I'm no expert but my understanding is that using a seperate memory area (instead of shared between two or more processes) is mostly of use if your process (or another process using the same area) hangs or crashes.
Using shared memory areas would mean that one hung 16-bit process will cause all other 16-bit processes using the same memory area to also hang. Using seperate areas is supposed to help prevent this 'all for one, one for all approach'.
Although the 'help' for 'start' suggests this only affects 16-bit processes, I think it can also affect 32-bit processes as well (although I could be wrong).
My own experience of using shared memory areas is that it can also affect the use of 'RunAs'.
For example, if you open a folder then, from the menu, select 'Tools > Folder Options' then select the 'View' tab you will see an option in the 'Advanced Settings' area for 'Launch folder windows in a seperate process'.
It took me ages (and several fruitless queries in this forum 18 months ago) to discover that this was a 'per user' (not system-wide) setting that needed to be set (checked) whilst logged on as 'Administrator' in order for my 'RunAs Administrator' scripts to work whilst logged in using a different account. My conclusion? This particular use of shared memory most definitely does have a disadvantage!
To summarise, if your script could possibly cause the perl interpreter to hang or if you ever need to run the script using different user credentials then I suggest using a seperate memory area.
Hope this helps...
