Yep, getting the error on that as well.
Hmmm - I must have been smoking something yesterday when I tried doing “set x to 1” because that’s not working now although I thought it was working yesterday. In any event, there is definitely something wrong here.
Wow-- I’ve found a configuration that works: if the host machine is run on Intel CPU AND the AppleScript Applet is compiled on intel (universal binary), then the client running 10.6.x can send remote event and get a response (see my simple test scenario above).
Now, the thing is, I could swear I tested this configuration way-back-when the problem first cropped up; it’s possible this only started working after 10.6.3. On the other hand, that was quite some time ago, my memory may be wrong…
Anyway, it looks like deploying the applet on an Intel host is a solution. Just for grins I also tried the re-compiled (universal binary) on my original host (an XServe dual G5) and it still doesn’t work, so it appears to not be just a compilation issue-- it must actually be running on intel architecture.
Interesting - thanks for the information. I don’t have another Intel Mac running Leopard to test with so I didn’t try that config. I’ll add it to the bug report at Apple, but unfortunately this pretty much cinches that it won’t get fixed. And, it doesn’t help my users, many of which run our software on older hardware that’s dedicated to the purpose.
Sigh… 