Collaboration - Gamified

This one is very similar to the Collaboration service, but this time I led my imagination flow a bit more. It could be a great tool for decentralized companies, where almost random people could subscribe to a task, where their contributions are partly awarded automatically by the system and partly by other contributors. In theory, it could work not only exclusively for software development.

MVP

• possibility to make tasks, which can consist of subtasks (TMO ontology)
• possible parts:
  • bool: backend or frontend (functional and user test)
  • functionality conditions
    • backend/frontend functional: input-output mapping (unit tests)
      • generate RDF by chat^^
    • frontend user test: some questionnaire?
  • var: finished/unfinished/under review/etc
  • submission date
  • last edited date
  • dates for all state changes
  • backlog-item (containing state-changes)
  • project defined labels
• a user should be able to assume a task, and submit a task submission
  • reference to task
  • contents
  • state (finished, only after successful test)
  • backlog
  • user defined labels

If all subtasks are successfully finished, then the super task can be tested too

If a original task is set to deprecated, then developers working on it should get a notification - or maybe also when there is discussion on the whole part.

Risks

• A lot of redundancy, not very efficient with manpower

  • Possible solution: Only forward an idea to the next round if it's deemed sufficiently promising (who decides that?)

• Fragmentation without a concrete end result

  • Possible solution: An expert panel could be helpful here.

• Too much focus on a single proposed but popular solution, due to lack of manpower

  • Possible solution: Involve critics as a fixed part of the process. If they have serious objections, they could call for an alternative; these could be, for example, hackers.

• People become demotivated if their solution isn’t eventually chosen

  • Possible solution: Following a gamification and "all contributions are valuable" philosophy, always give some kind of reward.

• People feel it's too reductionist of an approach

  • Possible solution: I think people should be able to decide when code should be addressed. This can be proposed by a user by establishing the input-output mapping.

Additional ideas:

• People should be able to vote on which areas need improvement.

  • Experts would get a separate status.
  • This should focus on different types of improvements (list from Chat):
    • Functionality
    • Security
    • Efficiency (scalability)
    • Architectural integrity
    • Maintainability
    • Observability (check for errors, etc.)
    • Portability/compatibility
    • Reliability
    • Testability
    • Compliance (legal/ethical)
    • Documentation
    • External impact

• What else is needed: A similar setup for interpreting the code.

  • This is necessary when discussing the impact on communities.

• Sometimes, you'd want to have a round for code improvement or even just migrate everything to a different (more specific) form (cleaning up).

#service