Couple of container groups expressed the need for a push model when dealing with content. The use case is like this:
- The user installed an OpenSocial app from a 3rd party onto the container site. - There is strict SLA level on the container site which requires the app to be rendered within a given time frame. This requirement leads to the need that all content used for rendering the view are locally stored. - The user performed some activities on a 3rd party site which results in some data about this user changed. - The app's view on the container's site needs to reflect the changes when it is rendered. There are several approaches for this problem: 1. Push Model: There are two alternatives in this model: a. Facebook model: Facebook provides setFBML api to let developer push personalized app view code to the container. This view code is stored inside the container and ready to be used when the app is rendered. b. Push Content Model: Instead of pushing the complete personalized view code, only the data got pushed. View's code is shared cross users. 2. Fetch/Cache/Invalidate Model: In this model, data are fetched and cached. TTL are respected in determining the life of the cache record. The container is also required to provide an api to allow developers invalidate certain cache record. On top of this, the container can decide wheither and when to do a pre-fetch. Are there any other approach for this problem? Which approach is most optimal or do we need a hybrid approach? Any other thoughts? -Charlie --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Implementing OpenSocial Containers" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/opensocial-container?hl=en -~----------~----~----~----~------~----~------~--~---
