Along with the generally positive response indicated above, my core reactions can therefore be distilled as follows:

At the conference it would be good to explore the first of these in detail, and in particular to hear how the different components of the project will be funded at one of three levels of effort and expenditure:

As for the second point listed above, I have reached the conclusion that the approach to annotation taken in the Online Chopin Variorum Edition project (see the description in Annex 2) may provide EVIA with a useful model. First of all, there are several types of scholarly metadata in our variorum edition: “Overviews,” “Source Descriptions,” discussion of “Key Features” and detailed “Bar-level Commentary.” As noted below, the “scholarly material presented in the resource is meant to be instructive and indicative rather than fully comprehensive,” an approach which we consider to be “more consistent with the aims of the project in general, i.e., the creation of a flexible ‘dynamic edition’ produced not by a fixed body of editors but rather through an individual’s creative interaction with the constituent sources” (231). It is my belief that this sort of selectivity would work well in EVIA, even if its aims and fundamental nature are quite different from those of OCVE. Not only would the inclusion of representative rather than comprehensive annotation content provide a convenient and (in my opinion) much-needed solution to the problems of information overload, peer review, and so on alluded to above, but it would also allow more of the available funding to be channeled toward EVIA’s primary mission, namely, the preservation of video content. Having a body of “core” (i.e., annotated) materials alongside a range of other video content without annotations would of course result in structural inconsistency, but, as in OCVE, this would or at least could be purposeful rather than a weakness.

Another OCVE feature of possible relevance to the EVIA team has to do with meta-annotations, which we refer to as “personal annotations” to distinguish them from the scholarly commentary. The description in Annex 2 indicates how these are fashioned. There are several points to stress in connection with EVIA:

In the future, we might try to develop a “music-editing forum” (see Annex 2), and, subject to the availability of funding, EVIA’s Summer Institute model would be an excellent one to adopt in this respect. For now, however, we regard the lack of monitoring/moderation as potentially unproblematic, although an eye will be kept on the material as it evolves to determine whether or not this policy is sensible. It goes without saying that the different nature of the material within EVIA may require a different means of presenting meta-annotations, but I would encourage the project team to consider a “light touch” approach at least at a pilot stage, provided that the segregation of material I have referred to is strictly maintained.

One final issue arising from the EVIA report and the project in general has to do with an area of research dogged by controversy during the past decade and a half: so-called “practice-led research,” also known as “practice-based research.” Annex 3 sets out relevant material from the Research Funding Guide of the Arts and Humanities Research Council in the UK, which has been an ardent supporter of practice-led research for many years (e.g., in the form of a Creative and Performing Arts Fellowship program, as well as a practice-led route within the Research Grants scheme). One of the main obstacles within this field has been the reluctance or inability of potential practice-led researchers to produce documentation that appropriately and effectively demonstrates the research content of their creative activity. As the AHRC Guide notes, “Work that results purely from the creative or professional development of an artist, however distinguished, is unlikely to fulfill the requirements of research,” for which the following must instead be satisfied:

Possibly the thorniest problem thus far has been to encourage practitioners interested in this form of research to produce documentation alongside and in addition to the creative output itself, describing the research process and identifying the conclusions reached in respect of the basic research questions underlying the endeavor. As stated in Annex 3, this form of “documentation, analysis, and reflection must be an integral part of the project,” leading to outputs that “can go beyond more traditional academic papers and can include such forms as journals or diaries; documentation on a website, CDs or DVDs, etc” (234)

In my opinion, EVIA’s annotation methodology offers an ideal solution to at least some of the problems encountered within this area. By extending to those carrying out practice-led research a new means of documenting the research as it is happening, of undertaking self-reflective analysis of the creative outputs, and of presenting the research findings in a manner that recognizes and reflects music’s time-dependency, the annotation methodology if applied to this field of work could fundamentally revolutionize how it is done and what it represents to those within and outside the practice-led arena. Here the annotators would not be “collectors” of material but the generators of that material, i.e., creative practitioners themselves. I see this as a potentially exciting spin-off from EVIA that may not have been anticipated by the project team themselves.

In closing, and by way of comparison, consider the Practice as Research in Music Online website described in Annex 4. This “cumulative research archive,” developed by the Institute of Musical Research in London, presents “full-length and excerpted rehearsals, workshops, performances, and demonstrations of various kinds” (234-5). The textual component is limited, however, to “a description and abstract giving a summary of the item's content and an insight into its contribution to current research” (234-5). Without wishing to dismiss the PRIMO initiative altogether, I cannot help but regard it as a missed, or at least unrealized, opportunity for presenting the kind of documentation, analysis, and self-reflection referred to above. In a nutshell, what one gets in PRIMO is footage plus a few paragraphs; but what could be produced thanks to the EVIA annotation methodology is a rich resource along the lines of EVIA’s own annotated video content, though used to different ends. Such a development would be highly significant to the body of creative practitioners whose working methods to date have been at odds with what funders like the AHRC require, and who, as a result, have not had the opportunity to share in an intellectually convincing and technically feasible manner the kinds of insights that underlie their artistic endeavors and the research insights that either feed into or arise out of them.

