Revolutionary reading, evolutionary toolmaking:(Re)development of scholarly reading and annotation tools in response to an ever-changing scholarly climate
James MacGregor, Michael Joyce, Brett Hirsch, Cara Leitch, Ray Siemens, Chia-Ning Chiang, and Rick Kopak
firstname.lastname@example.org | Public Knowledge Project, Simon Fraser University
email@example.com | Electronic Textual Cultures Lab, University of Victoria
firstname.lastname@example.org | Electronic Textual Cultures Lab, University of Victoria
email@example.com | Electronic Textual Cultures Lab, University of Victoria
firstname.lastname@example.org | Faculty of Humanities, University of Victoria
email@example.com | School of Library, Archival, and Information Studies, University of British Columbia
firstname.lastname@example.org | School of Library, Archival, and Information Studies, University of British Columbia
email@example.com | Public Knowledge Project, Simon Fraser University
KEYWORDS / MOTS-CLÉS
reading tools, text tools, knowledge environment, Professional Reading Environment (PKP), social networks / outils de lecture, outils des textes, environnement de la connaissance, Professional Reading Environment (PKP), réseaux sociaux
Open Journal Systems (OJS), an online scholarly research and publication software application developed by the Public Knowledge Project (PKP), was developed “to improve both the scholarly and public quality of referred research” (“Open Journal Systems,” par. 3) by providing a robust, durable and open source online publication and dissemination platform. OJS allows journal managers to control every stage of the scholarly publication process: article submission and author management; peer review; copy- and layout editing as well as proofreading; and final online publication. OJS can be extended via the use of additional plugins to support extra functionality, such as import/export functionality, metadata extraction and sharing, search engine optimization, and more. One of the more important areas that has been extended is that of user interaction: the point where readers visit the journal Web site and interact with the scholarly content therein.
Traditionally, this point of interaction online has been static and one-way: readers visit the journal web site; read the article they are interested in; and leave, with perhaps a citation at most under their belt. After such a visit all the journal (and perhaps the author) have is a record of one 'hit': one view of the article, maybe a notion on where that view came from geographically thanks to the journal's reporting software, but no other context for the visit, and definitely nothing in the way of reciprocal comments.
Conceived in 2001, the Reading Tools were originally developed to flesh out this point of interaction, and to create a more fruitful context for readers, journals and authors alike (Siemens et al., “It May Change My Understanding” 1-2). They appear to the left of an article when it is viewed, and offer access to article-specific metadata, author information, and a collection of discipline-specific search databases and catalogues, while also providing opportunities for limited email-based communication (to authors and to those interested in the article at hand). The tools were developed to serve manifold purposes: not only to provide contextof the study with the wider literature (Willinsky 8-9), but to provide that context to scholarly, professional and amateur levels of research alike (Willinsky 7).
The Reading Tools have not changed significantly since their first implementation, and although they fulfil their original purpose, the Web doesn't stay still. Social technologies have evolved significantly in the past few years, and users expect a certain amount of networked sharing and communication between applications (Siemens et al., “Prototyping” 26-27). In an online, scholarly context, there are a number of areas in which an investment in social networking tools can pay tremendous dividends: contribution and social validation of notes and comments; commenting upon the work of others publicly, with follow-up between peers and other academics (even, presumably, the authors or the editorial team); and not to forget the rich contribution that would be this captured, public history of communication.
OJS (and the other PKP applications) provide a great publishing framework, and the Reading Tools provide a worthy first step towards further discovery and communication; with a careful development plan based on solid research, the PKP, in partnership with the Electronic Textual Cultures Lab (ETCL) at the University of Victoria, will develop an evolved, forward thinking set of tools to provide a rich social academic framework on top of published online scholarly work.
By default the Reading Tools are disabled, and must be enabled and configured by the Journal Manager. Once enabled, they appear to the right of an article, nested in a frame. At its most basic, the Tools pane will display journal, issue, and article information at the top of the pane, and will have a search option at the bottom which will allow the reader to search across all journal content.
The remaining content falls into two components, which can be individually configured: Journal Item Tools; and Related Item Tools. The Journal Item Tools consist of the following individually enabled tools, available by clicking the linktext:
- Abstract : a link to the article's abstract or details page.
- Review Policy : takes the reader to the journal's Review Policy section in About the Journal.
- About the Author : launches a pop-up window containing biographical information for all authors.
