Lightweight Web-Based Conzilla
A read only and web-based edition of Conzilla is desirable for various reasons. Currently, displaying a context-map requires either a full installation of Conzilla, or alternatively an applet to be downloaded and executed from within a web browser. Both options are expensive in time and resources, if compared to a web site loaded and displayed by a browser. Users who just want to take a look at a context-map do not want to wait for an application to be loaded.
Another reason why a fully-featured application is often not necessary is the target group. Not everybody needs the possibility of editing a context-map, a viewer just wants to look at it and it's associated information. A read only tool with the basic and most necessary features would speed things up and make browsing context-maps a lot easier. The only requirement for accessing published context-maps is a current web browser, which is in place on the viewer's system most of the time anyway.
One of the integral parts of a lightweight web-based Conzilla will be the integration with the Collaborilla service and the availability of RSS feeds. With RSS feeds an interested person can subscribe to information reflecting the personal interests. E.g. a feed with information and links pointing to published context-maps of a specific author or contributor, a feed containing maps about a single subject, or a combination of several search terms.
- Read only mode, the presentation of context-maps is the main purpose.
- No Conzilla applet is necessary for showing maps and immediate display of context maps, this will speed up things dramatically.
- Users are able to browse through maps like they browse through web sites.
- Context-maps can be fully integrated into web pages, comparable to images. Alternatively they can be viewed as stand-alone pages.
- Quick links (short and direct URLs) for fast distribution of maps.
- Collaborative features, such as integration with the Collaborilla service and RSS feeds for updates, contributions, etc.
- Possibility to make annotations from within a browser
- Required: Conzilla server and Collaborilla
- Expose via WebServices (or another AJAX compliant technology)
- XFire http://xfire.codehaus.org, or
- Jetty, or
- something else
- Conzilla server (like Conzilla fat client, but no GUI)
- Communicates with Collaborilla
- Caches context-maps as browser-readable data (PNG, SVG, ...)
- Provides coordinates for imagemaps
- Enhance with search functionality (necessary for RSS)
- Eventually switch to a JCR (Jackrabbit)
- AJAX toolkit (GWT?)
- Develop new markers for information on concepts
- Small symbols which indicate additional information, annotations from visitors, etc
- Imagemaps (or perhaps server-side maps would make more sense?)
(Perhaps we can do extreme prototyping, see http://www.onjava.com/lpt/a/6797)
- Static prototype
- Most important pages
- Extended static prototype
- Logical data model
- Dynamic prototype
- UI adapted for chosen framework
- Working navigation
- Complete working UI without an implementation
- Service implementation
- Specifiy new APIs or adapt old ones, depending on the requirements
- Develop and document APIs
- Implementation of each service
- Integration of UI and backend