This is Todd Hanneken, Director of the Jubilees Palimpsest Project. This screen recording illustrates the major tools in Mirador for study of Latin Moses, a fifth-century copy of a collection of texts known by scholars today as Jubilees and the Testament of Moses.
Start from the homepage of the Jubilees Palimpsest Project and click on Mirador.
The first screen shows the objects available. Latin Moses, shelf mark C73 inferior at the Biblioteca Ambrosiana in Milan, has all the features we will demonstrate. From here we can jump to any of the first few pages, or click on the title for a gallery view of thumbnails for all pages. If we're looking for a particular page we'll want the index tab on the left side bar, which appears by default. This index shows the range of chapter and verse attested on that page, along with two page numbers. The first page number is the reconstructed page number of the original fifth-century codex. The second page number is the number given in modern times to the manuscript after it was palimpsested in the eighth century. If we want to study the earliest part of Jubilees preserved in Latin we click on Jubilees 13:10-15. We can also move between pages using the arrows on the right and left of the screen, or the tray of thumbnails at the bottom. We can hide or restore this tray by clicking the three dots above the center of the tray.
The zoom and pan features are intuitive. We can zoom in and out with click and shift-click, or the wheel of a mouse, or the buttons in the lower right. Panning can be done with a mouse or finger drag, or the buttons on the lower right.
In order to see the enhanced spectral processing and raking images we will need the “Layers” tab on the left pane. We can turn on layers for raking light coming from the North-East, South-East, etc. We can turn on transmissive to see the holes and scribal punctures, as if holding the folio up to a light. We can also turn on color enhancements of multispectral data, such Extended Spectrum and Keith Knox’s Ruby Image. This image shows the erased text in the darkest color, the overtext in royal blue, and the overtext from the other side in light blue. An image of Ceriani’s 1861 edition of this page is also available as a layer.
The annotations tab shows us more resources available. First, the WebRTI images are the interactive versions of raking illumination for each color process. Clicking on one of these will open a new tab. Again we can pan and zoom, but the main feature is the light-bulb icon. Clicking this turns the mouse or a touch screen into the control of a virtual flash light. If we click or drag to the upper right, the light comes from the upper right, and so forth.
Back to the annotations tab, we also have a few transcriptions and translations available. Ceriani’s 1861 edition has been converted to TEI XML with some reader aids, such as verse numbers. This is available in a simple view, or XML and HTML for EpiDoc purists. VanderKam’s 1989 translation of the Latin and Ethiopic are available for reference. The translation labeled “Latest” will diverge from the 1989 translation as new readings develop through the Jubilees Palimpsest Project.
Community-added overlay transcriptions also appear in the Annotations tab, but unfortunately may appear by tag rather than the main annotation. The best way to see these annotations is to click the “Dialog” icon for “Toggle annotations.” Now we see blue rectangles for all the annotations overlaid on the image canvas. Hovering the mouse over those rectangles shows the annotation, most often a transcription with tags for who contributed the annotation and the degree of certainty asserted. Much work remains to be done to study and annotate the images. Anyone can do so by clicking the rectangle in the annotations toolbar and dragging a rectangle around any region.
The final toolbar addressed in this video is the “image manipulation” toolbar toggled with the “sliders” icon in the upper left. These manipulations are not the same as spectral image processing. The “Rotate” icons might be useful if one wishes to read the overtext. The “Mirror image" icon might be useful if one wishes to read text showing through from the other side.
That concludes this introductory overview. Stay tuned for additional videos showcasing resources for study outside of Mirador, including the reconstruction of the original codex, the paleography chart, and options to search the index of overlay annotations.