- How to Cite Item : launches a pop-up window where readers will be able to cycle through available citation formats. Choosing a citation style (eg. MLA; Turabian; etc.) will display the properly-formatted citation for the article, while choosing a third-party citation software type (eg. Endnote; Procite) will present said format for download.
- Indexing Metadata : displays all relevant article metadata and how it corresponds to Dublin Core metadata types.
- Supplementary Files : displays any supplementary files and related metadata pertaining to the article.
- Print Version : displays just the article galley (typically HTML or PDF), without the normal journal header, footer and reading tool components added, ready for printing. It will also initiate the reader's Print protocol.
- Look up Terms : provides access to online dictionaries, where any given word can be looked up. (Alternatively, if the article galley is HTML, readers can double-click any word to launch the Look up Terms tool, with the clicked-upon word automatically added to the lookup box.)
- Notify Colleague : launches an email template pop-up window, where readers can add colleague email addresses. Suggested email body text includes a link to the article. Users must be logged in to use this feature.
- Email the Author : launches a pop-up window containing an empty email template; any text entered into the email body will be sent to the author. Readers must be logged in to use this feature.
- Find References : allows readers to search pre-selected scholarly databases for the cited work.
- Add Comment : allows readers to add publicly viewable comments to the article. Comments are displayed at the bottom of HTML galleys, or from the article's Abstract page.
Out of the above tools, only three could be considered 'social' in nature: the Notify Colleague; Email the Author; and Add Comment tools. Furthermore, only the Add Comment tool generates a public discourse attached to the article or to the journal Web site – the other two tools facilitate communication about the article in question, but opaquely, outside of the journal Web site itself. Any correspondence and/or relationships between reader, author, and other potential readers are not kept: there is no historical record of the communication taking place, and the potential for further discourse is correspondingly limited.
The Related Item Tools consist of 19 groupings of searchable databases and catalogues, mostly open access, against which readers can search for further contextual information. The journal can choose only one discipline-specific 'field' of Related Item Tools from the 19 default groupings (eg. Agriculture; Biology; Education) to be used across the entire journal; however, custom groupings can be created.
It should also be noted that a number of longstanding limitations inherent in the current Reading Tools implementation have been identified, and that an overhaul of the entire framework is slated to begin concurrently with annotation tool development. A full description of the prospective overhaul is outside the scope of this paper, but in short: the Reading Tools framework will be refactored into a plugin component, where each specific tool is its own plugin; the interface will be redesigned to reduce or eliminate frames-based display and facilitate the embedding of tools within articles; and the tools allow for full multilingual capacity, similar in function to the rest of OJS.
A number of new tools have been proposed for future development alongside the comprehensive overhaul briefly described above. Most of this proposed functionality is informed by research undertaken by Rick Kopak and Chia-Ning Chiang at the University of British Columbia (UBC), and also by research at the ETCL under the directorship of Ray Siemens. Most of these new tools contain, at their most complete levels, social networking elements, although basic elements of some tools may be implemented immediately, with social aspects to follow in a later development phase.
Ray Siemens et al. identify three “exemplary tasks” associated with humanities computing (and by extension, online scholarly pursuits): “(1) the representation of archival materials; (2) analysis or critical inquiry originating in those materials; and (3) the communication of the results of these tasks” (“Prototyping” 9). If the first point can be seen to correspond with the publishing within OJS of an article, the core OJS system executes the first task; the current Reading Tools execute the second, to a degree that could be improved upon; and the third is largely unsupported save for the three useful, yet insufficient 'social' tools mentioned above (Notify Colleague; Email the Author; and Add Comment).
Furthermore, Siemens et al. identify three strategies readers use as part of their research, yet which are insufficiently supported by any single social networking tool at the moment: creating an online identity by which one's work and commentary can be evaluated against another's (and through which one's scholarly reputation can be enhanced); allowing for clear connections between these identities to furnish communication, which can then be evaluated; and the management of research material, especially in a social way (“Prototyping” 28-30). These three combined aspects speak most directly to that third point above, yet they are also inherent to the development of the first and second points as well. That is, social discourse in and of itself is an integral part to further research, which in turn results in published scholarly work (“archival materials”).
