Here are two possibilities. 1) There is a copy of the .scx and .sct files in the same folder as the exe on the client's machine. 2) Someone made a build, fixed a bug and re-created the build with the same version #'s and then dated that new build to match the buggy one. That way the mistaken build can be covered up or it was done that way cause it seemed like a good idea at the time.

>>>>>>Has anybody ever heard of this?
>>>>>>>
>>>>>>>We installed a copy of our app at a client's office on 3 machines. On double checking, their copy of the exe is the same date, time and version as the one on my system.
>>>>>>>
>>>>>>>However, their tab order on one screen is different - different from mine, and different from the tab order defined in my development environment for that screen.
>>>>>>>
>>>>>>>There is no code in the Valid or LostFocus events of any of the controls that might change the focus of the 'next' control from the normal tab order.
>>>>>>>
>>>>>>>I'm at a complete loss in trying to come up with any possible reasons why this might be happening. In fact, it seems downright impossible.
>>>>>>>
>>>>>>>Anybody got any ideas?
>>>>>>
>>>>>>Are you sure that the machine put on the queue one TAB for every keypress ?
>>>>>
>>>>>I know it runs properly on my pc and on my colleague's pc (he doesn't have the developement environment, just the runtimes). I can only assume that if we have the same exe, we should see the same tabbing.
>>>>
>>>>you are a lot of optimist.
>>>
>>>I know what you mean, but thinking any other way makes me dizzy. ;)
>>
>>I understand, but the patter that you have shown (if I have understood well)
>>it happens with one shot him 4/5 tabs.
>>As I have seen more than once Excel go crazy in the movement of the cells...
>>
>>It checks that the window behaviour for the disabled persons it is not active.
>
>I'm going to go there tomorrow and see what the rest of the pattern looks like. I want to know if it's skipping tabs forever, or if it gets those skipped ones in a later part of the order.