3.2 Annotation tool

Our latest computational efforts have been focused on implementing an annotation system on the Manuscript Viewer. The goal is for everyone to be able to add comments on the transcriptions and, potentially, directly on the manuscript images. We believe that such system will offer a more direct and flexible alternative to the forum, in cases where comments are in direct connection to the text (for example corrections on the transcriptions) or the image (for example notes on image elements).

We initially researched existing annotation tools, as one of this project’s goals is to try to connect existing solutions and to work towards the standardisation of computational tools in humanities rather than redeveloping modules for unique use.

Text annotation is a relatively easy task, with a number of solutions already present. The tools vary from simple plugins to complete, visually powerful solutions. Even though providing a comprehensive list of the existing tools is beyond the scope of this post, it is worth mentioning hypothes.is, which belongs to the latter category. Hypothesis can be easily implemented on any website and provides an intuitive annotation interface (click here for a demo). However, the annotations are stored on the Hypothesis server and the website owner has no moderation rights over the added annotations. This complete publishing freedom has strong advocates as well as opponents.

In our case we decided against it and opted for a system where we would still have the possibility to moderate comments, mostly for spam control and filtering of inappropriate comments. We are currently working on the implementation of an annotation system based on the Javascript library Annotatorj and hope to be able to share it with you soon!

Example of annotation

Things are a bit more complicated when it comes to annotating images. Even though there are some tools focused on image annotation, they either include or are dependant on the image viewer. To understand why, we have to keep in mind that most modern viewers provide the possibility of manipulating the image (move around, zoom). Therefore, any annotation tool working with such viewer would have to “understand” the way the viewer manipulates the image or, in other words, to be compatible with the viewer’s encoding of the image’s coordinates, zoom level, etc.

A successful example of an annotation system adapted to a zoomable viewer is the Annotorious application on OpenSeadragon. However, there is no annotation tool currently compatible with iviewer, the image visualisation library used by EVT.

Even though our current focus is a stable text annotation system for the transcriptions of our manuscript viewer, future efforts will focus on extending this functionality to the manuscript images.