Upload is part of the interdisciplinary research project From Object to Icon, conducted at the Institute for Egyptology in cooperation with the research group Multimedia Information Systems at the University of Vienna (funded by the Austrian Science Fund, project number P 25958). It is based on the research that was initiated with the project MeKeTRE (Middle Kingdom Tomb Relief Evolution), in the course of which we have started to systematically collect, research, and study the reliefs and paintings of Middle Kingdom tombs of Ancient Egypt. For more details on the projects click here.
The data collected so far are available online in the MEKETREpository that has been developed in order to serve public use. It provides users with a collection of themes and scenes attested in the decorative programme of the tombs of officials datable to the First Intermediate Period and Middle Kingdom (ca. 2150–1640 BC) and encompasses plans, images (drawings and photographs), descriptions as well as references.
Upload constitutes a crowd sourcing approach that is ideally suited to enrich the MEKETREpository. It is a platform enabling users to upload unrestricted high-quality photographs depicting relevant art items in Middle Kingdom tombs, provide annotations, or suggest inclusion of new thesaurus terms. It offers an easy-to-use interface through which everyone can share private photo collections and perform simple repetitive but highly helpful tasks, thereby contributing to the scholarly enterprise. Upload is meant to engage both scholars as well as the interested public.
The expected results are twofold: First, we aim to acquire extensive material (especially photographs) that has the potential to complement the MEKETREpository. All data of sufficient quality gathered through Upload will be regularly transferred to the repository and certainly improve its usability. Second, the methods developed and applied in the implementation and data gathering process will constitute a contribution on their own, hopefully providing valuable insights about quality assessment and integration of data coming from citizen science projects.
Utilizing a crowd sourcing approach to support the work of Egyptologists is a novel yet promising way to assist the workflow of scholars.
Collecting data is often a tedious and time-consuming task but the effort can be reduced by utilizing crowd sourcing technologies such as Citizen Science that involves all users of the Web in solving scientific problems by performing simple, often repetitive tasks. In the case of the MEKETREpository , the dataset is manifold; we can, however, identify three main aspects of possible improvement: images, annotations and terms.
Images form a crucial part of the MEKETREpository because they serve as a basis for graphical annotations. However, it is often not possible to provide users with images of the scenes compiled because copyright restrictions prohibit free distribution of almost half of the uploaded images. Furthermore, images of some scenes are of poor quality or are impossible to locate with reasonable effort (e.g. because of out-dated and insufficient publications). Whenever images cannot be presented alongside a query, it impacts usability of the repository in a negative way; thus, we would appreciate if users could provide as many unrestricted high-quality depictions as possible.
Users are encouraged only to upload their own material and to acknowledge that every image, once uploaded, will become property of the Institute for Egyptology (University of Vienna) and will be licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International license. The users must confirm that no third party holds the rights on the material or provide accompanying license information.
The annotation facilities in the MEKETREpository application are used to entitle the scenes and motifs as well as to term icons (smallest pictorial elements such as various animals, plants, tools, etc.). They have been designed with a strong focus on usability and were widely used. Nevertheless, coverage of all depicted icons is far from being complete, but the corpus can easily be extended by users willing to annotate images available on the Upload platform. A more detailed set of annotations would provide scholars with accurate data suitable for future research and better query results (in terms of recall and precision).
The terms currently available in the dataset establish a controlled vocabulary, capable of categorizing and describing icons in a standardized way. These terms are organized in a hierarchical relationship and many of them are available in different languages in the Thesaurus. While the terms used for categorization are already highly nested, such a structure is less evolved, e.g., for icon annotation terms. With an increasing number of annotations, this structure is expected to grow further in both number of entries and structural complexity.