Why consider a New Environment?

• existing Data Analysis Environments have been designed more than twenty years ago
• they have never been thought as "global environments", but as data analysis systems
• their algorithms are oriented towards a particular wavelength range (optical, vs. radio, vs. X-ray).
• often use internal data formats, and only revert to FITS for data import/export
• user Interface was not an issue at the time of design. Efforts have been lately made to "superimpose" GUIs to such systems, but they are still not part of the system
• they have no facilities for web access
• addition of new algorithms is far from being straightforward, even if most of them have scripting language facilities to chain existing operations
• they have no data base facilities, nor direct interface to data bases

Adding the specifications outlined in the UserScenarios on such systems would require a very big effort, and the result is uncertain.

Designing a new environment in which the UserScenarios specifications are taken care of already at design level, would allow to embed such facilities from the beginning.

At the same time, such environment could inglobe the existing data analysis systems, allowing for communication between them, thus treasuring the work done

Why Free?

• commercial general purpose environments include many, but not all the facilities outlined in the UserScenarios
• commercial products are usually "close" and do not allow extensions
• a commercial product esplicitely tailored for our needs would be enormously expensive, as the number of potential users is low (compared to word or excel potential users)
• software freely distributed can benefit from the contribution of a wide community (not necessarily limited to the astronomical one), and can thus grow in an exponential way (linux and linux based applications are a powerfull example)

Why Open Source?

-- BiancaGarilli and PrebenGrosbol - 13 Sep 2004
to top