One thing that I've seen many radio's do, and not locked into one vendor is when they have major problem or low battery (for portables) is sometimes this exact thing can happen.
Also doesn't rule out that as people exited the patrol vehicles that the radio mic ended up weged in between something (wouldn't be the first time).
Its been awhile since I have programmed an EDACS radio, but I am sure that a TOT is an option.
Interesting how this all focuses to equipment. While the technology can provide secondary protection and correction, policy failure seems to be the issue here. Perhaps no policy on standardized programming, as well as the normal one that seems to surface in every situation, no policy on training and continued training for dispatchers / operators of the equipment.
I have yet to see a trunked systems subscriber units not programmed differenently. Every single one I have touched follows a specific template . This is to prevent hand programming each one, or customization so that any one person can pick up a spare or someone elses and can operate it correctly. I doubt thats the issue here.
Policy/Procedues only go so far. Not every failure or incident that goes bad is a policy failure. In our system we hold an entire day class on use and do practical stuff. Works pretty well. We also turn off advanced features that are not used to lessen people playing or accidently selecting features that might throw them for a loop.
I wouldn't throw the policy thing into the wind. Most CA departments are almost over policy'd and Oakland certainly has many of them, amoung a very high crime rate.
FWIW, if you haven't seen previous threads, Oaklands system has had many complete and partical failures in general.