I've got a couple of questions regarding data synchronization. The architecture doesn't appear to become obvious relating to this:
Does Google Gears provide from automatic synchronization of information (from client to server and server to client)?
Whether it provides for automatic data synchronization, then can one write functions to hook in to the sync mechanism (around the server as well as on the customer) to manage what information is synced?
Can there be anyway to partition the information around the server? for instance a client should have the ability to download only individuals rows inside a table that are associated with it (for instance there's 'hostname' column within the table, and just individuals rows that have the hostname from the client attempting to download the information ought to be downloaded into it)
Essentially things i am searching for is: does Google Gears Synchronization offer the data synchronization features like SQL Server Merge replication? [ http://technet.microsoft.com/en-us/library/ms152746.aspx ]
Gears doesn't provide any synchronization abilities. I believe I recall hearing a podcast interview and among the gears designers if this first arrived on the scene, and that he stated they wanted use a framework to assist with synchronization, but recognized that everyone's synchronization needs were sufficiently different that it might be too difficult to get a 1-size-fits-all solution.
Yes, you are able to partition the information around the server nevertheless, you want... due to there being no automatic synchronization
I initially believed that Gears provided synchronization abilities. But recognized it doesn't.
Should you consider the Google Gears paperwork you'll notice that the neighborhood SQLite database isn't writable for you personally. It's writable for Gears, therefore it can't even assist you with local caching of information that'll be delivered to the server. It's for caching read-only data, at this time.