Thanks for following this up Colin it is an interesting response. At least people are aware of this issue. Many thanks for all the other comments especially about the icing problem.
I’ve just bought one of these, tried it our for the first time this weekend (in woods - no snow here unfortunately) and very pleased I am with it.
I’ve tried reproducing the fault mode described by Colin and, despite being as careful as possible with positioning the slider, I can’t do it. There is a preliminary flash when the slider is half in position, but I can’t get it to go into its bootup sequence without fully engaging the lock. I don’t know whether this is updated firmware or not, but I don’t see the problem on my unit.
I did find a problem in the practice scenario though. When the first victim is located, if that “victim” then moves about, e.g. while the search is still active for other burials, the multiple-device identification does get confused and starts showing more burials than are actually present. It takes about 30-60 seconds to recover from this and show the correct number of burials. This is probably less of an issue in a real rescue scenario though.
I’ve been testing 2 of my units for about a year now.Although I try to be a thorough and fault-oriented, I have never encountered any problems with the switch. Probably because I haven’t been looking there. Now that I know about the possibly dangerous course of events, I can only say two things -
1 - True, it is possible to inadverently turn off a seemingly working unit. But:
2 - To do that - you have to be pretty deliberate.
Never ever has my unit switched on without locking the switch securely at the same time. If you turn it on the right way - by pushing the protruding switch bar into the casing - it will always securely snap and stay in ON position.
As a bottom line - It is my firm belief that everyone is responsible for his deeds - which includes properly wearing and using his equipment.
Some things my be bombproof. But nothing if foolproof ‘cause fools are so inventive 
Be safe out there…
Odd how some people make such a big deal about a potential failure that requires almost deliberate action to achieve, yet just a few stray drops of water that refreeze can render the beacon incapable of searching—I repeat my previous post below:
*****
True, the Send mode is activated on my two Pulse units before the switch is *locked* into Send mode, but locking the switch into Send produces a very solid-feeling and fairly loud-sounding click. By contrast, sliding the switch into Send yet not allowing it to lock into the Send position requires an almost deliberate and fairly delicate action. Plus the switch is still visibly protruding from the beacon housing.
Overall, I think a Pulse owner should be aware that the switch should be locked into Send mode with a very clear click, but that’s already very obvious even after turning on the beacon a single time.
By contrast, just a few drops of water onto the sliding switch, which then disappear below the switch and can freeze, will render the beacon incapable of being switched into Search or even Off without a few minutes of rewarming first. (This actually happened to me in the field because of a partially unzipped jacket during an extended period of light snow.)
So any Pulse owner has to be careful not to allow water to get onto the beacon, and if any water is suspected of being in there, periodically switching the beacon briefly to Search might be a good idea to prevent and/or break up any hidden internal ice formation in the switch.
Atom
MyYahoo