Yes we'd all wish for a true data separation model, one that's native and fast. But the very first reason we want it, is to be able to develop on a standalone database, and then, when ready get all the solution data, complete with correct serial index preservation, in our new solution version in one click.
So here's my proposal that won't need to change anything in the structure, won't force us to trade performance for data serapartion workarounds.
What if, you'd be able to ask Filemaker Advanced, in the "Developer utilities" you could designate a master folder containing all the new solution files (evn if subfolders), and designate a second old solution folder, and then Filemaker Advanced woud pour all the data of the old solution in the new one. Of course that's up to the dev to not delet essential fields etc. If a problem Filemaker could write and error log.
Obviously, like all my resquests this would have to be scriptable, and server side compatible. Because all new functionality must be fully scriptbale and server side scrip compatible (fo the non interface related one of course).
Yes, what I'm asking is for a native RefreshFM from goya whioch is excellent http://www.goya.com.au/refreshfm, but does some settings and would be slower thna a native version.
Moreover a native version will be much more usefull to much moe people.
Also it would tremendously help in case of corruption. We could take an older clone, a remake a clean clone, and then get back the data (cleaned data).
So we would get most of the benefits of data separation, and that's an easy feature for FM Inc to implement.