Col-an-der n. a bowl with a perforated bottom for straining or rinsing foods.
[Latin colum sieve]
It can be used like the various programs and portal sites that collect and summarize newsfeeds, but it is more flexible than that. Instead of just newsfeeds, it can take any arbitrary data, and can make sense of it if given a plug-in for that particular resource. Currently, it can understand and parse XML, RDF, RSS, HTML, FTP, and raw streams of bytes. This covers a lot of ground. Since plug-ins can access the Internet, they can be used not only to gather information, but to send information out, trigger events on other servers, and run programs. All of this is done within a controlled security environment.
As a very simple example, this summary HTML file was created by running Colander.
Users have accounts containing assignments that specify all of the tasks they want performed. They can then log in with a standard web browser and issue a command to perform those assignments and subsequently view the results. All functions, account maintenance, and file transfers can be managed with any web browser. The server is designed to operate on a remote host supporting multiple users, but it is envisioned that most people will run Colander on their own machines.
The motivation behind this project was a long standing desire to automate the various tasks I tend to perform every day. What makes this unique from personalized portal sites is its flexibility and that control remains with the user, providing benefits such as better availability and privacy.
Colander is free software released under the GNU Lesser General Public License.
Colander requires the Java Runtime version 1.4 or higher.
Colander and its source code is available for download here .
The jar file is self-extracting. Type
java -jar Colander-version.jar
or double click on the jar file.
A list of answers to Frequently Asked Questions is available here.
Please see the project page for up to date news, support, and information.