User Needs

Project Information

Agreed Goals

  • Provide a repository for all UQ Research data. The repository feature set will be attractive to potential users of the system so that they will prefer Fez to their current ad-hoc repositories (such as personal web-sites and file shares).
  • Research data to be stored sustainably - i.e. descriptions of file formats and information needed to view the data in the future should be stored.
  • Easy to store, retrieve and link to data.

Environment

What is the system's business environment?

Research generated by the university should have maximum availablity to increase contribution to academic research and attract more funding and researchers.

Researchers publish their work digitally to make it more available for purposes of peer review and citation to help build their career and attract funding to their area of research.

It is labour intensive to digitise the data and record metadata, so Fez has a goal of trying to reduce any rework that might arise from data-loss or changing data types.

What is the system's physical environment?
Fez is a web application. The servers will be located in the UQ Library LTS server room. The actual repository will be part of the LTS SAN.
What is the system's technology environment (hardware and software)?
Fez will mostly build on and use open source software. This includes Linux Operating System, Apache Web Server, MySQL Database, PHP Web Scripting Language and Fedora Digital Repository. The system will make use of open standards and file formats as much as possible such as XML and standard XML formats such as XSD, RDF, OAI and Dublin Core. Part of the sustainability requirement is that the system shouldn't rely on proprietary software and systems which may become unavailable in the future.
The software will run on rack mounted web server hardware making use of a fibre network connection to the SAN and internet.
Most users will be using personal computers running a web browser to access the repository. Though sometimes, a computer on the internal network will connect to the SAN directly to transfer data into Fez.
The repository may also be accessed via a web service interface to integrate with external repositories in partnership with UQ.

Stakeholders / Actors

All
All stakeholders share the following key needs:
  1. Easily view and retrieve repository content
Library

The Library as an organisation uses Fez to provide a standards complient digital repository that will provide access to data for the forseable future without the need for a lot of re-work.

Key Needs:

  1. Meets current metadata standards (Dublin Core, OAI)
  2. Add new standards for new types of data
  3. Sustainability Metadata to guarantee that file formats can always be accessed
  4. Data integrity
Archiver

The archiver uses Fez to store collections of data. The data would generally not be in a digital format so they will scan it or photograph it with a digital camera or record it digitally etc...

Key Needs:

  1. Upload multiple items
  2. Add metadata about items
  3. Keeps track of which items need metadata
  4. When archiving similar objects, can pre-fill forms intelligently
Author

The author uses Fez as a place to publish and safely store digital research material. They are not always technically skilled so require a simple interface that guides them through the process of storing their material. They want a way to easily give others access to their material and reference it in print or on the web. They are also interested to find out who is using their research and how often.

Key Needs:

  1. Upload research material to the repository
  2. View research material they have added to the repository
  3. Reference the material and share with others
  4. Statistics on use of their research material
  5. Update or remove their research material from the repository
  6. Undo updates they have made
Viewer
The viewer uses Fez to find research material in many different ways. They may want to browse a topic, find research from many topics that has some similarity or crossover between disciplines, find research from a particular author or group or find certain types of research data according to meta-data such as the origin, age, size or format of the data.
Key Needs:
  1. Search the repository by keywords with dependencies on meta-data
  2. Bookmark and return to sets of search results
Administrator
The administrator ensures the data integrity of the repository. They want to control security so that material can only be submitted by authorised authors. They want to set up flexible and robust schemes for labeling and grouping research material. They want to create rules that govern the meta-data requirements for research material.
Key needs:
  1. Manage users of the repository
  2. Create authorisation schemes for users and types of research material
  3. Create meta-data schemes to be associated with research material
  4. Create workflow schemes to manage the status of research material
Reviewer
The reviewer ensures that the data being submitted to the repository meets quality requirements. They will want to make sure that data in the repository can be viewed sustainably. They will want to manage the status of data according to a review process. They will act as part of a workflow defined in the repository.
Key Needs
  1. Review material assigned to them
  2. Manage material as part of a work-flow by setting it's status according to workflow rules
  3. Add / Modify meta-data about submitted material

User Stories

Archiver adds a Conference Paper

Alice has been asked to archive a conference paper. It is a Word document. Since word is a proprietary format, she converts the document to PDF before uploading it into Fez. To upload it, she uses a web form to indentify the document on the hard-disk of her computer from where it is uploaded into Fez. Fez requires her to add some metadata about the paper using Dublin Core records. It also automatically creates some metadata such as the size, format, time of upload, word-count etc...

Archiver Uploads a Collection

Adam is scanning a set of slides to go into Fez. The images must all be scanned as TIFF files so that there are no jpeg compression artifacts. The resulting collection of high resolution files is quite large. Rather than try to upload the collection through the web interface, he copies them to a network drive called 'Fez incoming' (actually part of the SAN). He groups them in a directory. He then uses the Fez interface to ingest the files from that directory. Fez creates the collection and all the items in it very quickly because it doesn't need to copy the files right away. Adam then gets to work on the metadata. Fez copies the files from the incoming area to the permanent archive in the SAN in its own time, creating thumbnails and web displayable images automatically.

