This page provides vivid descriptions of scenarios we expect to see in the use of DSpace.
Joining DSpace | Simple submission into DSpace | Preventing submission into DSpace | Submission into DSpace with workflow | Retrieval with rights management | Retrieval for cash | Retrieval across multiple communities | Administering DSpace | Administering a community | Defining a new bitstream format | Defining a new compound content type | Extending a content type | Creating a new DSpace fulfillment service | Creating a new storage mechanism within the archive | Creating a new DSpace query service
A Department, Lab, Center or group of people becoming a community within DSpace for the first time
Adam, the administrator for the MIT Center of Teleportation Research (MIT-TR) has heard about DSpace, from a friend down the hall. He visits the DSpace web site and decides he wants to learn more. After a brief telephone conversation with the DSpace Faculty Liaison, a meeting is set up. At the meeting, Adam and the DSpace Faculty Liaison discuss DSpace services, and review the existing MIT-TR web pages. They identify the workflow of research papers in the Center: from author through various informal reviews, and finally to being published in outside journals. They agree that the DSpace storage, retrieval and preservation services would fit in with the existing MIT-TR publishing processes, and add value. Since Adam is also the MIT-TR web site administrator, the various aspects that define the MIT-TR look and feel are easily copied out and given to the DSpace Faculty Liaison, who during the meeting uses the DSpace Configure Tool to create an initial DSpace site for a new publishing community: The MIT Center for Teleportation Research. At the next group meeting for the Center, Adam shows off the DSpace site, and gets feedback on how it needs to change before it can be used by most members of the Center. After some email exchanged between Adam and the DSpace Faculty Liaison, the community is completely configured. Using a combination of automated database extraction and some hand-tooling, the community is populated with the knowledge of who is in the Center for Teleportation Research, and what roles each person plays. For example, in the submission process, there are roles of author, reviewer, copy editor. In MIT-TR there is the person who has final approval over what gets published in the official MIT-TR Reports. Adam's DSpace identity reflects that he has two roles at different times: copy editor, and final approval for MIT-TR Reports.
Simple submission into DSpace by a member of a community
Tom is a researcher in the MIT Center of Teleportation Research (MIT-TR). He has just finished a working paper that he wants to share with his colleagues around the world. Since he has an identity in the MIT-TR community, he simply goes to the MIT-TR DSpace submission web page, and clicks on the "SUBMIT Working Paper" button. Tom is prompted for information through a web form that is similar to other submission processes he has experienced when submitting publications to print journals and preprint archives. This time he chooses to take advantage of the ability of DSpace to preserve not only his paper, but the datasets that he used in the preparation of the paper. Shortly after he clicks on the "No more items" button, his paper and all the associated datasets are available in DSpace for retrieval.
Rejection of submissions into DSpace from people who are not members of an appropriate community
Clifford has been experimenting with Teleportation at home. He happens across the MIT Center for Teleportation Research DSpace page while surfing the net. He attempts to submit a paper he has written to the site. He clicks on the "SUBMIT Working Paper" button. DSpace pops up a web page thanking Clifford for his interest and informs Clifford that he is not recognized by DSpace as a member of the MIT Center for Teleportation Research community. The page offers links pointing to forums that are related to Teleportation but not access restricted, and to a page with information on how to join the Center for Teleportation Research community.
Submission into DSpace by a member of a community into a controlled collection
Sally, a researcher in the MIT Center for Teleportation Research (MIT-TR) would like to publish her working paper, "Flogiston Induced Teleportation" as an official technical report of the Center. Sally goes to the MIT-TR DSpace web site. She reviews the instructions on how to prepare a submission. They look a lot like what she is used to from when the Center for Teleportation Studies published only on paper. She already has most of the proper file formats in hand, and follows a few simple steps to convert one file to a preferred format. She then fills out a form that looks like a web version of the form used on paper the last time she submitted something to the Center for publication and clicks on the "SUBMIT" button. She is prompted for the names of the files that contain the paper and various datasets being submitted. Mary, is the submissions director for the Center for Teleportation Studies. She receives notification that Sally, has uploaded a working paper for submission as an official technical report of the center. It is Mary's job to see that the document is readable, and that there is enough metadata to make the paper reasonably searchable. Mary sends Email to Sally asking for a few more keywords to describe the paper, and points out a few paragraphs that are difficult to understand. Ralph is a copy editor who works for the Center part time. He too receives notification that there is a paper he needs to look at. He reads over the paper and sends Email to Sally requesting correction of a few typographical errors. Sally submits revised draft that incorporates Ralph and Mary's feedback. Mary assigns an official Center for Teleportation Studies Technical Report ID to the paper and flags the submission as "Complete". Shortly thereafter, Sally's paper appears in DSpace for all the world to see as the latest Center for Teleportation Studies Technical Report
Retrieval from DSpace by people who are identified as having rights to content in a particular community
Riley is an Associate Professor in the MIT Department of Mechanical Engineering. He has registered his Email address with DSpace in the Center for Teleportation Research (MIT-TR) community, asking to be notified when new Technical Reports are added to DSpace. He receives notification of the posting of Sally's paper. He goes to the Center for Teleportation Research DSpace web site. DSpace recognizes Riley's MIT affiliation. The MIT-TR has set up the rights management of its technical reports to give free access to everyone at MIT having an academic appointment. Riley downloads a copy of Sally's paper and all the associated data at no charge. Some of the data intrigues him. Since he is a Research subscriber he is allowed to fetch her datasets, which he uses to attempt to reproduce her findings. Riley later goes on to write a doctoral dissertation based on some of Sally's research that also gets submitted to the DSpace Collection of MIT Theses.
