The BID project team will use these scenarios to frame discussion about the technical integration work between SAKAI, Fedora and SRB.

This scenarios were taken from the JISC CRIG Wiki Page.

Scenarios 1: Learners

L1_Single Sign On (One Username And Password)

From Repo Wiki Source: ACDT Suggestion Box, Summer 05 / Adapted by Neil Jacobs

A Chemistry student logs into his institutional VLE system with his local username and password. He is instantly presented with his personal area which lists a set of links to the web pages that he access the most. From this page the student clicks on a link to an online journal that he regularly uses to see if there are any articles to help him with the mid term paper that he is writing. The journal system instantly recognises that he is from Oxford University and lets him access the full texts. Having reviewed some abstracts, he identifies three papers of potential interest, and tags them as such. He then leaves the journal and clicks to perform the same search, in turn, on a global search of open access research papers, and on the university library database. From viewing abstracts during the open access search he identifies two further papers of potential interest and tags them as such. He then opts to download and/or print the tagged papers, depending on the licence conditions associated with each. Finally, the search for books related to his paper on the local library catalogue throws up a number that look useful he decides to hold some of the items to pick up later. The database can identify who the student is from their initial log in to the system but refuses to let him hold the books because he has fines on a number of items that he did return but were overdue. The student clicks on the "pay fines" button and enters the overdue amount. The system automatically debits his bank account and lets him hold his books. Before the student leaves the lab to pick up his books he clicks on the link to the course registration page to book himself on an advanced excel course. Again he is not required to enter any personal information as the system recognises who he is, the same is true when he checks his university web e-mail account.

Solution

insert solution here...

L2_Leaving University

From Repo Wiki Source: ACDT Suggestions Box, Summer 05 / Adapted by Neil Jacobs

Student Y is just about to graduate as a doctor of medicine from university. She has been storing all of her notes, annotated reading lists and work in the institutional repository for the past three years and wants to retain access to all of her files. The repository provides a service whereby student Y can download the entire contents of her area in the repository into a zip file which she can place on her personal computer. The zip file retains any hierarchy of data that was present in the repository via a folder structure. The repository also provides a service whereby it offers ongoing access to this material to Student Y, and (under her control) to anyone else she selects, perhaps using an e-portfolio tool. The IPR is appropriately managed.

Solution

insert solution here...

L3_Store E-Portfolio Artefact

From Repo Wiki / Adapted by Neil Jacobs

Student Z is using an e-portfolio (or PDP system such as LUSID) and wishes to store an artefact (for example, Word Document, Photo, Video Clip, Sound recording). The student clicks on the 'store my evidence' button and is faced with a screen or series of screens where they upload the artefact and add some metadata. The default visibility would be that only the owner can see the artefact, although in another scenario we will see that the student may like to share the artefact with others, and in this case there is a process whereby the rights are managed appropriately.

Solution

insert solution here...

L4_Resource Discovery Service

From Repo Wiki source: ACDT Social Sciences Project

Mrs. E is a distance learning student studying for a diploma in social policy at Oxford University. She needs to find unemployment data for the previous century. She accesses the institutional repository, which recognises her as a student of the university. She performs a search within her subject domain of social policy to discover reliable datasets of unemployment figures. The search results that are returned contain datasets from the UK data archive which is publicly available. It also returns datasets which have been uploaded into the institutional repository, and are only available to members of the University. The search also returns datasets from the ICRSP service, which is a subscription-only service which the University subscribes to and Mrs.E therefore has rights to access. The search results show the name and author of the dataset, as well as a brief description and a URL pointing to where the data files which are held.

Solution

insert solution here...

Scenarios 2: Researchers

R1_Private Personal Details

From Repo Wiki Source: ACDT RDS project / Adapted by Neil Jacobs

Dr. Y is working on a highly sensitive and controversial research project that is likely to elicit hate campaigns from various lobby groups. Whilst he has a number of research students and colleagues with whom he would like to share some of his papers, data and findings, he needs to be very selective. He stores all of his digital work in an area of the institutional repository. Whenever he wants to make a document available to someone else he is able to select that object and give individual people permissions to access it. They cannot see any thing else of his when they access that area. The conditions under which they are able to access the material are clear to them when they access it.

