ScholarsBox/EssaySeries/CurrentState

In Essay 2 (and probably a few following essays), I will describe the current state of the ScholarsBox (SB).

What is the Scholar's Box
Overall characterization of current state of ScholarsBox
How we got where we are
Technical current state
Screenshots
Opening screen
Simple search
Return of search results
Sources

CDL
amazon.com
google and melvyl (should be separate)
NSDL
CalPhotos
Flickr
MetaLib
METS
IMS-CP
RSS/Atom feeds
Desktop files
Drag and drop from browser
Scholar's Box FF Extension
Opensearch

Collection Handling

basic handling
collection menu
Annotating an individual item in the collection
looking at more images/items
XML export

Outputs

HTML
OO.o
Microsoft Office 2003
Endnote
PDF
IMS-CP
Output to weblogs
RSS
Chandler Parcel

Options
Help

The goal of this essay is to describe how the Scholar's Box works right at this moment. We will first show the current interface (as a series of screenshots) and then present a critique and plans for a redesign. (Although there is some current information about the Scholar's Box to be had in the various presentations that we have been making (see MyIuWork/PresentationsAndTalks), we have yet to produce detailed and current documentation of the ScholarsBox, which is what I'm aiming to write here.)

What is the Scholar's Box

We at the Interactive University Project are building the Scholar's Box.

The Scholar's Box is a tool that gives users "gather/create/share" functionality, enabling them to gather resources from multiple digital repositories in order to create personal and themed collections and other reusable materials that can be shared with others for teaching and research. The Scholar's Box can currently perform the following functions:

Gather: From California Digital Library, amazon.com, google.com, NSDL, RSS feeds, METS (digital library), WWW, CDL’s metasearch system, and the local file system.
Create: Data and metadata gathered, annotated, and organized into personal collections via drag and drop
Share: IMS-CP, OpenOffice.org Presentation or Text document, PDF, HTML, a METS document, a set of Endnote references, Chandler Parcel, or sent to a weblog via the Blogger API

I will continue to build out the Scholar's Box as a prototyping system to demonstrate technical interoperability for our various other projects (e.g., the creation of reusable collections of images and bibliographic metadata; testing the metasearch API from CDL). We at the Interactive University are in process of releasing the Scholar's Box as an open source project so that others can benefit from the overall architectural framework as well as any concrete implementations of services (e.g., creation of OpenOffice.org presentations from collections).

We are writing specification documents for the Scholar's Box, which we break into three parts:

/StrategicIssues, which lays out the organizational issues around Scholar's Box development, including project plans and the rationale behind this work
/FunctionalSpecification, which describes how the Scholar's Box will work entirely from the user's perspective.
/TechnicalSpecification, which describes the internal workings of the Scholar's Box.

Note that some of the most active description of the ScholarsBox in my wiki is to be found in An Occasional Essay Series about the Scholar's Box.

[ScholarsBox]

Overall characterization of current state of ScholarsBox

The interface is at the "edge of usability." The emphasis of our work in the Scholar's Box has been to produce a proof of concept rather than to hone the tool into a finely usable interface. There is a lot of good functionality, but the various functions need to be presented in layers to make the interface more usable and intuitive.

How we got where we are

We started with a focus on images with the plan of generalizing to other data types such as bibliographic references and texts.
In redesigning the interface, we are moving from a collection of functionality to a smooth blending of this functionality in a good UI.

Technical current state

A detailed outline of our current object model will be offered. Such documentation will be very useful in any architectural refactoring and wrapping the Scholar's Box in alternative interfaces (such as a Web-based version). Also, such documentation would be helpful to anyone interested in implementing the services of the Scholar's Box in another language, such as Java.
The Scholar's Box is currently written in PythonLanguage. We have been asked why we are using the why PythonLanguage? The answer is a topic for a future essay. Brief answer: Python is a marvelously flexible language for agile development, prototyping. Also we will very likely want to tap into the ChandlerProject for infrastructure.
We wanted to leverage existing infrastructure where possible.
We have pursued a philosophy of PracticalInteroperability.

Screenshots

SB is currently a desktop client, written in PythonLanguage and WxPython. In theory, it should be easily portable to the Mac and to Linux -- though we haven't done that port yet. The following screenshots are how SB will in the Win32 version -- so Mac and Linux users should adjust their expectations accordingly.

Opening screen

When the Scholar's Box is started, the user is presented with a choice of four actions:

start a search
do a metasearch
create a collection
open an existing collection.

We wanted to prompt the user with something to do rather than just let the user flounder by looking through the menus. (Maybe there are other ways to make things clearer. This prompting is what happens at startup with Endnote.)

The Scholar's Box depends on search as the primary way to find items to gather. There is no browse functionality per se -- though there are ways to send references and extracts from FireFoxBrowser to ScholarsBox.

Simple search

Prompt for a search term and selection of repositories to choose from. Current choices:

amazon.com
californiadigitallibrary.org
Melvyl: According to Melvyl Catalog Help:

The Melvyl catalog contains records for materials (books, journals, movies, maps, music scores and recordings, computer files, dissertations, government documents, etc.) held by the libraries of the ten UC campuses, the California State Library, Hastings College of the Law, the California Academy of Sciences, the California Historical Society, the Center for Research Libraries, the Graduate Theological Union, and the Lawrence Berkeley National Laboratory. All publication dates are included. The database contains over 25,000,000 records, and most campuses update their holdings on a weekly basis.