Retrieval from DSpace by people outside a community who buy the right to retrieve
Courtney works for American Transportation Unlimited, a company working on commercial teleportation. He has registered his Email address with DSpace in the Center for Teleportation Research (MIT-TR) community, asking to be notified when new Technical Reports are filed. Courtney received notification of Sally's new Technical Report. Going to the Center for Teleportation Research DSpace web page, he reviews the abstract, and the excerpts of the paper available for free to everyone. He decides he wants a complete copy for himself. Clicking on the "Buy this Document" button, he is taken to a secure web site that presents several offerings at different prices. Courtney enters his credit card information and selects one of the mid-priced offerings that enables him to download the paper and a single dataset.
Retrieval from DSpace not limited by community boundaries
Charlie is a freshman at MIT. He is interested in Teleportation. He goes to the MIT Libraries DSpace page. Using a tool developed jointly by the DSpace team and the MIT Libraries, he performs a search across multiple DSpace communities on topics related to Teleportation. He follows hyper-links in his search results to Tom's working paper, Sally's published paper, Riley's dissertation, and papers from the Mechanical Engineering Department at MIT.
Performing maintenance, monitoring, and administration of DSpace. Jim is the systems administrator for one of several DSpace systems
Via a secure http connection, he points his web browser at the control page for the DSpace system he is responsible for. DSpace recognizes his role as a system administrator and is prepared to give him special privileges. He peruses the logs to determine that all services are up. He checks to see how many hits his favorite community received today. He goes to the replication control page, notices that the copying to the Cornell DSpace site is complete, but that the copying to Stanford and Berkeley seem to be bogged down. He slides a control slider to tell his DSpace to throttle back the transfers to Berkeley for now. In preparation for a budget review, Jim asks DSpace to generate some reports on how many accesses have been made both to free content and to paid content. The amount collected over the past quarter will be of particular interest. Luckily, the review is internal rather than with external sponsors, so the full power of the DSpace logging and report generation facilities need not be used tonight.
Performing maintenance, monitoring, and administration of a community within DSpace
Mary, the publications director for the Center for Teleportation Research (MIT-TR) is preparing for tomorrow's group meeting. She goes to the admin pages for the MIT-TR DSpace pages, and generates a report that tells how many people have accessed the Technical Reports of the Center. Of particular importance is the count of how many accessors actually paid real dollars and obtained their own copy of a Technical Report.
Adding awareness of a new leaf-level bitstream format to DSpace
Margret, MIT Libraries' faculty liason within the DSpace operations team, has been working with several researchers from the Media Lab's center for Automotive Applications of Holography. These researchers have increasingly been using a new international standard for digitally capturing holograms. Hewlett-Packard company and several competitors have introduced products that read this binary digital format and display holographic movies.
Margret and the researchers decide that it would be desirable for DSpace to offer support for the holographic description format (".hdf").
Margret, through an administrative interface, adds ".hdf" to the list of bitstream formats that DSpace supports. The admin interface queries Margret for references to the international standards that describe this file format. Subsequently, DSpace offers this format as a choice to individual submitters as they submit content files.
