Annotatist Simon Post
Summary Findings for AnnotatIST - Annotate ISTanbul gathering
Workshop title: Future for the Annotation of Digital Objects
Organizer: Dr. Yuk Hui, Simon Worthington, Hybrid Publishing Lab, Centre for Digital Cultures, Leuphana University Lüneburg
Participants: Bob Stein (Institute for the Future of the Book, SocialBook), Christina Kral (Hybrid Publishing Consortium), Claudius Teodorescu (University of Heidelberg), Andre Gaul + Nico (PaperHive), Thomas Kollatz (DARIAH), Paul-Emile Greffroy (IRI of Centre Pompidou), Johannes Wilm (Fidus Writer)
Date: 12-13 May 2015
Notes: https://consortium.titanpad.com/44
*Framing and Conclusion*
- Annotation happens in many different ways in a wide variety of sectors, this has an impact on approaches to the application of standardization. - The general model of the use of annotation in digital systems is as a subordinate and secondary activity following on from the model of comments, forum and bulletin board systems. This general approach should be challenged as it is limiting to the role of annotation. - There is a shared interest in the workshop, put forward by Yuk Hui, for annotation to be considered as key to teaching and learning contexts. This relates to raising the perceived value of the annotation as equal to the object being annotated, for there to be no need for ontological split between objects in 'collection management' and annotations as an example. See: http://grahamlarkin.info/art-conservation/ . This elevation or reassessment of the annotation is worth exploring in terms publishing workflows and in reception in terms of context like library collection organization.
Technology issues: 1. A standards for annotation, 2. Media location, 3. Long term accumulation of annotation on an object, 4. Distributed or decentralized annotation on changing documents.
Interesting uses of annotation: 'Lignes de Temps'- IRI of Centre Pompidou, Paris. http://www.iri.centrepompidou.fr/outils/lignes-de-temps-2/
*Technical*
The basic technology frameworks, standards, conventions for annotation and necessary interoperability appear not to be resolved to allow development projects by groups to make robust, interoperable technology implementation or annotation content projects that can operate outside of limited and specific technology environments. e.g. a book distribution system like Kindle or a Digital Humanities study project using one technology stack or software application.
These missing basics are:
1. A standards for annotation A stable and widely used standard set is missing. W3C has Web Annotation Data Model but this is still in DRAFT http://www.w3.org/TR/2014/WD-annotation-model-20141211/ There is also the open annotation 'Open Annotation Data Model' http://www.w3.org/ns/oa US Library of Congress controlled vocabularies like VRA also do not seem to have accommodated annotation adequately.
Note: our discussion raised the issue that standardization in terms of enforcing uniformity might not be appropriate as annotation has such a wide set of applications. But interoperability is desirable.
2. Media location The issue of being able to reliably reference a media type object, then a location within the object, even for common media types such as text is unresolved.
There are a number of interesting approaches: A. Hypothesis - go for a layers combination of elements to resolve a location in an Open Web document. B. Models of Stephanus references convention of citation for Plato's text seems reasonable. In that an arbitrary document version is set as the master and then vectors assigned–page number, place, or lines on page etc. (See: http://plato-dialogues.org/faq/faq007.htm and in visual form http://plato-dialogues.org/stephanus.htm) C. The model of 'Revisioning Systems' seems the most promising as location, and revision can be located and reconstructed. But this has the limitation of the media needing to be in revisioning system.
3. Long term accumulation of annotation on an object If a text has many annotations over time, how are the annotations sorted and filtered, for example in a social reading environment?
4. Distributed or decentralised annotation on changing documents needs to better solutions. Synchronization of annotations, when document/digital objects changes. Files and change logs (deltas) for moments in time T1 and T2. Examples on GitHub and Google Docs. Using Javascript range API (https://dom.spec.whatwg.org/#interface-range). A selection in a document can have more ranges, see https://dvcs.w3.org/hg/editing/raw-file/tip/editing.html#selection.
*Projects*
The most interesting projects (IMHO) were from 'IRI of Centre Pompidou'. For example 'Lignes de Temps' for offering contra opinions on a video http://www.iri.centrepompidou.fr/outils/lignes-de-temps-2/ See others at line 289: https://consortium.titanpad.com/44
*Writing*
Collective Individuation: The Future of the Social Web (Yuk Hui (Leuphana)/Harry Halpin (W3C): http://www.digitalmilieu.net/documents/Hui_Halpin_Collective%20Individuation.pdf
Bureau for Book Liberation
The Bureau for Book Liberation (#bbl) looks to free publications from the confines of the bound paper codex and monopolies of distribution.
It seeks to establish new forms of reading and accessing information while envisioning alternatives to restrictive corporate copyright.
https://github.com/consortium/Bureau-for-Book-Liberation
In a nutshell #bbl is a design prototyping process run as a series of book remixing workshops, initially from Feb. to Jul. in a Public Library, Berlin. http://apubliclibrary.org/
The germ of the prototype idea is to design a book of the near-future.
“a reading experiences where transmedia references are immediately made available in a document—inline, in full, or with full media playback and with a full citation and revision history.“
Currently the state of play for the prototype is the model of personal note taking, using a desktop/device text editor and annotating ones own reading. This is then synched using GitHub and connected to Zotero for shared bibliographic data. This is our quick route to testing out book remixing, with Github being used combine the 'personal notes pads' in simply styles HTML/CSS/Javascript documents. The results are still very much in the pipeline. Already we see that the process can shape the library collection, feedback on the collection and help explore the library as a place of care for maintaining objects and annotation as part of the learning activities in the library––recommending titles to others, showing routes through various titles etc.
