Primary Applicant: Jason Fleming

Secondary Applicant: Michael Jay

 

Acquisitions and Licensing Department

 

UF Library Tagging Project

 

Funds Requested: $5000

Cost Share $500

 

 

 

Narrative

The objective is to create a functional service that is easy to use and would allow students, staff and faculty at the University of Florida to add to, and draw from a database of stored information. The resulting database will be a warehouse of bookmarks generated by the user. These bookmarks will link users to resources they have previously discovered from using the Library catalog, purchased resources, or research materials from the internet. Information will be organized by USER ID and by assigned Tags that identify the information for the user and others. Keywords assigned to bookmarks are considered tags that serve as index points into the list of saved resources.

 

Several types of similar projects are available for inspection on the internet. Commercial applications include such websites as del.icio.us and Flickr. The closest example in the academic community can be viewed at PennTags (http://tags.library.upenn.edu/). The PennTags project allows users to tag, or bookmark instances of information in the catalog and on the internet. The UF Tagging Project would perform a similar function, but would have some added features.

 

Most tagging implementations allow the user to only tag items on one level. I am proposing a system that would allow the user to create two sets of tags for any given resource. The first set would allow them to create social tags that are designed to allow others to find and use the same information. These keywords would describe the object in such a way that it would be evident to other users what the underlying concept is. The other set of tags that could be associated with the item would be personal tags. These “personal tags” would allow the user free reign to create a new way to organize the information, and would not be shared. Examples of this kind of tag include “toread”, “free”, or involve the use of special characters.

 

A controlled vocabulary will be put in place to monitor the social tags. If tags are not contained within the controlled vocabulary the user will have an opportunity to decide to temporarily add them pending review, or to tag them as personal. Social tags can be viewed by the entire community, but personal tags can only be viewed by the user. If a user wishes to keep a tag private then they will tag it as personal.

 

Tagging systems allow us to create new tools to aid users with their resource discovery tasks, or to help users organize the information they have amassed. Tag Clouds allow users to see at a glance what subjects are popular in the community. This can be accomplished by increasing the font size of a tag as the number of instances increases. An Interest Tracker is a timeline that would allow users to see visually what keywords they have been using, and with what frequency since they started using the tagging system.

 

A reputation scale will be employed to give users an additional means to decide whether a resource is valuable or not. A user’s reputation will fluctuate as other users rate them by what they have tagged.  If another user finds the tagged item useful then the original poster’s rating will increase and vice-versa.  Librarians will start out as power recommenders giving their tagged resources a greater degree of authority in the system.

 

To increase usefulness, links to citations and proxy services will be employed as much as possible. The user should be able to easily migrate to resources no matter where they are. They should also be able to keep track of citations they use when writing papers or working on projects.

 

This project is aimed at fulfilling the library’s goal to simplify the search and discovery of library resources. Users will have a tool at their disposal to keep track of the items they have already discovered, and a means to find out what other users have noted as worth looking at. The interest tracking feature will guide users through their past search habits to help them keep on track with how their research has changed over time. The Tag Cloud feature will help them see what topics they find most relevant to their research and what the University as a community is studying.

 

This project would also increase the visibility of the libraries for the UF community. Users will see what a valuable resource the library has made available to them, and this will also increase their use of the resources available because they will be more exposed.

 

The Tagging system will enable the community of users to more effectively manage library provided resources. This grant will give us the opportunity to strengthen our programming skills and an opportunity to network with the creators of other tagging projects. As a result of this funding we will have the time and resources to follow through with a plan to implement this service for the UF community by next fall.

 

Resources Needed

Michael Jay and I will be working on the development of this project. We will require assistance from Systems as needed during this phase. We will need the help of library employees during testing phases to ensure that the product is ready for the UF community. We will call upon the Marketing Committee when the time is right to get inform users of the available tool.

 

Currently we have a Linux server from Systems, and a KVM switch is being ordered to interface with my desktop. We may need additional software and hardware items as found necessary. The developed tagging systems that we will first examine are open source and freely modifiable.

 

Travel will be necessary to meet with representatives involved in other tagging projects to get ideas and keep up to date. Travel money will be needed to attend training events and conferences that will be useful.

 

Timeline

November 2006

Install products and examine them to see what elements are useful for the project goals.

• FreeTag
• FarFetch

Initiate talks with PennTags project coordinators to see what assistance they might be able to provide.

Install KVM switch to desktop server

Find out what training opportunities are available in the area

 

December 2006

Find out what programming language we should focus on

Find training opportunities to learn necessary aspects

 

January 2007

Start Intensive training with programming language

Begin preliminary development of codebase

Start working on mechanics that will allow Gator One Ids to be used as primary key identifiers in database.

Work on Login and interface.

Develop web tools that will allow users to use system

 

February 2007

Develop API (Application Program Interface) to allow power users to call information from system.

Present preliminary findings to library community

 

March 2007

Work on integration of services

Intensive testing to ensure that objectives are being met and rollout will be successful.

 

April 2007

Finish Beta product and perform quiet launch to ensure that the service will not crash

Work on last minute concerns that might hamper staff participation

 

May 2007

Roll out soft launch of product to Library Staff

Give presentations demonstrating the product and explaining how to make it work

Start to get review system in place for controlled vocabulary

 

June 2007

Data from library users should give us a base to work with and develop new aspects of system such as the tag cloud feature.

Users will be solicited individually for ideas.

Formal survey will be developed to send out to users

 

July 2007

Survey will be sent out at the beginning of the month and returned by the end of the month.

 

August 2007

Work on enhancement requests generated by survey

 

September 2007

Examine how changes have fixed any perceived problems

 

Prepare for October rollout

Work with marketing committee to ensure that everything is in place

 

 Create your own educational wiki!Support CommunityPublic, not yet premium Privacy Policy RSS