A number of conclusions may be drawn from the work done in the first two phases of this project. These relate to the conduct of the project itself and to the final findings and recommendations that this project will have to offer.
During the first two phases it has become clear that much work has been done by other projects which directly relates to the scope of this project. Also little consolidation appears to be taking place between some of these key activities. This project has done some valuable work and this can be reinforced by relating it to the experience of others and perhaps acting as a catalyst to draw together the conclusions of these various activities. It should therefore examine, analyse and consolidate the output of these other initiatives for inclusion and/or reference. These other activities include but are not limited to:
The sector is predominantly a user sector and therefore it is not technically innovative as regards on-line access to its digital material. It simply does not have the resources to support innovative development. The sector will therefore make use of existing technology. In doing so, it will have to make trade-offs. For instance there are trade-offs in digitising technology between cost and resolution, cost and colour quality. There are also many pitfalls to negotiate to avoid for instance going up a technological road that leads nowhere. Our findings have uncovered a need for general guidelines and easy to understand practical advice in order to facilitate the wider uptake of the results from numerous other related but uncoordinated studies. This will assist institutions to make good investment decisions. The MDA IT Survey and part 5 of the Magnets report are good starts in this direction but much more is needed. This sector has also been involved in a number of pilots. Many pilots seem to have been and end to themselves, and do not have financial or management commitment to extend to a real 'production' environment. The danger is then that they investigate 'interesting' technology rather than concentrate on cost efficient bulk processing, which seems to be a key deliverable in this context. There seems to be only limited interest in either cost or equipment sharing initiatives - a sort of mobile production environment which the VASARI project kicked off, but does not seem to be continuing at present.
Institutions have been very consistent in the on-line publication of their images. Low resolution images are the norm on the Web. Technology does allow parts of images to be magnified (zoom) but the high definition images to make use of such facilities sensible are generally not available on-line. Institutions understandably want to protect the IPR of their images especially those of high quality and resolution. There are a number of current initiatives under way (e.g. IMPRIMATUR, see https://www.imprimatur.alcs.co.uk/imprimatur/about.htm) which are addressing this issue by the use of such techniques as digital watermarking and WIPO has recently developed a treaty to outlaw tampering with such techniques but until standardised market driven solutions become available, this will remain a barrier.
The digitisation process is expensive and institutions are looking to protect their investment in digital images so acquired. Often an institution will use leading edge technology to obtain the highest quality images obtainable for a given price. Sometimes that technology may not lead anywhere in the future. Institutions therefore want to ensure that such images will be reusable in the future and are interested in longevity and upwards compatibility as newer technology is introduced to the market. However, this has not always been the case and institutions are understandably cautious in their approach and this situation does present a barrier to further exploitation.
In providing on-line access to collection information and images, institutions wish to provide differing levels of access depending on who wishes to access. Thus a general member of the public may only have access to a small proportion of what is available, whereas a professional researcher or collaborator may be granted a much wider access. Payment also has a key significance plus items such as 'pay per view' and 'time expired' viewer technology. Access control is generally implemented on an ad-hoc basis at present. There are a number of initiatives underway in this area (e.g. OCTALIS which is also dealing with copyright protection issues, see https://ns1.tele.ucl.ac.be:80 /OCTALIS/index.htm.). Inhibitors will remain in this area until a recognised framework of access protocols becomes generally available.
A clear message that came from the field survey was the need for guidance on best practise or process related guidance. It is also one of the most significant concusions. For example, digitisation of images is now a common activity but still a slow process. This procedure needs scaling up so that large collections can be processed effectively. Guidance is needed on establishing the process based on existing experience. This requires guidance not only on equipment and software but also on other things such as environment, organisation and procedures.
The cultural heritage sector is primarily interested in data content standards. The AAT is mentioned many times. Several institutions are extending this thesaurus for their own use. It seems that some coordination of these extensions would be a good idea. Translation of thesauri is another area ripe for development to support multilingual use. Associated with this is the need for integration of terminology across linguisticbarriers, and discussion of how to handle semantic differences between translated terms.
There is beginning to be a requirement for definition and use of common meta data standards among a large number of institutions as a means to facilitate synergy and information interchange. Many institutions probably have only scant knowledge about applicable meta data standards.
Another area where standards are needed is for composite data capture and organisation both for qualitative and quantitative information. For instance, in the archaeological sector, there has arisen a need for standards to integrate spatial geographical data, 3-D topological information, remote sensing data and information from digital excavations.
Object oriented computer science methods and databases are potentially well suited to handle composite Cultural Heritage Information. There is therefore a potential requirement to start defining some general abstract object classes and methods to be used in Cultural heritage Information systems. (Similar work is underway to define general business object classes and methods).
Data standards are needed to ensure high level interoperability but there is also a need to ensure interoperability by means of general system interface standards and tools. Some mechanism should be put in place which would allow the results from isolated implementation efforts to be combined and built upon. The CIMI test bed project is a good basis for a start in this area. It concentrates on interoperablility and common tools.
One example of the need for standards in this area has come from the archaeological arena. Demand for solutions in this area is beginning to arise from other sectors. Not only are technical standards required but also best practice guides.