...
To oversimplify, the goals are:
- Provide a flexible authorization system to allow collaboration between instructors and librarians.
- Whenever possible, store generic information (ISSN) rather than site-specific information (link to a specific vendor's page for a journal) to facilitate long-lived links that work across semesters, courses, and in some cases institutions, and to avoid "linkrot."
- Have a single instance of each item that appears in a guide. Information that is used to construct a link/citation is static and used by all instances; data such as the title and descriptive text are guide-specific and can be overridden.
- Provide medium-specific editing tools for guide authors that take advantage of external services (from Sakaibrary and elsewhere) to make adding and updating items as painless as possible.
- Pursue a strategy of reuse, making it easy to use items in Sakaibrary, whole or part of another guide, and item descriptions entered elsewhere in the system.
- Provide output in a number of useful formats, including a well-described XML format that can be used by a sophisticated client (such as Sakaibrary) to take full control over the rendering of an SRG.
...
- An authoring system, for gathering information about included resources, authorization/authentication of authors, and storage of the guide structures themselves.
- A Sakaibrary servlet, which will provide web-services access to data and resources already gathered and structured as part of the Sakaibrary project
- An optional service that can be used to perform simple transformations of data (e.g., turn an ISBN into a link to the local OPAC)
- A rendering system (tightly coupled with the authoring system) which will produce SRG output in a variety of formats such as HTML, machine-optimized XML, PDF for printing, etc.
| Panel | ||
|---|---|---|
| ||
|
It is intended that in addition to being a crucial part of the authoring process, Sakaibrary will serve as a target for a rendered citation (number 4) as well, taking advantage of other Sakaibrary infrastructure to provide specialized views of some items (e.g., a database could potentially include a Sakaibrary searchbox targeting that database).
...
When possible, the SRG will rely on external services to turn generic, guide-neutral information about a resource into a specific, local pointer to that resource. Sakaibrary already collects and generates much of this information, in the form of database information, code needed to generate search boxes, OpenURL generators, etc. The Sakaibrary functionality can be coupled with other (optional) local services to allow the SRG to simply describe the resource (in much the same way a citation is a description of a specific article or book), and turn that description into a valid local pointer/link at the time the guide is viewed.
| Panel | ||
|---|---|---|
| ||
|
Creating a rendered research guide
Rendering an SRG (to HTML or other format) will reverse the process. Sakaibrary and any other external services will be called to fill in item metadata with the most recent edits (if any), making sure that links are active. This will then be output to whatever format is called for, usually HTML for a web page interaction.
| Panel | ||
|---|---|---|
| ||
|
However, a well-specified XML format will also be available that includes all the various UIDs and other metadata available for each item. This will allow a "smart" renderer such as Sakaibrary to add searchboxes, canned and constrained searches, references to existing citation lists – anything available to Sakaibrary that makes sense in the research guide context.
At the other end of the spectrum, a print view would remove all interactive elements but include URLs in a visible form to facilitate the guides' usefulness on the printed page.