Basic Functional Requirements for Cross Search Service
M.Moffat (M.Moffat@hw.ac.uk) -Ver 1.0 (25/05/07)
Contents
- Introduction
- Functional Requirements for End User Interface
- Functional Requirements for Administrative Interface
- Functional Requirements for Machine to Machine Interfaces
1. Introduction
The PerX pilot service was used as a test-bed to explore the practical issues that would be encountered when developing a full scale subject resource discovery services. As a result of setting up and maintaining the pilot a list of basic functional requirements for a full scale subject based cross search service has been produced. The functional requirements are divided into three main areas - requirements for the end user interface, requirements for the administrative interface, and requirements for machine-to-machine interfaces.
2. Functional Requirements for End User Interface
2.1 Search Functionality
- Basic Search Functionality
- Search via keyword(s)
- Search via resource type category (e.g. articles, technical reports, thesis)
- Support for Google like syntax (e.g. +,-,?)
- Advanced Search Functionality
- Exact phrase searching
- Boolean searching (AND, OR, NOT)
- Ability to specify individual and groups of collections for cross searching
2.2 Search Results
- Viewing of search results from individual collections
- Viewing of search results from merged collections [not achieved by PerX]
- Relevance ranking of search results
- Provision of full text availability indicators
- Ability to edit previous searches
- Ability to refine search results (e.g. via result clustering) [not achieved by PerX]
- Record selection functionality (e.g. ability to store selected records) [not achieved by PerX]
- Outputting of Selected records in formats suitable for reuse (e.g. RSS, XML, plain text) [not achieved by PerX]
3. Functional Requirements for Administrative Interface
3.1 General
- Ability to accommodate a range of target types (including Z39.50, OAI-PMH, SRU/SRW, Bespoke APIs etc). [see sections 3.2 and 3.3 below]
- Ability to add new target details (e.g. name, description, logo etc)
- Ability to delete targets
- Ability to browse targets details
- Ability to search targets details
- Ability to temporarily suspend targets
- Secure administrator access
- Limited functionality guest access
3.2 Harvested Targets (OAI-PMH)
- Manual Harvesting
- Ability to manually identify, harvest, normalise and index OAI-PMH targets
- Ability to specify particular sets for harvesting
- Ability to add/delete sets for harvesting
- Ability to delete entire harvested content
- Ability to delete selected harvested content (e.g. individual sets)
- Ability to deal with XML errors in harvested content
- Ability to perform general safe transforms ('normalisations') on harvested data [see OAI-PMH Metadata Augmentation Report]
- Ability to enhance harvested metadata via collection specific transforms [see OAI-PMH Metadata Augmentation Report]
- Automated Harvesting
- Ability for administrators to automate the harvesting process for OAI-PMH targets [not fully implemented by PerX, see Setup and Maintenance Report]
- Ability to specify automated harvesting type (e.g. incremental or full reharvesting) [not achieved by PerX]
- Ability to specify automated harvesting periodicity (e.g. weekly, monthly etc) [not achieved by PerX]
- Automated harvesting logging facilities (e.g. dates of previous harvests, harvesting details etc). [not achieved by PerX]
- Error handling and administrative reporting of harvesting process.
3.3 Distributed Search Targets (e.g. Z39.50, SRU/W)
- Ability to manually test distributed search targets in real time.
- Automated testing of distributed search targets with reports to system administrators [not achieved by PerX].
4. Functional Requirements for Machine to Machine Interfaces
4.1 General
|