of web pages is improved. HTML is too easy to write badly, and insufficiently structured; it works without many visible standards;
expense is the key: it is expensive to implement standards; and just as it is cheap and quick to mount rubbish, so it is expensive to write proper catalogue entries;
fugitive: it is easy to mount and change material - a flexibility which has its down side: the web needs to impose concepts of timeliness and out-of-dateness; and who will maintain suites of pages in 50 years time, after the death of the server or even its owner?
democratic but crowded: we need dedicated academic networks with reliable transmission speeds - the more so if video and audio are to become widely used media in academia;
anarchic: this has helped the vitality and rapid development of the web - but anarchy militates against collaboration, and we cannot build Virtual Art History without an acknowldged lead of some sort (such as cataloguing and page-writing standards);

Art Historians using the web should:

develop an understanding of the technologies they wish to use. They cannot even specify, let alone control, what programs and programmers can offer them if they do not understand the potential and the drawbacks of the various technologies;
promote inter-university cooperation as the only possible use of scarce resources, which shouldn in any case go as much as possible on staff;
use the web as the cheapest efficient path to flexible learning: getting caught up in various proprietary technologies, especially those which work on only one platform, is unnecessarily restrictive, and can be a waste of resources;
evolve programs which make student use of the web a viable alternative to 35mm slides, providing not only flexible databases, but quiz setups; at the same time we should be careful to make clear that bookss are unlikely to be replaced by the web very soon, if ever - that is, to demonstrate that the web should be used

But does the Web work as a learning Medium?

I can illustrate such points by reference to ArtServe, a machine running Linux which concentrates on the Art of the Mediterranean basin, contains over 130,000 images (i.e. circa 48Gb of material), and takes an average of about 90,000 hits per day. It offers programs written in-house for the bulk-processing of digital images, and their regimentation via databases into HTML pages, for example using:

rdbweb: allows the user to view a set of thumbnail images on HTML pages, and provide data records for them by typing the details into a web browser; in other words, as the name implies, it is a relational database front-end where the cataloguing is done directly into an HTML form and - such is the nature of the web - where the cataloguer need be nowhere near the stored images;
salami: takes a data file for an image database, and slices up the records into HTML web pages of any length, with any number of images per row and per column, together with indexes, and previous-next hotlinks. The big images are accessed by clicking on the displayed small thumbnails;
illuminate: takes a text-file with references to images, and produces almost-elegant HTML pages with the images displayed as clickable thumbnails, wrapped in a tabular format containing any or all fields from the data record;
light-table: allows the user to interrogate an image database over the web and, from the images viewed, select and rearrange chosen images for dumping into a HTML page. This is useful for live lecturing over a video feed; for private study, or for printing to paper;
quiz allows students to test themselves on multiple-choice image materials from a variety of databases stored on the server;

Technologies for Learning Art History

If using computers for digital imagery is a good idea, and employing the web as the most convenient distribution medium, then what kinds of solutions of use to Art Historians may be employed? These may be listed as follows, from the simple to the more complicated:

Simple web pages with thumbnail images (click on the thumbnail: this brings up the larger version), such as my rubens' home page; where money has been available, such images will be accompanied by database details, sometimes encapsulated in smart tables;
Even video images can be used because, in spite of their low resolution, the amazing zoom capabilities of digital video cameras (up to 20x manual and yet more digital is the norm) and their extraordinary light-gathering ability allow the photographing of distant objects high-up in gloomy churches - in those instances where even a bad image is better than no image, as in these examples from Toro;
web multimedia are the key; and use browsers to access CDROMs;
there are drawbacks to using both video and audio - large file sizes, and very large file sizes and transmission problems if bigger than postage-stamp video is required;
simple images can be extended by the use of panoramas, because they are simple to construct, and give added context; although narrower panoramas are often adequate, 360 degrees is almost like "being there". Hotspots can be added to panoramas which give links to text, other images, or other web sites. Again, panoramas may be linked together, allowing the user to move from one space into other spaces, and back again - i.e. to take a virtual tour. Indeed, various other bells and whistles can be added for web use;
stereo imagery: needing red-blue glasses, stereo offers a more "realistic" view of three-dimensional objects such as sculptured capitals - although the usual "bleaching" effect means that it is not of much use where accurate colour rendition is required;
constructed worlds using languages such as VRML: these "worlds" can be complicated, as seen in these views of my Borobudur Project mentioned above;
presentations incorporating several of the above technologies;

Of all these techniques, panoramas are the easiest to construct (in both time spent and skill required), and they can be divided into several types including:

simple image-strips, offering a wide-angle view of a scene;
zoomable images, so that the user may move closer to the items in the panorama, and examine them closely;
panoramas with links, so that objects in panoramas can be linked to HTML targets such as images, video, or other web links; by extension, one panorama can be linked to others, offering the ability to move through a sequence of panoramic spaces;
large panoramas - over one megabyte - can be impressive - if you have a powerful enough machine, a fast network, and if you don't mind the time spent loading;
panoramas (essentially imagemaps) are usually horizontal, but may also be constructed vertically, very usefully allowing whole wall sections of buildings to be imaged and hotspotted;

What is virtual reality?

The use of a computer to construct a version of the world we see with our eyes. We can interact with this world by giving directions to the software to move, pan and zoom the scenes. Uses include education, process control, training (e.g. for surgery, piloting, firefighting).

