Digirati have been working with the Wellcome Library, The British Library and the National Library of Wales to further develop the viewing technolgies introduced in the Wellcome Player.
Please see IIIF and the Universal Viewer before implementing the data model described on these pages.
We can't avoid the imaging part of the workflow, but we can automate two key parts of the back end.
The first is to use an image server to generate the tiles on the fly. IIPImage is an image server than can generate the deep zoom xml file and the tiles. This means your server just needs the source images, you don't need to slice them up in advance. An image server also allows you to offer the features on the Wellcome Library site - arbitrary downloads of cropped regions of the image, or downloading the whole image at various sizes.
The second is the generation of the package data for the Player. We can't be prescriptive about this because it all depends on how it's stored (METS files, databases, other library systems) but however it's done, server-side processes need to transform the raw data into the serialised JSON format consumed by the player, as described on the data model page.
Future development of the Player will involve:
Cut out the tile slicing step - just upload your hi-res image files, and use a friendly UI to order the book, preview it and give structure.
The GitHub project already includes an example of a deep zoom image where the Player is consuming IIIF image data rather than Seadragon dzi. The Player uses OpenSeadragon, which supports the IIIF image API.
However, there is also a new format for metadata for entire works (not just images) - the IIIF Presentation API. This is conceptually similar to the package model described in the data model section, although the Wellcome Player package model is designed for all digital library objects, not just images or image sequences.