Solution

insert solution here...

R2_Cross-institution Research Group Collaboration

From Repo Wiki Source: Oxford University Shibboleth Case Study document

Dr. P is member of a cross-institutional research group studying gene therapy. He is able to share his material with his peers in other institutions, and view their material held in another participating institution's online repository. Access to the research group's material is necessarily independent of the host institution's LDAP system as the group consists of members from various institutions. Group members roles within the group and rights within the system are not defined within the system itself, rather they are determined based on externally supplied information regarding the members. Group members’ rights within the system may include uploading, editing, deleting or even system administration abilities.

Solution

insert solution here...

R3_ReusableVideoClipsCollection

From Repo Wiki Source: ACDT Dynamics of Rotating Fluids Project

Dr. B has a large collection of digitized archive footage of experiments carried out across the last century. He has made all these clips available within the institutional repository and searchable online, so that that any relevant clips can easily be found and used in research and teaching. In many cases he has made sub-clips from existing clips and has associated them together such that when another user looks at a subclip they can easily find other sections of the same footage. Also when relevant searches are performed, all related clips and sub-clips are found. Additionally, other people are able to make their own records relating to a video clip or clips by adding their own annotations.

Solution

insert solution here...

R4_Transfer Resources To PDA

From Repo Wiki Source: Conversation with Dr. of Archaeology

Dr. H of Archaeology is going on a dig in Niger. The area in which they will be working has no internet access but she will need to access a number of items she has stored in her area of the VLE, such as excel spreadsheets and archaeological maps she is plotting. Occasionally Dr. H will visit the capital Niamey where she will have access to the internet at the University and from her hotel. Rather than print out the documents and excel spread sheets she needs in paper format (which are likely to become damaged during the dig), Dr H downloads everything she needs before she leaves onto her PDA. Whilst on the dig she updates various excel spread sheets and maps, and takes a number of photographs with the PDA's inbuilt camera. When she has access to the Internet in Niamy she connects her PDA to a PC and deposits all her resources back in VLE. Every time she adds new resources to the VLE her research team are sent an automatic e-mail telling them new data is available.

Solution

insert solution here...

R5_Interface Between Research Expertise Database And Full Text Repository

Author: St Andrews / AKC

Background

The University of St Andrews Research Expertise database contains research-related information for staff across the University. It includes metadata for over 10,000 publications.

The University is now working with the University of Edinburgh on implementing a dSpace full-text Repository. The University wishes to provide a single interface for researchers to add/update details of research publications. It also requires a one-off batch process to upload selected publications from the Research Expertise database into the Repository. The end result of both of these should be :

• the metadata in the Research Expertise database
• the full-text deposited in the Repository with relevant metadata
• the URI to the full-text version in the Repository stored in the Research Expertise database; this URI should be to the final full-text version (whatever that is depending on copyright issues etc)

Key requirements for the interface between the two systems are :

• both systems will continue to function without the other; indeed there will be many examples of full-text deposits which have no relevance to the Research Expertise database
• we could switch to using different Repository software/database with limited re-working of the interface
• the interface will be one-way; ie the Research Expertise database will drive the process; this has the problem of how to deal with (or prevent?) a change to a record in the full-text Repository as the Repository has no knowledge of the record in the Research Expertise database linked to it's URI.

Solution

insert solution here...

R5a_New Publication added to the Research Expertise database

A researcher has a new publication that they wish to add to the Research Expertise database. The researcher logs in and enters the details of the publication (year, title, authors, location, etc). After they have saved the details they are prompted to deposit the full-text version of the publication. The system will first check for matching record(s) in the Repository; if none found then it will create a record in the full-text Repository and pass back a unique Repository ID to the Research Expertise database. If a matching record is found, the system will update the metadata in the Repository and pass back a unique Repository ID and the URI (if known) to the Research Expertise database. In both cases, the system will place the record (new or updated) in the relevant position in the Repository's mediation/workflow process. For example, for the new record, additional metadata may need to be completed, and the full-text deposited.