Kopak and Chiang have identified social annotation tools as a primary consideration for further scholarly communications development, and have furthermore described what such tools should look like, and how they should act and interoperate. Their survey of twenty-plus existing note and annotation tools defined four major design element categories: highlighting, notemaking, linking and workspace, of which the first three have been identified as “an aspect of annotation rather than an expression of annotation in a more complete sense” (“From Retrieval to Interaction” 1). The fourth 'workspace' element isn't an annotation element itself: rather, it is seen as a composition space supportive of writing notes and annotations (“From Retrieval to Interaction” 1), and in the context of OJS will likely be developed outside the reading tools proper. However, the Reading Tools' proposed annotation tool will rely on it heavily, and so it is described along with the others below:
- Highlighting : Highlighting and underlining text in different colours for different purposes. Highlights should be overlappable, especially in consideration of future social annotations ("From Retrieval to Interaction" 1; 4-16).
- Notemaking : The notemaking function will allow readers to attach a text note to the article – specifically to an anchor (eg. a bookmark), or alternatively, to the entire document. Notes should be taggable, and should be sortable by tag and/or colour of the attached highlight or underline ("From Retrieval to Interaction" 1; 16-34).
- Linking : The linking function will allow readers to interlink notes and bookmarks within a document to other items within the document; to other items within the journal Web site; or possibly to other items across the internet. These links may optionally also describe the relationship between items ("From Retrieval to Interaction" 1; 44-48).
- Workspace : The workspace function will give readers an area within OJS to create, organize, publish and delete annotations and bookmarks, and in which to see the relationships and interactions between one's own items and those of other readers ("From Retrieval to Interaction" 35-44).
Kopak's proposed annotation design corresponds neatly with Siemens' description of an ideal scholarly social networking environment: it gives readers an environment and tools in which to promote their academic reputation, manage their research and transform reading into knowledge; and allows for relationships between readers, researchers, authors and even editorial staff to be identified and nourished via linking and especially the linking of annotations with comments etc. The implications of this at a 'local', journal/site level alone are interesting; to think of extending the reach of these tools across the existing 3,000 OJS journals (“Journals by Continent”) and beyond is staggering.
Redevelopment of the OJS Reading Tools will be split into phases to make development tasks more clear and manageable. While this article provides no firm (or even estimated) completion dates for these phases, dates will be decided upon shortly and will conform to the duration of the SSHRC-funded Synergies Project, of which this development partnership is a subset.
Phase I will serve to familiarize the UVic development team, consisting primarily of Michael Joyce, with the OJS code,  and will consist almost entirely with the initial implementation of annotation and marginalia tools. This implementation will be based primarily on Geof Glass' Marginalia tool and Kopak and Chiang's annotation design proposal (“From Retrieval to Interaction”). It should be noted that while Glass' Marginalia tool was extended to work with OJS as a plugin, thanks to the SSHRC-funded “Navigating Information Spaces” project and under the direction of Kopak, the plugin code as it currently stands is quite hefty, and deemed by the ETCL to be easier to replicate than to integrate as a maintained OJS plugin.
During this first phase, the annotation tools will be implemented at their most basic form: readers will be able to write, save, edit and read their own annotations across a journal (or multi-journal install), and will be able to bookmark elements of a document (or the document itself); but will not be able to share their notes or bookmarks in any fashion. Additionally, a clear workspace will not yet be in place: as in the current Marginalia plugin, the annotation tools will only be available from the article itself.
Phase II development will focus on the extension of current tools into a comprehensive social network at the local (individual journal or site) level. Readers will be able to publish their annotations and bookmarks for other readers to see, and will be able to link (and describe relationships) between published items. Readers will also be able to tag their annotations, providing an additional contextual bridge between information.
A basic workspace will also be developed during phase II. This workspace will be available from a reader's User Home page, and would list all bookmarks and notes, as well as contextual information for those items (for example, whether they have been made public; whether they are linked to; etc.). The workspace will also show common tags used by the reader, and indicate whether and where other readers are using the same tags; similarly, it will show incoming links from other readers. As such, the workspace will act as a central communications hub between readers.  The accumulation of [a] the ability to annotate “archival materials”; [b] interpret and subsequently manage the relationships between one's notes and others, goes a long way towards satisfying the broad outline of a full-fledged academic social networking tool as defined by Siemens, while also vastly improving Willinsky's original vision of providing a contextual framework from which to further research.
While the details of Phase III have yet to be firmed up, they will most likely focus on further Reading Tools refinement, the workspace environment, and further social networking.
