Feasibility test of using trello as project management and issue's backend

[hughie]

We are discussing about using trello for project backend. I mentioned on last meeting. Julian agreed with using trello. but, when we are planning how to integrate with trello, then several issues are occurred.

• Account association(session)
  • Every page has checking code for trello
  • User token management
• Dependency with trello
  • We like trello, but if we are integrating as project management and issue backend, then it is not small dependencies.
  • User should have trello account.
• Synchronization issue
  • If we make limit like it only can change one side(CENTO or Trello), It is easier, but It is inconvenience. It's seems not good solution.
  • If we are making both side can be update each other, then it has many technical issues. It is difficult for us, now.
• Database handling
  • Trello is providing good Restful API but, we should use fixed filed which is trello defined.
  • Synchronization issue

So, we want to compare with making our own user interface.

[adamantivm]

@hughie, is this to analyze using Trello as backend database only or also
as project management UI?

[hughie]

@adamantivm Actually, we considered both. we also want to show in different way with Trello. Because, we think Trello is not showing relationship between tasks. I thought U/I Implementation and DB handling is taking much time at the first time. Now we know it's easier then we thought.

[adamantivm]

@hughie I think that makes sense. I would suggest if it is not too much problem to subdivide or create sub-issues / related issues for the specific (1) back-end for task management and (2) front-end integration.

For the front-end integration, I agree with you that Trello would only be able to show a subset of the functionality, like for example it would be missing relationship between tasks.
What I think would be interesting is to explore the possibility of using Trello's default UI to visualize boards that make sense to represent part of Cento, even if it is incomplete, and also allow to do modifications (e.g.: add comments, archive cards) so that when you go back to Cento, then those changes are reflected in Cento and still make sense from the Cento point of view.
I think this is possible, but we need to experiment.

As a final note: we need to be careful to not get too distracted about this for the Teaser. I don't think this is necessary to be able to show a demo, but I do think it is necessary for the Design Plan.

[hughie]

I would suggest if it is not too much problem to subdivide or create sub-issues / related issues for the specific (1) back-end for task management and (2) front-end integration.

Okay. It seems more clear for discussion.

For the front-end integration, I agree with you that Trello would only be able to show a subset of the functionality, like for example it would be missing relationship between tasks.
What I think would be interesting is to explore the possibility of using Trello's default UI to visualize boards that make sense to represent part of Cento, even if it is incomplete, and also allow to do modifications (e.g.: add comments, archive cards) so that when you go back to Cento, then those changes are reflected in Cento and still make sense from the Cento point of view.
Synchronization of both side

Our idea was similar. If we are using trello, then we can try it this way.

Now we are just checking if we are using trello, then what can be issue for implementation and if we are making our own frontend how difficult it is.