Solution

insert solution here...

R5b_Existing Publication updated in the Research Expertise database

A researcher wishes to update some of the details of a publication in the Research Expertise database (eg change the title, add journal volume and pagination details etc). The researcher logs in and find the relevant publication and makes the changes. After they have saved the changes the system will do one of two things : (i) if there is a Repository ID stored against this publication, the system will update the metadata in the Repository (ii) if there is no Repository ID stored against this publication, the system will follow Scenario1 above to add/update a record in the Repository.

Solution

insert solution here...

R5c_Batch update of URIs in the Research Expertise database

Because any new deposit into the Repository is required to go through certain mediation/workflow processes there will always be a delay in creating a record in the repository and knowing the final URI of the deposited full-text. This scenario explains how the URIs can be updated in the Research Expertise database. An administrator selects a subset of publications in the Research Expertise database to update; this could be, for example, those publications added in the last week. The subset will only include publications with a Repository ID in the Research Expertise database. The system searches the Repository for matching records and returns the URIs for each match. The URIs are then stored in the relevant Research Expertise database records.

Solution

insert solution here...

R5d_Batch update of metadata from Research Expertise database to full-text Repository

This is a one-off scenario which describes how the Repository can be populated with metadata from the Research Expertise database. An administrator selects a subset of publications in the Research Expertise database to upload to the Repository, for example all those for a particular Researcher. The subset will exclude any with a Repository ID in the Research Expertise database. If we can be sure none of these publications already exist in the Repository, the system can just create a new record for each publication. If we cannot be sure that none of these publications already exist in the Repository, the system will need to search for matching record(s) for each one, and only create a new record if no match is found. It will need to update the metadata for existing record(s). In either case, the system will need to pass back the relevant Repository IDs to the Research Expertise database and place all the new/updated Repository records in the relevant position in the mediation/workflow process.

Solution

insert solution here...

R6_Research Workflow

Author: Neil Jacobs

A researcher signs into their local system (this is the only time she enters authentication/authorisation information in this scenario). She communicates with colleagues via email and skype. She notices that she has been asked to review a dataset recently deposited in a national data archive, so she accesses and assesses this, in the process comparing some of the parameters and values with datasets held elsewhere. She assigns a peer review value to the dataset. She then checks her own dataset, held for the moment at her institution on a departmental / laboratory repository; this dataset is automatically being updated by a sensor array at remote locations. By applying an analysis tool across elements of the dataset, she confirms a trend that she has previously suspected, discusses this with colleagues online, and identifies the potential for a conference paper. To start drafting this conference paper, she sets up a new document in her institutional repository, whose metadata is automatically populated with her and her team’s names, the project name, persistent links to relevant datasets, funders and relevant grant codes. Her team (based at a range of institutions) have edit rights to the new document, and there is full version roll-back and audit. The whole team is emailed to alert them to the existence of the new document.

Solution

insert solution here...

R7_Reuse Of Research Items

Author: Neil Jacobs

On reading a paper from a repository, a researcher is curious about the claimed findings and runs what should be an equivalent analytic procedure on the source dataset that is linked from the paper. Finding an anomalous result, she combines the original dataset with some additional environmental data, and notes that the anomaly is accounted for by an environmental variable not considered in the original paper. The researcher publishes a new paper describing these findings, deposits it into her institutional repository, together with the revised dataset, with appropriate rights management and control throughout. (NB it is possible that a demonstrator based on this scenario would fail because the IPR framework is not yet established, but this might be avoided?)

Solution

insert solution here...