For example, it is worth noting the newly-announced, Mellon-funded Open Annotation Collaboration (OAC) project, which seeks to develop an online annotation specification (or set of specifications) for sharing annotations across the Web and between diverse applications and clients. The project will dedicate their first development phase to defining a digital annotation framework, and to create an environment where the framework will thrive. This and similar projects will be carefully tracked by the UVic/PKP collaboration, and implemented if it makes sense to do so.
It is also not beyond the realm of possibility (although perhaps beyond the realm of Phase III and Synergies funding) to develop a client-server software arrangement similar to that between the PKP Harvester and OJS: in other words, a standalone user/annotation server system to tie together disparate but sympathetic scholarly systems.
One other area that demands further thought is the workspace. After a basic workspace area has been developed, it can serve not only to help the reader with the research process, but, in a nascent way, to help with the writing process: thoughts and notes are collected and can be organized in one easy-to-manage interface. This can be further extended, with additional tools providing a 'document pool' or bookshelf for notable works; clipping services to collect fragments of text and more from various resources; comprehensive document creation and editing functions further enriched by these tools; and a publication workflow to move documents from private to public. Thus, the workspace (and the associated Reading Tools) would not only assist in scholarly, professional and citizen academics in their research, but also assist in their subsequent organizing of that collected wisdom into a tangible, valuable output – ready to be fed right back into the research pool.
There will be significant challenges in implementing what are at the moment still very theoretical ideas on social networking and workspace development as applied to OJS and the other PKP applications. The most interesting and perhaps largest technical challenge may revolve around privacy, ownership and access to information rights, issues that have and doubtless will continue to trouble social networking sites all across the Web. Security also becomes a much larger issue when user information sharing is considered. These are integral issues when developing an open, collaborative environment, and will be carefully considered on the way to, within, and beyond Phase III. But for the moment, we have a clear path ahead: evolutionary steps to improve what will continue to be a revolutionary tool, based on clear research and a sound design plan.
“Journals Using Open Journal Systems by Continent.” Public Knowledge Project. 2 April 2009. Simon Fraser University Libraries: Public Knowledge Project. Web. 10 Aug. 2009 < http://pkp.sfu.ca/ojs-geog>.
Kopak, Rick, and Chia-Ning Chiang. “An Interactive Reading Environment for Online Scholarly Journals: the Open Journal Systems Reading Tools.” OCLC Systems & Services 25.2 (2009): 114-24. Print; rpt. Web. <http://pkp.sfu.ca/node/2065 >.
––––. “From Retrieval to Interaction: Recommendations for Annotation and Linking Tools in OJS.” Forthcoming.
“Marginalia Web Annotation.” geof.net. Web. 10 Aug. 2009 <http://www.geof.net/code/annotation >.
Open Annotation Collaboration. Web. 10 Aug. 2009 < http://www.openannotation.org/>.
“Open Archives Harvester.” Public Knowledge Project. Simon Fraser University Libraries: Public Knowledge Project. 2005. Web. 10 Aug. 2009 < http://pkp.sfu.ca/?q=harvester>.
“Open Journal Systems.” Public Knowledge Project. Simon Fraser University Libraries: Public Knowledge Project. 2009. Web 10 Aug. 2009 < http://pkp.sfu.ca/?q=ojs >.
Siemens, Rick, Cara Leitch, Analisa Blake, Karin Armstrong, and John Willinsky. “'It May Change My Understanding of the Field': Understanding Reading Tools for Scholars and Professional Readers.” Forthcoming.
Siemens, Ray, Brett Hirsch, et al. “Prototyping the Renaissance English Knowledgebase (REKn) and Professional Reading Environment (PreE): Past, Present, Futures.” Forthcoming.
Willinsky, John. “Open access: Reading (research) in the age of information.” 51st National Reading Conference. 2003. Web. < http://pkp.sfu.ca/files/NRC_Galley.pdf >.
 As a sort of pre-Phase I development exercise, Michael Joyce has developed an addThis sharing tool option ( http://addthis.com), currently in CVS, which is configurable from the journal manager's Reading Tools management page. It places a “Share This” button at the bottom of an HTML article: readers can click the button and choose to share a link to the article, along with any comments, with sites such as Facebook, Slashdot, Digg, etc.
 While this workspace area is invaluable for users who want to work with these annotation tools, as a development effort it does fall outside the Reading Tools purview and will likely not be developed as a Reading Tools-specific plugin.