>I find it interesting that using the "=" seems most corrosive and that any of the variations
>actually make a difference (not doubting the submission, just see it as odd).
>In the back of my mind I recollect one bug report that was bypassed by inserting the "=" in
>the method call and yet another that was bypassed by coding the parens. at the end of a method
>call, so these things obviously *can* have some impact on execution.
>
>Do you have any considerations on that aspect of the report?

I think the = sign got around a problem with Outlook Automation. The other sounds vaguely familar, but I can't quite place it.

Of course I'm taking this into consideration, but my test did hundreds of requeries with no waitstate and the syntax that apparently causes the most trouble, and didn't hit the error. It might be a question of speed on my machine, my network, my server... but until I have code that definitely causes the problem, it's less than useful to report it. If it fails somewhere, that gives me a place to start looking. The code I put together hasn't failed yet, so I can't dig into it looking for problems.