▪ The AJAX request operation is a self-defined operation and does not support an operation defined in a processor. This is because a self-defined operation has a better reusable granularity.

▪ In an AJAX request, it is suggested to just do READ operation with the out-scope context. This means that, for a self-defined operation, it maybe leverage two part of context to finish task, one is the context exactly defined in the self-defined operation, another is that it assume that it will be provided with out-scope context during the runtime. E.g. AgetBalance operation needs to leverage the user account info that is just available during the runtime, when the operation executed, the proper context tree will be build, so it can get the useraccount info directly. So, you know the Ajax operation can control the out-scope context during runtime, and maybe many concurrent Ajax request access the out-scope context at the same time, it is dangerous to update the out-scope context in an Ajax request, unless the developer confirm the security.

▪ In a page, more than one AJAX request can be processed at the same time. If a customer wants to navigate to the next page, HTML channel requests cannot be processed until the processing of all active AJAX requests is complete.

▪ When an HTML request is being processed, all incoming AJAX requests are blocked. In an HTML page state, all AJAX request previous state triggered will be ignored.

▪ File upload data definition must be exactly align with the linked file handler's implementation. Different handler may need different data structure.

▪ Only one file can be uploaded through the AJAX channel at the same time. If more than one file needs to be uploaded through the AJAX channel, the files must be uploaded one at a time.