We've developed a reasonably tidy solution between two 4D servers in a commercial project, where the content is transferred using a BLOB. For non-4D servers, we're using HTTP transfer and though it works seamlessly, the performance is poor: approximately 30% performance against a local A4D file. I wouldn't recommend it, except as a last resort.
I'd be interested to know more about the HTTP transfer you mention. How does this work?
It uses custom code and the ITK method referenced by Alex.
Regards Mark Howells
