Evaluation critera were discussed partially in the December 5th meeting (slides).
The goal of this project is primarily learning and evaluation first, then preferrably to reduce the number of choices to a set of preferred choices for the automotive industry. For both evaluation and comparison/selection, we need to define and gather data on relevant evaluation criteria. Let's propose them here:
- Defines own IDL?
- Reuses IDL?
- IDL is Franca, IDL is other, ...
- Defines data encoding (datatype to bits)
- Defines a (stable) programming API
- Has code-generator
- for C++,... Java... etc.
- ... (continue)
Project Evaluation Criteria:
How much is the technologies being used by other parties (not necessarily related to automotive). This is important because the vehicle is supposed to communicate with external entities (might evolve in the future).
- Performance: This is generic as criteria are not established yet. Can the chosen technology meet certain performance goals (example: speed, bandwidth, CPU % etc...).
- Popularity among automotive companies
- Available resources - documentation/support/etc.
- Number of associated implementations
- Quality evaluation of associated implementations
- OS Compatibility, even between Linux variations / hardware variations
- Encryption →Integrity, Confidentiality,...
- Ability to implement access control
- License (Software / Specification, etc)
- Maintenance evaluation
... feel free to add more.
- Scoring criteria as above
- Technical evaluation → feature modeling → comparison
- Scoring of each characteristic as above
- Have you used technology X in Research/Prototype?
- Have you used technology X in Production?
- List preferred technologies (engineer's personal viewpoint)
- Guidelines - which technology to use for specific scope – how?
- List of technologies with bullet-point descriptions
- Tech Briefs and Whitepapers
- (indirect) contribution to GENIVI code projects