Some of the key points in Alan Burdette’s paper are summarized here for ease of reference.

Since its inception in 2001, EVIA’s primary mission has been “to preserve ethnographic field video created by scholars as part of their research,” and its secondary mission has been “to make those materials available in conjunction with rich, descriptive annotations,” thereby creating “a unique resource for scholars, instructors, and students.” This has required the building of tools and infrastructure to preserve and document recordings and “to make them part of the scholarly enterprise through a unique form of peer-reviewed online publication.” By its very nature the work has been highly collaborative. To date, seven collections have been made available online, comprising seventy hours of annotated video, with a further 1,200 hours “in various stages of completion.” All of this has required “significant investments” in funding and effort.

Collection development has occurred in two main ways:

Development has also resulted from projects collaborating with EVIA “for preservation and access services”; exceptionally, collections in this category “are not peer reviewed,” although some portions may be “more highly annotated and then peer-reviewed.”

The use of peer review for video annotations is one of EVIA’s distinctive features and is intended to create both “a stand-alone publication” and “a resource that can be used in conjunction with other print or online materials.” The annotation process involves the following stages:

The annotations can be extensive, equivalent in some cases to a “small monograph.” The EVIA project’s vetting processes require significant up-front investment of time and money prior to completion of an author’s collection. Moreover, “preservation transfers, video transcending, file and data management, training, the summer institute fellowship, and peer review management all consume a great deal of money.” There is doubt about the perceived value of the published work to the scholarly community at large because of general problems with the ways in which “scholars are assessed and rewarded,” not to mention the inferior status sometimes accorded to online publications. Nevertheless, the inclusion of EVIA project work in certain tenure and promotion dossiers has had positive results.

The software tools developed within EVIA include the Annotator's Workbench and the Online Search and Browse Tool, along with smaller-scale applications for “controlled vocabulary maintenance,” technical metadata collection, etc. The latter are intended to serve other disciplines on an open-source basis.

The report describes numerous challenges to do with sustainability, understood by the EVIA team in terms of “preservation, access, publishing, collection development, infrastructure, and funding strategies.” Each of these must be addressed “if a project is to successfully move from being just a project to being a trusted resource, a dynamic social space, or a viable channel for scholarly communication.” Apparently a “means of sustaining the project” has been created, although few details of the funding strategy are given.

The following observations do, however, appear:

Alongside these sustainability challenges, EVIA faces conceptual dilemmas which potentially “will have an important impact on the continued success of the project” while also shaping its fundamental aims and objectives. These include the following:

The report ends by noting that EVIA’s “software platform” and “suite of services” were designed to meet “pressing needs within a core group of ethnographic disciplines,” but that these are also intended to satisfy a “wide range” of disciplinary requirements.

This description of the OCVE project has been taken from the Final Reports on both the pilot study and the first developmental work phase. Further details can be found at www.ocve.org.uk.

The first developmental phase of the Online Chopin Variorum Edition (OCVE) was funded by the Andrew W. Mellon Foundation from November 2005 to September 2009. It followed on from an eighteen-month pilot study also funded by Mellon from May 2003 to October 2004. In both of these work phases OCVE’s principal aim was to facilitate and enhance comparative analyses of disparate types of musical source material, attaining a level of manipulability outstripping that manifested in extant printed editions of Chopin’s music and indeed of any composer to date. The research exploited emerging technical capacities for text/image comparison as well as recent musicological advances in cognate projects such as Chopin’s First Editions Online (CFEO; www.cfeo.org.uk), funded by the Arts and Humanities Research Council from March 2004 to September 2007, and the Annotated Catalogue of Chopin’s First Editions.2

The chief priorities of the first developmental phase (hereafter referred to as Phase 1) were to extend OCVE’s content and to build the technical tools and frameworks for the display and manipulation of that content. The end result is a new type of “dynamic edition.” Users themselves have the ability to construct a unique “edition” of their own, combining elements from the constituent source materials and consulting the scholarly apparatus that we have provided for the sake of greater insight and understanding. In addition, there are tools for adding personal annotations and thus for creating flexible, idiosyncratic “Critical Commentaries.” The emergent system is intended not only to facilitate research on musical sources but also to encourage wider modes of comparison and the reconstruction of creative histories to an extent which could not be easily achieved outside a digital environment.

One of the recommendations made by external participants at early project workshops was the inclusion of more scholarly content than had originally been envisaged, in the form of detailed commentary on the sources themselves, on the philological significance of the variants revealed through the juxtaposition of sources, and on the interpretive issues arising from those variants. The project team incorporated this suggestion within the pilot itself, whereas other recommendations were addressed for the first time in Phase 1, including the implementation of personal annotation tools.