Scenarios 3: Teachers

T1_Digital Resource Aggregator

From Repo Wiki Use Case Name: Digital Resource Aggregator Actors: Lecturer

Summary A lecturer compiles a virtual collection via a combination of uploads and searches across multiple repositories. The resources are displayed as one single collection. Basic Course of Events A Lecturer logs into the system The Lecturer creates a new collection The Lecturer uploads a number of their own resources to the collection They then perform a federated search to find other appropriate resources for the collection. The Lecturer imports selected objects from their search results, according to any rights conditions The Lecturer makes their collection appropriately available to appropriate groups of students.

Alternative paths If the Lecturer only want to include their own objects in the collection. If the Lecturer want to add objects to existing collections. If the Lecturer want to add references to other objects rather than the objects themselves to the collection.

Assumptions The lecturer has a collection of digital resources that they have the right to use. Preconditions Standards compliant systems to search. Postconditions The lecturer is able to present disparate resources in one coherent system.

Author: K Lindsay / J Talbot Date: 10/08/2005 Use case number: US1 Use case category: Search, Discover, Storage Scenarios:

Solution

insert solution here...

Scenarios 4 : Managers

M1_Publications Database IR_Cross-Check

Author: Neil Jacobs / Amber Thomas

A repository manager initiates one of a series of regular cross-checks between the institution’s publication database and their (separate) repository of research papers. The scope of the two systems is not the same; neither are their metadata schemas. However, both are registered in the Information Environment Service Registry, and have metadata schemas defined in the Information Environment Metadata Schema Registry. In addition, the repository is register in OpenDOAR. The cross-check detects publications that are described in the publications database, whose full text could (in terms of scope) be held by the repository, but which is not held by the repository. Of these, the cross-check distinguishes between those publications that have previously been identified and those that have not. For each publication in either of these lists, the cross-check also uses the Sherpa-RoMEO service to check whether any (and if so, which) version of the publication can be made available OA via the repository. Where it is possible to make such a version available, the cross-check sends an email to one of the (perhaps many) authors who work at the university, requesting them to deposit the appropriate version of the publication into the repository, bearing in mind any conditions on deposit set out by the Sherpa-RoMEO service.

Solution

insert solution here...

M2_Populating Departmental Website

Author: Neil Jacobs / Amber Thomas

The manager of a university departmental website sets up appropriate links to the institutional repository so that there is a dynamically updated list of the repository content displayed in a box on the departmental web page. The contents of this box can be configured to represent the most recent items in the repository, items from a particular research group or researcher, items on a particular topic (using either a controlled vocabulary or ‘canned’ keyword search outputs, or items selected by a manual trawl of the repository by the manager or their support staff.

Solution

insert solution here...

M3_Target Metadata Only Records

Author: Neil Jacobs / Amber Thomas

At the instigation of the repository manager, a list is generated of all records in an institutional repository of research papers, where that record does not have full text, and refers to an item that was published more than 12 months ago. One author (who works at the institution) of each of these items is emailed to request deposit of the full text of the most complete version of the paper, providing that is allowed (by reference to the Sherpa-Romeo listing). The emails are coordinated / collated, such that no author receives more than one. The responses to this email (in terms of items deposited) are monitored and a month later a report is generated for the repository manager, breaking the responses down by department. This enables her to target her advocacy efforts efficiently.

Solution

insert solution here...

Scenarios 5: Editors

E1_Quality Marking Papers

Author: Neil Jacobs

An editor trawls the contents of a known and trusted set of repositories and identifies a paper whose qualities he wishes to have expertly assessed, to see whether it can be included in a thematic journal issue that he is planning. Having selected the paper, he clicks on ‘have this item reviewed’. A dialogue box offers him a range of options, such as who should review it (from the journal editorial board), what is the deadline, and what – if any – exceptional criteria should they use beyond the normal journal criteria. The selected academic is sent an email and asked to review the paper by completing an online form. As a result of this activity, a potential reader of the paper is able to assess its qualities and discriminate between it and other papers on the basis of specific criteria of her choosing (from a prescribed list of criteria). (NB- the scenario is technical; there may not be a business model that would support this at the moment)

Solution

insert solution here...