How to ensure that a bug is fixed in a timely fashion, given that the Polarsys WG does not have enough ressources to provide such services ?
A solution could be the following:
- Users that want professional support subscribe to a professional services offer provided by a Branded Polarsys Provider, which could include:
- Advice, explanations regarding the use of the tools;
- Bug fixes;
- In order to distinguish from the Polarsys WG and therefore be able to have a sustainable business model, Branded Polarsys Providers must offer additional added value, which could be :
- the provision of validated binaries that are not provided by the Polarsys WG, and
- the provision of customer-specific releases when there is an urgent and important bug to be fixed;
- To provide for VLTS, support contracts should include the following best practices:
- All patches should be given back to the Polarsys LTS branch,
- unless there is a good technical or confidentiality reason provided.
- The Polarsys WG would provide to its members a demo version only to avoid competing with Branded providers.
- Support Contracts are accessible only to Polarsys members ?
- who controls the Polarsys VLTS branch ?
- What if there is no Branded provider in a given area ?
Feature requests should be treated the same way:
- When someone needs a new feature, he does it as part of a contract with a Branded Polarsys provider;
- The development should be given back to the Polarsys LTS branch, unless there is a good technical not to do so; this way, users are sure not to pay two times for the same feature;
- Is it possible to find a system that allow users sharing the development costs ?
- If we use a system based on good will, we are likely to end up in a situation where urgent needs are funded by individual users, no matter the Polarsys roadmap, and non-urgent needs will remain unfunded forever;
- A mandatory system is likely not to be approved.
Configuration Management is a transverse issue to Bug Fix and Feature Requests
Too many versions tends towards a custom development model and lowers economy of scale. In order to save money and allow affordable support, we need to tend to COTS.