NOTE: This proposal is the version that was attached to the contract with MINITEX for the MDL SSR project. To how this actually turned out, please see the report. Thanks! ...Eric

The Minnesota Reflections site brings together a rich set of resources from institutions across the state. Each institution must not only provide images and documents, but also the metadata to describe these items. Providing even basic metadata is a challenge, full descriptions of each item are nearly impossible to generate. For a number of years planners of the Minnesota Digital Library Coalition (MDLC) have wondered whether it would not be possible to open up the description of items to public input. This project explores this "social side" of Minnesota Reflections, public input of item descriptions.

The first question we have is whether the public is interested in contributing descriptive information to the MDL. Individual interactions the planners have had with members of the public at the state fair and in other settings leads us to believe there are people out there who would invest time if the doors were open to such contributions. Some other Minnesota explorations of this kind of input, such as Hennepin County Library's Bookspace and Minneapolis Public Library's teen "make your own list," have already met with great success. This project will take the first steps toward demonstrating this interest in the MDL community. At this stage we only need a simple interface for quick annotation. If such a simple system is used, that would justify MDL investment in a more sophisticated publicly accessible annotation system.

I propose that we use wiki technology as the basis of a simple annotation system. This wiki would be completely separate from the main MDL database and only linked to entries of that system through the use of common item identifiers. In other words, we would use the same image ids on the commentary wiki as Reflections uses internally in its database. By keeping the commentary separate from the main database we would eliminate any potential security risks of a system open to the public in some way compromising the existing professionally populated database.

A simple link could be provided from Reflections that takes users directly to the commentary system and allows them to add comments for the item they were inspecting. If desired, the Reflections system could also display comments from the wiki without having to add any of that data back into its own indexes. Of course, if we would want to include public comments in the searchable index of Reflections, we could later, after review, harvest this data.

The risks of building such a system are relatively low, partly because it would be relatively inexpensive as well as well isolated from the MDL's core systems. Creating a commentary wiki will require system development and administration expertise, choices about wiki technology, integration with the current Reflections database, and web hosting which could be provided by the University of Minnesota or another party. The remainder of this proposal lays out the details of these components.

Project Description

The Social Side of Reflections (SSR) will be a wiki that provides a page for each item in the MDL collection. These wiki pages will be editable by anyone in the world who registers with the SSR site. This wiki, unlike the Wikipedia, will not accept completely anonymous contributions.

Any registered user will be able to edit the whole page associated with an item. In other words, the description space in the SSR wiki will be shared space. Only textual content will be allowed on the wiki pages, users will not, for example, be able to upload additional images. Even most HTML will be prohibited on the SSR wiki. The goal will be to keep the content as simple as possible in case it needs to be transported to some more sophisticated system in future MDL iterations.

The MDL will add links to the SSR wiki to each detailed item record in Minnesota Reflections. These links will take users directly to the commentary space and allow them to either register or add comments. It will also be possible for the Reflections system to poll the SSR wiki for the current SSR content for a given item. MDL planners may or may not use this capacity to display SSR content within Reflections.

Project Audience

Anyone in the world may have information about an MDL image that they would like to contribute. While the SSR wiki will be open to the world, we expect that the primary audience will be citizens of Minnesota. In fact, our early adopters are likely to be staff of the very historical societies and cultural institutions that have contributed content to the MDL collection. The system should be designed for non-professional use, with simple instructions and quick rewards for contributions. The most important quick reward will be that a user's contribution must become immediately visible in the SSR system.

Project Approach

Setting up the SSR wiki will require some planning and coordination with the MDL technical team. We must make sure that the team has some level of comfort with the wiki software chosen, so that they could maintain the system beyond the conclusion of the project. The MDL technical team must also be prepared to develop the SSR page link within the Reflections system and the format of that link must be agreed upon. Finally, the MDL team will likely want control of the server hosting the SSR wiki, and so must be fully aware of the specifications and requirements of such a server. By the conclusion of the Planning Phase of the project a wiki engine and server will have been selected.

The Development Phase of the project will proceed on two parallel tracks. On one track the MDL technical team will develop the item-level link to the SSR wiki. The other track will be the focus of this project: the development of the SSR wiki system itself. For the most part this will require little interaction from MDL planners until a test system is made available. However, the wiki will be developed in a location fully accessible to MDL planners and input and feedback on progress will be welcome at any time. Once a working beta of the system is ready, MDL planners will be invited to test it and offer requests for changes and enhancements. After a few weeks of testing and iteration, the Development Phase will conclude.