Advantages: sense of "being there"; real-time adaptation when necessary; the foundation for hotspotting to extend the learning experience;

Disadvantages: difficult and expensive to construct and display properly; uses expensive computer resources, or takes a lot of time and patience to construct; in the future: the disappearance of small computer monitors in favour of wall-sized displays much nearer the scale of the buildings represented;

The Classroom/Student of the Future

Here are some predictions based on several years experience of using computers, the net and now the web as teaching and learning tools:

classrooms will probably survive, because people need people;
small portable computers with video, audio etc, and used for uploading material from the web or a bancomat-type machine;
lectures delivered using digital images pulled from the web and video-projected into the lecture theatre (as I have been doing for three years);
students issued with CDROMs of course images, and initial unit documentation; the web is the noticeboard for all augmentations, changes, updates;
programs which check that essays have not been downloaded from the web;
seminar presentations mounted in advance as web pages;
theses presented on CDROM with multimedia where appropriate (NB I have a 1994 Geology PhD from Stanford submitted this way;
computer monitors quaint and exclusive, very like the standard "renaissance window"; replaced by wall-size right-angled screens offering virtual reality, it is possible that VRML and associated technologies will indeed win in the classroom. always assuming they become much easier to build and "fit out" with hotspots and links;
stereo glasses for the study of sculpture, bas-relief and architecture? after all, the Walkman looked funny when introduced;
conceivably, robot cameras at important sites, controlled from the lectern or by individual students;

Some useful links, including 3D projects

The Web 3D Consortium;
The VRML Repository;
Software to make 3D models from photographs:
1. PhotoBuilder;
2. Photomodeller, and examples;
3. ShapeCapture;
4. Canoma, with examples of some Canoma models;
3D models from measured drawings with photographs added: Borobudur;
"Easy" technique: Reconstruction from uncalibrated photographs: the full paper is here;
An archaeological site reconstructed using VRML: Virtual Sagalassos;
stereoscopic photography: anaglyphic stereo imaging;
software for making panoramas: Ulead Cool 360; PhotoVista;
stitched panoramas: Piazza del Popolo or Castel Sant' Angelo;

Research links in 3D modelling:

Abstract

Aims:

To promote (and illustrate with concrete examples I use everyday) the advantages of the web as the only efficient way of offering course materials to a student population whose work practices differ greatly; to insist that lecturers must fully understand the technology in order to commission innovative applications and them to apply them in suitable ways; and to argue for greater co-operation (especially with image-banks) in an age of decreasing funding.

Summary Description:

A practical paper relating my experiences over the past four years in using image databases in many of my taught Art History units, and lecturing from web images as well (using a feed to video projectors in the lecture theatre). The paper discusses the pluses and minuses of such online units from thepedagogical point of view, and also deals with updating, unit management, and financial and quality issues. It concludes with a look at the likely future, with wearable and palmtop computers capable of virtual reality providing new study scenarios which, using wireless modems, may well free the user from a complicated university network infrastructure.

Central Arguments:

Art Historians should:

develop an understanding of the technologies they wish to use;
promote inter-university cooperation as the only possible use of scarce resources;
use the web as the cheapest efficient path to flexible learning; and
evolve programs which make student use of the web a viable alternative to 35mm slides;

Digital materials:

supplement the printed book and are often more convenient, especially for art-historians, because they offer multimedia possibilities, at the very least integrating text with images from databases; see my High Renaissance web-based unit;
can sometimes help re-integrate art in its context; see my VRML model of Borobudur;
in association with email and listserv-type code, offer the only sane way of administering courses, offering visual tests, returning marks, etc; and
can be "layered" to address different levels of knowledge.

I shall illustrate my paper by reference to Artserve, a 30Gb machine running Linux which concentrates on the Art of the Mediterranean basin, contains over 80,000 images, and takes about 50,000 hits per day. It offers programs written in-house for the bulk-processing of digital images, and their regimentation via databases into HTML pages, for example using:

Rdbweb: allows the user to view a set of thumbnail images on HTML pages, and provide data records for them by typing the details into a web browser; here is a web page showing the cataloguing of Piranesi prints in progress: the four images at the top have been catalogued, the one in the middle is the current target, and the four at the bottom are next. Type-in boxes and pull-down lists offer two ways of populating the data record;
Salami: takes a data file for an image database, and slices up the records into HTML web pages of any length, together with indexes, and previous-next hotlinks. The big images are accessed by clicking on the displayed small thumbnails; a sample is here;
Illuminate: takes a text-file with references to images, and produces almost-elegant HTML pages with the images displayed as clickable thumbnail, wrapped in a tabular format containing any or all fields from the data record; a sample of the output, from The Borobudur project, is here;
Light-Table: allows the user to interrogate an image database over the web and, from the images viewed, select and rearrange chosen images for dumping into a HTML page. useful for live lecturing over a video feed; for private study, or for printing to paper; a sample is here; and
Quiz, which allows students to test themselves on multiple-choice image materials;

Relation to the Themes of this Section:

directly addresses teaching, learning and publishing; and research tangentially, since we must research ways of presenting digital learning materials more effectively if the discipline is to prosper iunder continuing funding and staffing restraints.

Art History Webmasters ASSOCIATION des webmestres en histoire de l'art
Research and Communication Tools in Art History
Outils de recherche et de communication en histoire de l'art
Since November 14, 1997. Depuis le 14 novembre 1997.