Generally, the described system and process enables resolution of conflicts in a synchronized folder. Within the described mesh operating environment, each of the devices may be configured to do the same processing so that the file system view of the synchronized folder looks the same on all devices (pending local capabilities). Updates that cannot be immediately realized to the local store due to conflicts may be deferred for later attempts when, for example, additional updates at the system level or local level are made to resolve or eliminate the conflict for the update item. Generally, further changes may be propagated by a user in addressing a particular conflict that the user is notified about (e.g., via a selected winner that the user disagrees with). Alternatively, the conflict may resolve itself when a further update occurs that overrides or renders moot the previous update (e.g., a deleted item having a modified enclosure, where the enclosure had a previous concurrency conflict).Depending on the local file system, the described system and process may perform additional fix ups for file name, attributes, etc. In this case, the described system and process may be configured not to update a main replication feed since the updates are local or node application specific. Resolution may primarily depend on the user. When the user does resolve a conflict, the feed may be updated and the resolution propagated to all nodes.