Free vs. Licensed - Open vs. Restricted

When considering a future environment, it is important to state clearly to what extent one requires a freely available system and at what level dependencies on products with associated license fees are acceptable. Our considerations are:

• that there should exist a base version of the environment which can be used without the need for paying license fees. This version should be fully functional but may not be optimized and not provide some optional features. This would ensure that even institutes and groups with limited funding can use it for their astronomical research.
• that it should be possible to extend and enhance the basic version with commercial products. The two main areas would be to a) replace free base implementations with optimized commercial versions, and b) provide additional features which are considered desirable but not mandatory.

Another issue is whether or not to accept any restriction in the access to code and specifications. Our preferences are as follows:

• all science code should be open source. It is essential that the scientist can check the actual code if in doubt of the results. Pure mathematical functions may be an exception to this if they can be verified by independent means.
• all interface specifications should be openly available and eventually controlled by an independent body. This would ensure that alternative implementations of the environment could be made (e.g. with special optimization) and still be compatible with already available versions.

-- PrebenGrosbol - 21 Oct 2004
to top