Scholarly commentary written by members of the OCVE team exists at three levels within the Phase 1 resource. First, each work has an “Overview” section which describes the general character and provenance of the individual sources relevant to it. Second, each witness has a more detailed “Source Description” including catalogue metadata, which is stored in the database in a structured XML format which could form the basis of a structured search at a later date. Finally, descriptions of “Key Features” and detailed “Bar-level Commentary” text devised for nominated works within the Phase 1 resource have been entered into the newly developed annotation system. In addition to highlighting salient details of a given source, the “Key Features” text typically provides relevant background information to it. Reference may be made under this heading to other sources (whether or not they appear in OCVE) for the sake of comparison. Significant modifications and errors will usually be highlighted; however, in no case is the discussion exhaustive, nor—as noted above—are “Key Features” identified for all OCVE sources. Instead, the (extensive) scholarly material presented in the resource is meant to be instructive and indicative rather than fully comprehensive; this intentionally selective approach is more consistent with the aims of the project in general, i.e., the creation of a flexible “dynamic edition” produced not by a fixed body of editors but rather through an individual’s creative interaction with the constituent sources.

The new annotation system allows the bar-level comments to be associated with single bars or bar ranges across single or multiple sources, and so represents a logical development of the annotation system developed in the pilot study. However, the level of sophistication is now greater, likewise the potential of this material to serve as a model for individual users in constructing their own “Critical Commentaries” in the form of personal and/or public annotations (see below). All of the individual scholarly comments pertaining to a given source can also be viewed in an aggregate form approximating a standard “Critical Commentary.”

It should be noted that OCVE’s editorial approach is neutral in respect to the available sources, in that these are presented without qualitative judgments being made in terms of hierarchy or respective pre-eminence. As OCVE is not presenting a single version of a music text, individual editorial decisions of this kind are not required. We do, however, report on relevant variants and corrections of errors and omissions in one or more sources, and typically the user is invited to compare one source with another when looking at a set of bar-level comments.

Personal annotations are fashioned, at a low level, upon the model developed for the OCVE pilot study. Registration is required in order for end users to be able to create annotations, which can either be publicly visible (i.e., available to any user of the web resource, whether registered or not) or private (i.e., visible only to the user-creator). The annotations themselves make use of a standard client-side, rich-text editor which allows the user to express basic formatting and structure in their annotation (including the creation of links), which is then stored as XHTML in the OCVE database. Each annotation can additionally be given a title, and each user’s “My OCVE” page offers a quick overview of the annotations created within the system.

One of our key aims in developing the web-based annotation mechanism in Phase 1 was to ensure that the process of creating annotations was as cognitively undemanding as possible,3 and primarily for this reason we decided to make annotations attachable at the level of the bar only (rather than to specific coordinates). This allowed us to solve a further problem: the development of an interaction model sufficiently expressive to allow users easily to select multiple sources to which they want to apply an annotation. Users simply click the bar images of the sources to select those to which they wish their annotations to be relevant (or click a second time to unselect).

A tension between accessibility and authority is evident in respect to these meta-annotations. If a free, online resource enables users to “create their own editions” and integrate their own comments in a dynamic edition environment, it also requires an effective system for dealing with comments, changes and annotations. Self-policing works most effectively in popular, large-scale fora: given that only a relatively small group of scholars and music professionals might take part in a music-editing forum, the monitoring of comments along the lines of Wikipedia might prove difficult and, ultimately, unsustainable. But if such comments are clearly identified as separate from the core scholarly resource, a lack of monitoring/moderation is not necessarily problematic.

The following text has been adapted from the AHRC’s Research Funding Guide (www.ahrc.ac.uk/FundingOpportunities/Documents/Research%20Funding%20Guide.pdf).

The AHRC provides funding for research:

Research of this kind should:

Work that results purely from the creative or professional development of an artist, however distinguished, is unlikely to fulfill the requirements of research. [p. 64]

For practice-led projects, whilst creative output may be produced and practice undertaken as an integral part of the research process, the Council would expect this practice to be accompanied by some form of documentation of the research process, as well as some form of textual analysis or explanation to support its position and to demonstrate critical reflection. This documentation, analysis, and reflection must be an integral part of the project and must be carried out during the award period. These outputs can go beyond more traditional academic papers and can include such forms as journals or diaries, documentation on a website, CDs or DVDs, etc. A clear rationale for the appropriateness of the form of the Principal Investigator’s critical reflection should be provided. [p. 19]

This text comes from the website of PRIMO at http://primo.sas.ac.uk.

PRIMO describes itself as “a new platform for musical research in sound and vision” and as “a cumulative research archive.” Designed for research, study, and teaching, it presents full-length and excerpted rehearsals, workshops, performances, and demonstrations of various kinds. Some files are as short as five minutes; others last over an hour (these are segmented for ease of use). All are self-sufficient pieces of research, even when they represent a phase of a longer- term project. Some files will contain explanatory texts, but the point of the repository is to capture sonic events. Each item is accompanied by a description and abstract giving a summary of the item's content and an insight into its contribution to current research. PRIMO is subscription-free, peer-reviewed, and managed on behalf of the community by the Institute of Musical Research.

PRIMO aims to provide:

Users of PRIMO are likely to be those who have an interest in composition and performance as research processes; those who study the ergonomics, psychology, ritual play or ethnography of performance; and those interested in performance practice. PRIMO welcomes research involving disciplines other than music with the proviso that musical research questions must lie at the core of the submission. It equally welcomes single, self-standing items and series of items documenting research processes.