|Group refactoring project||Florian Gnägi||6/22/12 1:53 AM|
Dear OpenOLAT community
I'm glad to announce that we started a project to implement a fundamental refactoring regarding the group data structure in OpenOLAT. We are aware of the problems in the group management for a long time now and I presented the necessary changes several years back at one of the OLAT user days in Zürich. Until now we never had the chance to actually implement those changes.
The main problem with the group data structure is a weird concept called group context. Instead of having direct relationships between groups and courses, groups are bound to the group contexts and the group contexts are bound to courses. This is all fine as long as one only works with course internal groups. Unfortunately, in many scenarios this is not the case. E.g when your real-world domain language uses the word "class" rather than group, you usually want many courses to be associated with your class.
Today you could theoretically use the so called course comprehensive groups. But this is so weird and leads to so many errors because people do not understand the concept of the group context that we hide this course comprehensive group management tools for our clients.
The aim of this refactoring is to finally solve this dilemma and free the groups from the group context cage.
After the refactoring:
- Groups are fist-class citizens like courses, independent of courses or contexts
- The group types (project group, learning group, right groups) are gone. Why decide when you can have the features in any case? It is just a matter of the rights you have which group features you can enable.
- OpenOLAT will have a global group management tool, all groups can be managed by admins
- The course internal group management stays as it is
- Learning areas are a course internal concept not sharable across courses
- Users will not see big changes, the changes are more visible for admins / teachers.
The migration of existing groups will be done automatically when installing the new release. If you have modified OpenOLAT in that regard, please contact us to make us aware of potential problems. We encourage you to start with early builds as soon as the code is available to give as feedback about issues you might encounter.
The time frame for this refactoring is summer 2012. We expect to release it somewhen in Q3 / Q4 2012, most likely as part of OpenOLAT 8.2. With this refactoring done, OpenOLAT will be much better suited for many scenarios and simpler to use.