Mode problems leading to error

[Richard G]

I have found the simplest and most reliable street shooter is my meterless M2 with a wide angle lens loaded with Tri-X, exposure preset and scale focussed. It is always on. I haven't accidentally left it on 1/30s and forgotten that I am now outside again and I haven't switched back to A. I also don't still have it on ISO 1600 from last night. And I haven't forgotten to code my 21 as a 21 instead of the 50 1.5 I used last night. And I am not shooting precious material in black and white jpegs, thinking it was raw. And, if I have left the lens cap on, I am not stuck in the middle of a 25s exposure plus another long exposure for noise reduction.....All of the latter problems can be encountered on something so simple as the M9. The M240 Leica no doubt has more modes to forget in the heat of battle.

The M2 could have the following mode problems: no film, or incorrect assumption about the film type or speed; few frames left from having not re-zeroed the film counter; film misload (never happens with an M2); lenscap on; wrong shutter speed selected from an indoor shot. The point is, that I expect to have to do everything with the M2 and so make fewer errors. Pilots used to expect to do everything too before automation with the autopilot. Automation of such complex activities has helped them and has reduced human error, but has created a space for human error at critical time points in flight.

This mode amnesia, misperception, seems to have been at the core of the Asiana jet stall and crash at San Francisco airport. The autopilot has multiple modes, and thinking it is in one, when it is actually in another, was missed and saw the pilots complacent about the control of the aeroplane. Apparently a large percentage of jet crashes in the US in the last five years have been due to this mode awareness error.

I thought of this in relation to a recent online video from a senior photojournalist who gave seven tips for better photographs, one of them being to be shooting before the action is likely to start. This engages the photographer who sees the results with his digital camera, and allows him to iron out problems ahead of the main action. And another was to practise, again to be in the groove and know the equipment and its limitations and what that particular camera can do and how to get the best out of it.

This could have lessons for the pilots. Some sort of retraining to manage stalls seems to be high on the agenda, but perhaps also there should be some drill to run through precisely where the autopilot mode selection is and what that is currently taking care of and this could be a routine 20k from the destination.

Here is the article from The Australian newspaper about the mode problems in airline crashes.

http://www.theaustralian.com.au/bus...t-from-autopilot/story-e6frg95x-1226774020871

 

[lynnb]

All valid points Richard; humans are not good at automated process monitoring and then switching to full manual mode when there is an exceptional event. I follow the practice of using a particular camera shortly before starting a critical project, such as paid work. (Changing from one system to another is like having to re-wire the brain!).

It's a philosophical trade-off: statistically, automation may produce more successful outcomes for the majority of operators, whether in aircraft or photography. But mode amnesia, and the de-skilling of pilots and photographers brought on by heavy reliance on automation both increase the risk of things going pear-shaped when exceptional circumstances arise.

In these circumstances the decision on what is the problem, and what course of action is required to rescue it, is made more complex - and time-consuming - when automation is involved. And time is often critical. In flying it is made far more potentially dangerous when the automated systems are compensating for some problem, therefore hiding the symptoms from the pilot.

In photography I once was photographing a wedding with a 5D when the viewfinder blacked out and a strange and rather loud noise came from the camera. There was an obscure error message on the rear LCD. Time was critical as this happened when the bride was just about to walk down the aisle. Of course I grabbed my backup and kept shooting. But it would've been a very unhappy story if I was less well prepared. I didn't have a manual in my camera bag - I had used the camera for some years and thought I knew it inside out - and it wasn't until later I discovered the problem was in the lens (an EF 24-105mm/4L IS), not the camera (the centre element group had failed).

Automated systems can be fine until some obscure error message threatens to ruin your day. Made me appreciate my film Nikon SLRs - simple, reliable, and you are making all the decisions, not the camera.

 

[Richard G]

You made a good point about the bewilderment when a malfunction does occur in a complex system. The temptation, and in a plane, the necessity, is to try to solve the problem. For the wedding photographer the solution is counter-intuitive: drop the problem, immediately, and pick up the other camera. In the Airbus 380 QF32 near disaster de Crespigny writes in his book that flying the aeroplane old style became the only way to be in control when the computers were spitting out multiple error codes and likely misinterpreting inputs from only a minority of intact connections. Your 5d might have been up and running with a lens change, but who would have thought of that? The M9 SanDisk 8G lockup problem requires a separate M9 formatted card to make the camera return from the dead. Won't find that solution on the fly at a wedding either.