Database Schema Changes

Does anyone have any details on the client server interactions of making database schema changes?

The reason I ask is because, while developing a solution file on my laptop using wifi, the wifi has on two occasions dropped out in the middle of processing a schema change.

This means the file becomes totally corrupt, and I have to go back to a backup file.

I would have thought that that sort of thing would have been transactional, ie the FM client sends all changes at once to the server for processing, the server checks it's a complete and valid request, then all the operations done exclusively on the server, with it just sending informational progress updates to the client.

Instead of what appears to happen is the client controls the changes as it goes. Meaning any network connection loss between that server and client results in corruption.

It would be interesting to know if there is any particular reason for the method FM currently uses over a more transactional style approach?

Layton