I don't have a link, sorry. It was sent to me directly from Trimble's customer service, although it may be available online somewhere. I've attached the file sent to me, and these were the instructions given:
"In regards to the ongoing issue with the offset points being shifted after check in with GPS Analyst:
Success!!
I received a fix from development today. I have tested this fix and it does resolve the shift issue!
Make sure that you have GPScorrect 3.15 installed. You do not need to re-install if it is already on the unit.
Using ActiveSync or Windows Mobile Device Center, copy the attached cab file over to the my documents directory on the Geo.
On the Geo, use file explorer to find and double tap on the cab file to run it.
The offsets shift issue will now be fixed. When data is checked in with GPS Analyst , the offset points are no longer shifted.
P.S. make sure that if you have your declination set correctly on the Laser. It is advised to have it set and applied on the laser as opposed to setting it in ArcPad. ArcPad will not remember the setting after it is closed, so it must be set over and over. Make sure that it is NOT set on both the laser and in ArcPad ( double applying declination)."
A couple additional notes, I'm not sure why they sent it as a cab file, since you'll probably find that you need to rename it to an exe before you can copy it to the handheld and run it, I did anyway. Also, I haven't had a chance to test it with ArcPad 10.0.02 yet. I'm also not sure of the compatibility with different products. I'm using a 2008 series GeoXH, Trupulse 360B, ArcPad, GPSCorrect, and GPS Analyst. Obviously, I had nothing to do with the development, and so I can't say what else it will work with. Use at your own risk.
About the averaging of the GPS offset point, after firing the rangefinder, ArcPad reads the signal and pops up the offset menu. If you then click on what I'm going to call the survery button (the little tripod looking button), it opens a new menu that allows you to recapture the point (by tapping on the satellite) and does use averaging. But, I'm not absolutely certain what it's averaging. Could this be a method of averaging the offset point? My understanding is that the rangefinder already introduces some error, so I'd hate to add more by only being able to get one shot.