Researcher Uploads Data

Reg has heard that Fez is the way to go because many more people can find and use her research and she can keep track of statistics on how and where it's being used. She decides to check it out. An email she received about it has a link to Fez. She clicks on it and sees the Fez front page. She sees a clear link to a section inviting her to put her research into Fez. She logs in using her UQ username and password just like when she reads her webmail.

It is easy for her to follow the instructions to put her files into Fez and create notes about what is in the files. The system fills in a lot of the information automatically and remembers information she has typed before. She uploads some papers she's written and some tabulated data from surveys she conducted. The interface clearly shows that as she puts the research in, she is making a personal Fez area which she can get to at any time by clicking 'My Fez' from the main menu.

Researcher Checks Statistics

When Rod views his personal Fez homepage, he sees the most recent research data added in his name (an archiver may have added the material to the repository for him). He can page through and view older items as well. Next to each item he quickly sees how many downloads each has had and sees links for doing common work on the items such as updating them.

Each item has an obvious link for the referencing of research - with one click he can email an item link to a colleague or generate a citation reference he can use in a bibliography.

Rod's Fez homepage also has a Recent-Comments link where he can read comments people have left about his research. He can configure his account so that he is emailed in future when someone comments on his Fez items.

Viewer Researches a Topic

Vincent is working on a literature review about a topic. He searches for keywords in Fez and finds a result with several promising leads. He bookmarks the search and tries another. When he searches, he can add special words and symbols into the search to build boolean searches that are restricted to certain fields e.g. Dublin Core fields like title, creator, subject etc... combining them with 'AND', 'OR' and 'NOT' operators. He bookmarks several searches.

As Vincent views the research material he found in his search, he annotates the documents in Fez. His annotations are private and he can review a list of his annotations later. He also visits the Fez profiles of some of the authors to see what other research they have published. He annotates his thoughts on some of the authors. When he annotates a research paper or author, the items are bookmarked in his private Fez.

Vincent puts some tags on the items he finds to help himself and others find them later. He tags them with his topic of interest. The tags are publicly viewable and consist of one descriptive word about the research. (This concept has been used with great success on websites such as www.flickr.com)

On some of the items, he writes a public comment summarising his thoughts on the research. Others have left comments about some of the research, informally reviewing it, noting limitations or assumptions made by the author or commending it's strengths which Vincent finds useful to evaluate whether the documents and data are useful to him.

Each item has an obvious link for the referencing of research - with one click he can email an item link to a colleague or generate a citation reference he can use in a bibliography.

Researcher Restricts Permissions

Rod's research group is working on a project that is not ready for publication but which they would like to share with one another. They decide to put it in Fez so their colleagues in Antarctica can review the research material and comment on it through Fez and they all can access that discussion. They set up a collection in Fez and use a form to set permissions on the collection as a whole so that only their group can see it. They also change the permissions so that their corporate partners can view the collection but make no changes to it. Later, when they are ready to publish, they make the research available by lifting the access restrictions.

Reviewer Checks Some Research

Cheryl logs into Fez and sees that there are a number of submissions to be reviewed - these items have been uploaded by others (such as self submitting researchers) and perhaps lack completed metadata. She clicks on each one in turn and follows the workflow instructions. These include checking the metadata, making sure the files are readable, and that they conform to the correct standards.

When she is happy with a research item, she clicks a button to progress the item to the next stage of the workflow (perhaps it will be checked again or its status will change to a completed state). For one of the items, the document contains a diagram written in an unknown application. She clicks to assign the research item to the administrator who deals with the problem of convincing the author to save the item in a different format or by adding support for the format if it is an open standard.

Administrator Adds Metadata Options

The Entomology research group would like to add a collection of cricket recordings they made on Moreton Island.

Adam works with the entomology department to find the best way to digitise and store the collection with the most useful metadata. They find out about an XML format called BugML which is being used by other universities for this purpose. Adam obtains an XSD that describes the format. Adam goes to the Fez administration page and adds the XSD file to the accepted format for Fez. He then configures how the form to make a BugML file will be displayed. The web interface allows him to link the fields from the XSD file to HTML form elements such as drop down boxes and text fields. He is able to place constraints on the fields so that the entry form will check the validity when the archivers / researchers are uploading their data later. Adam doesn't have to do any programming and the system automatically sets up the form from the XSD file so he only has to tweak it here or there.

Performance and Capacity Needs

At this stage we don't know what the capacity of the system will be. A gross estimate is about 300GB per year with about 200 new authors per year (after the initial sign-up of maybe 2000 authors).

Traffic to the site will vary according to student and academic interest and may surge with media attention to research or after large conferences.

Company Proprietary
Copyright © 2003-2004 Jason Robbins. All rights reserved. License terms. Retain this copyright statement whenever this file is used as a template.