>Erik,
>
>The way I understand DCOM is that object is actually instantiated at the client's machine. This scenario would be much slower. Please correct me if I am wrong.

I don't know that that is necessarily true. For DCOM to work, the object has to be registered on the client machine, and so the interface is defined there. All operations on the client operation are actually manipulated a proxy stub that marshalls the request and response to the real object that is running on the server. This process is pretty quick. I have not compared or seen comparisons done between DCOM and other methods of remote object manipulations.

As far as XML being inherently faster over the wire, I doubt it. XML, especially for sizable datasets, suffers from the considerable bloat inherent in it's syntax (opening and closing tag pairs).


>>DCOM, or HTTP?
>>
>>
>Does this mean the form can be set to private datasessions and MTS will handle the rest?

As far as datasessions in the client interface, this is completely irrelevant (my efirst answer was addressing Datasession for server object). You don't need data sessions at all unless you are translating server data into VFP cursors on the client. If you are doing this, I do recommend private datasessions, but only for all the same reasons I recommend private datasession to anyone, even for single-tier apps.