During the Deployment Phase the SSR wiki will be available for public input. This project will focus on troubleshooting any issues that come up and making sure the wiki operates as expected. The MDL planners may want to publicize the availability of the commentary system in some way, though the most interesting question will be whether users of Minnesota Reflections notice and use the commentary link without special prompting. Note that even though the project will enter a Review Phase, we expect that the commentary system will remain deployed and operational from here on out.

After a few months of operation, the project will enter a Review Phase in which assessment and reporting are the main objectives. Assessment will include at least a summary of usage statistics comparing Reflections usage with SSR usage and describing overall SSR contributions.

Project Assumptions

The consultant for the SSR wiki will need complete access, including shell access, to whatever server hosts the SSR wiki. This access should include the ability to install cgi scripts, php programs, MySQL databases, and other tools that are typically part of wiki engines. Details of exactly what tools will be required will be discussed with the MDL technical team during the Planning Phase.

The MDL technical team will be responsible for any work in Reflections to accommodate the SSR wiki URL on each item record.

The SSR wiki will not accept completely anonymous contributions. In part this is so that any abusive use of the site can be blocked. Therefore some sort of user registration will be required. One possibility is using the University of Minnesota's "guest account" system as the mechanism for users to get accounts for the SSR wiki. Other options can be considered during the Planning Phase.

The SSR wiki created by this project is not intended to be the last word in the application of social networking technologies to Minnesota Reflections or other MDL efforts. Instead it is a first step that can help us understand the promise and challenges of opening to our users in this way. Further exploration of other options, such as an integration of MDL collections with the Flickr social photo network, while outside the scope of this project, will be vital for the MDL.

Proposed Project Timeline

Planning for this project would take place during the first three weeks of November 2007, with a web server procured and available by 23 November. Development of the commentary site could begin 19 November, even before the server is available, and will continue through 31 December, including a few weeks for testing and revision. Availability of the publicly accessible site would begin in January with a period for assessment of the project starting 25 February. A report the the MDL will be provided by the end of March 2008.

Task

November

December

January

February

March

Project Management

CRS

Planning Phase

Project Planning

CRS

Procure Server Space

Development Phase

Reflections Link

System Development

Testing

Deployment Phase

Operation

Review Phase

Assessment

Reporting

Items listed as "C" above are the responsibility of the consultant. Items listed as "R" and "S" fall to the MDL technical team. The "R" tasks are those that must be completed by staff maintaining the CONTENTdm/Reflections system. The "S" tasks can be done by whatever staff will manage the SSR server long-term, perhaps staff of MINITEX since it serves as the umbrella under which MDL operates. Project planning and operation will be jointly managed and are listed at "CRS" or "CS" above.

Proposed Project Budget

The complete project will require $15,000. I believe we could take advantage of the web services of the Office of Information Technology at the University of Minnesota. These services are provided at minimal cost to MINITEX.

Deliverables

Phase

Consultant Cost

MDL Cost

Project Management

Planning

$ 500

Consultation Costs

Planning

2,000

Project Management

Development

800

200

System Administration

Development

500

Software Installation

Development

1,000

Modification to Reflections

Development

1,000

Wiki Programming

Development

4,700

Graphic Design

Development

800

Project Management

Deployment

400

100

System Administration

Deployment

1,000

Project Management

Review

400

100

Consultation Costs

Review

1,000

TOTAL

$ 13,100

$ 1,900

The MDL Costs are estimates and would be payable in whatever way the MDL handles internal costs. They are intended to cover the costs of managing the changes to and interactions with the current CONTENTdm/Reflections system.

The Consultant Costs will be invoiced as follows: Planning Phase costs will be invoiced once a choice of wiki software has been made and agreed to by the MDL (likely November 2007), Development Phase costs will be invoiced once a commentary system is made available for MDL testing and feedback (likely December 2007), and Deployment and Review costs will be invoiced when a final report is presented to the MDL (likely March 2008).

Conclusion

A wiki-based exploration of the Social Side of Reflections provides planners with a low-risk assessment of the interest and value of community contributed descriptions for MDL items. A positive outcome to this project would indicate that further investments in more sophisticated commentary mechanisms are warranted and should be built into future grants.

As a former member of the MDL planning team and a developer of wiki sites for over three years, I would be eager to devote energy to this task. I look forward to working with the MDL on this project.

Eric Celeste

Eric brings over 15 years of library and 25 years of technology experience to his consulting. At MIT Eric shepherded the creation of DSpace, open source digital repository management software developed with HP and now deployed at hundreds of institutions worldwide. At the University of Minnesota Libraries he encouraged the development of the UThink blog service, a wiki-based staff intranet, LibData, and the University Digital Conservancy. He works with non-profit institutions on appropriate uses of technology for informing, communicating, and collaborating with their constituencies.