The first video mentioned the reconstructed page numbers of the fifth-century pre-palimpsest codex. The reconstruction of this codex can be visualized in HTML using the zoom feature of web browsers. First, click on “Reconstruction of the fifth-century codex” on the project home page. Each box represents a page, and is gray if the page is lost and white if preserved at the Ambrosiana. The numbers show the quire, bifolio, folio, and page number in the reconstruction, along with the range of verses and links to view the page, if preserved. If we zoom out until there are sixteen boxes across we will see the codicological structure with one quire per row. Zooming can be done by holding down “control” on the keyboard while turning the wheel of a mouse. The view is symmetric because the bifolia were kept intact during the reuse of the parchment. Whenever a page is preserved, so is its other side and partner across the fold, which is consecutive only in the middle of a quire of four bifolios, or sixteen pages. The first five quires were lost, but otherwise the shuffling of bifolia into the preserved and now lost quires is random. The reconstruction was first proposed by Rönsch and repeated independently by Hanneken on the premise of counting how many lines of Ethiopic Jubilees correspond to a page of Latin Jubilees, and aligning with some preserved quire numbers. The extrapolation works remarkably well for the gaps, and suggests that Jubilees started on the second folio. Along with the index demonstrated in the screen capture of Mirador, this is a good way to determine if a passage of interest is preserved and to get to the images if it is.
There is no one way to perform a search, but different ways of searching the transcription, translation, and user annotations. In each case the most reliable way to search is simply the “Find” (ctrl+f) function of a browser. The transcription and translation links in Mirador take us directly to the page from which we came, but the entire transcription and translation are above and below on the same page. Thus we can search the entire work quickly from that window. From there we can find the chapter, verse, and page number (which differs for the fifth-century and nineteenth-century paginations) and browse to that page in Mirador.
There is a similar but separate way to search the community-contributed overlay annotations. From the project homepage click “Index of overlay annotations.” This index is dynamically updated and is never more than a refresh away from the most recently added annotations. We could search for annotations tagged “contraction” with ctrl+f and typing the search term. The annotations are grouped into the page numbers on which they are found, so we can quickly see where they are and jump to images of that page. We could also search for what annotators have said about the mysterious figure in the Testament of Moses, Taxo. In this case it has been suggested that Taxo could also be read Taao.
We could check how the letters X and A typically appear in the manuscript by consulting the paleography chart linked from the page About the Jubilees Palimpsest on the project home page. Here we can see that an A is typically a thick descending line with a light loop on the lower left. An X has a similar thick descending stroke and a lighter rising stroke that crosses near the top. In the case of Taxo or Taao, there is no clear loop to confirm an A, nor is the upper right of an X visible. It would also be unusual for the strokes of an X to cross so low. It could be either an X or an A. Attempts to identify the figure using gematria should consider both possibilities.
If you are wondering how long it took to crop page images into single letters for the paleography chart, you might be interested in learning more about the IIIF Image API. The original image is left intact on the repository, while a URL specifies the region desired, in this case a single letter. Although Mirador is the most powerful viewer, especially for annotations, some valuable tools can only be accessed through the IIIF Navigator. A link to IIIF Navigator is prominent on the project homepage. To use the Navigator we need to identify a manifest, in this case Latin Moses. We can show all resources in the manifest by clicking the appropriate radio button, then “List resources in manifest.” This can be thought of as a more direct visualization of the information available in Mirador. Mirador does not show the URL source of a IIIF Image or provide tools to manipulate it for use elsewhere on the web. If we wanted a single letter for our paleography chart we could click “crop” and then manipulate the page and selection until the region we want is highlighted. The link in the upper right shows the IIIF Image API address for just that region. Now we can send or use the web address for that region of the image without copying, editing, or attaching images.
If we want a web address not just for a region of an image but a complete view in Mirador we can copy and paste from the address bar. Notice when we navigate Mirador that the address in the address bar changes when we change manuscripts and pages, which are called manifests and canvases in the IIIF Presentation API. We can bookmark or copy and paste these web addresses to share a particular page. The ability to specify a particular layer or region may be added in the future.
Special thanks to Annette Y. Reed for suggesting the creation of these videos at the Jubilees Palimpsest Project workshop at Notre Dame, May 2018 (LINK).