See also "adding a new transformation service to DSpace"
Maintaining a set of definitions of the types of compound items that DSpace supports
Margret, MIT Libraries' faculty liason within the DSpace operations team, has been working to understand and meet the needs of medical researchers. These researchers generate, on an ongoing basis, interpreted X-ray images that are representative of certain degenerative conditions. They are motivated to make this corpus available online and relate the interpreted X-rays to their published work in the field. Through interviewing these researchers, Margret determines that: X-ray images submitted for this purpose should always contain: an image in one of several image formats instance metadata in an "x-ray-descriptor" schema. Researchers have defined this schema to capture information about the x-ray such as the medical conditions indicated by the X-ray, and characteristics of the patient such as age, sex, prior conditions, etc. an ascii file containing notable comments about the X-ray contextual metadata that links this X-ray to one or more studies published in one or more articles Adding a compound content type for interpreted X-rays makes sense for the library, the researchers, and for DSpace. That is: This content type has a sufficiently wide or important audience that Libraries wishes to take it on as a new type of holding. Libraries are willing to invest the energy to define the characteristics of such holdings. The benefits gained by the availability of the interpreted X-rays exceed the startup costs of adding support for the content type to the DSpace system. There is some mechanism for funding or recovering the startup costs involved. Margret creates a unique name for the interpreted-x-ray compound content type, and describes its information requirements and constraints by writing an instance of a content-type-definition schema. Margret validates the content type definition and, using an administrative interface, adds it to the DSpace system. Upon addition of the new content type definition, DSpace subsequently allows collection administrators (through the collection admin interface) and individual submitters (through a "detailed" version of the submission interface) to choose this content type definition to govern specified submissions. That is, they may require that submission(s) conform to the information requirements described in the new content type definition.
Augmenting existing holdings to be conformant with a new content type
Shortly after creating the interpreted-x-ray compound content type, Margret becomes aware that a large corpus of pre-existing "publication" holdings that meet most of the information requirements to be considered "interpreted-x-ray"s, but only about 20% contain "x-ray-descriptor" instance metadata (this schema has only been around for a year or so). After conferring with the medical researchers, the researchers decide that it would be helpful to their research agenda to annotate a portion of this corpus with the missing x-ray-descriptor metadata. Using a DSpace submission interface, one of the researchers annotates about 300 X-rays with new "x-ray-descriptor" instance metadata. Margret then runs a DSpace query, finding all the DSpace holdings that now meet the information requirements for interpreted-x-rays. Using a batch process, she annotates each of them as being a conformant interpreted-x-ray.
Adding services that extend DSpace's capability to deliver holdings of particular content type(s) to retrievers
The researchers that led to the creation of an interpreted-x-ray initiate a dialogue with some of their colleagues at the University of Chicago school of medicine. Between them, they decide that they have sufficient resources to fund the creation of a viewer for interpreted-x-rays. Some programmers at the University of Chicago write a program that accepts the URL of an image, an ascii english prose x-ray description, and an instance of x-ray-descriptor metadata in XML. The program produces an html page that displays the x-ray with its annotations, and links the annotations to a medical dictionary, hosted at Harvard Medical School, that describes the various conditions. [much handwaving follows. Let's firm this up?] The programmers define an interface to the program, and register each of the methods in the interface in UDDI. They also send an email to the DSpace administrators describing the interface. The DSpace administrator creates a DSpace service interface (adapter?) for "x-ray-descriptor" that maps to the University of Chicago implementation. When an end-user accesses an interpreted-x-ray via DSpace, DSpace offers the new rendering service as one of the fulfillment options. When the user selects that fulfillment option, the program at UC is discovered and invoked via RPC, the interpreted-x-ray is rendered there, and transported (via RPC return stream) back to DSpace, who delivers it to the end user.
Adding services that extend DSpace's capability to deliver holdings of particular content type(s) to retrievers
12 months after DSpace goes live with the MIT community, DSpace administrators realize that a significant amount of video material has accumulated within DSpace holdings. Further, the performance demanded by users accessing video holdings exceeds the performance demanded by users accessing holdings that are primarily textual. MIT Libraries decides to invest in new storage hardware to better stream the video holdings within DSpace to its end users. The hardware is purchased and installed. Programmers write a storage adapter, a program that accepts a bitstream and a DSpace URN, and stores the bitstream on the new hardware. Alternatively, the program accepts a URN and returns the corresponding, previously stored bitstream. The adapter is associated with the "video" bitstream format within DSpace. DSpace administrators perform a query that identifies all video bitstreams in current holdings. These are each stored, via the storage adapter, on the new media, and then removed from the previous media. An end-user submits a compound object that contains a video. DSpace resolves the storage for the video, through the adapter, to the new media. The remaining portions of the document are stored as before, on the previously existing media.
Adding new ways of discovering DSpace resources
MIT Libraries partner with one of the working groups of the World-Wide-Web
consortium to create a new query service that enables users to enter relationships
among semantic concepts from "WorldNet", and return articles believed
to satisfy those relationships. Relationship to "Creating a new storage
mechanism within the archive": Implementation of said search capability
might reasonably involve the addition of some new storage/access mechanism
with different performance characteristics than existing DSpace storage
mechanisms (e.g. RDFdb, jenna), and causing storage of certain content
types to be diverted or replicated within that storage mechanism. Relationship
to "Creating a new DSpace fulfillment service": Alternatively, the query
capability might be implemented wholly as an external service, based on
either bulk exports of data, or periodic pushing of data from DSpace to
the query service as new content enters DSpace.