Weekly Roundup, Friday 27 March 2015

Good morning

This is (almost) a single topic newsletter this morning.  The usual varied compilation will be offered again next week.

It is amazing, isn’t it, how we can change our perspective on air travel in an instant.  There we are, sometimes going a year or longer without any fatal crashes, at least not within ‘first world’ countries or in scenarios that we feel have any relevance to us, then all of a sudden, a single crash and we’re looking askance at the safety of air travel entirely, even though it is and remains by far the very safest form of travel possible.

The Rush to Publish, Facts or Not

The shocking nature of the Tuesday A320 crash into the French Alps certainly gives us all pause for thought.  And perhaps pausing, both for and before thought, is something that is called for here.  Journalists now have ridiculous deadline pressures – they don’t have a daily deadline for tomorrow’s newspaper, nor a several time a day deadline for the next scheduled news broadcast.  Instead, they have a fresh internet deadline every second of the day, and enormous pressure to get stories ‘out there’ as urgently quickly as possible, with their competitors not only being other ‘real’ journalists but also the burgeoning ranks of, well, let’s be kind and call them ‘citizen journalists’ – people who gleefully post and tweet whatever comes their way with very little quality control or thought.

This meant that some news outlets felt pressured into cheating, and published video and still images purporting to be of the plane and crash scene which in fact where nothing of the sort, as revealed in this article.

It is one thing for mistakes and misunderstandings to occur – that’s pretty much what passes for normal in early reports of any air crash, but deliberately faking images is a step way too far.

Is it Really as Simple as We’re Now Being Told

Now, for the stunning apparent cause of the crash, itself, and there have been some assumptions made about that, too.  They may prove to be correct assumptions, but we need to understand exactly how much we do know and what we don’t know, and clearly separate the facts from the guesswork.

We do know that the pilot left the cockpit, and then when he came to return back to the cockpit, was unable to gain entry.  Fact.  We do know the copilot did not respond to the pilot’s requests to be readmitted, and we know the copilot, while silent, was still breathing.  Fact.  We also know that the autopilot was set to descend the plane from 38,000 ft where it was cruising, down to 96 ft (with mountains some thousands of feet below rather than land at sea level, this would guarantee a crash if not countermanded at some point in the descent), and the plane obediently flew down that descent path, albeit making a few complaints towards the end about the pending crash.  Fact.

But we don’t yet know that the copilot did this as a way of committing suicide.  We do know that unfortunately it is not as rare as we would wish that a pilot flying a plane full of passengers does commit suicide by crashing his plane and killing everyone.  Fact – see this article.

But we don’t know why the copilot apparently reprogrammed the autopilot and refused to allow the pilot back into the cockpit.  We also don’t know why the pilot was unable to get back into the cockpit.

There’s one explanation that is being overlooked in the rush to blame the copilot.  There is a possibility of the plane’s cabin pressure dropping, either suddenly or slowly.  In such a case, the onset of hypoxia induces a state of drunken euphoria, and people do stupid things.  If the plane quickly depressurized, there is only about 15 seconds (at 38,000 ft) before this sets in – that doesn’t leave much time to urgently put on your oxygen mask, and then, after the 15 seconds, you might decide you don’t actually need the oxygen mask because ‘you feel fine’ and maybe it is just an instrumentation error.

It is possible that a depressurization event caused the pilot to forget how to open the cockpit door, and caused the copilot to do foolish things in response – simply arranging for a not particularly steep descent and hoping that very soon, he’d have enough oxygen to be back to normal and able to pull the plane out of its descent.

This possibility is well explained in this article.

I’m not saying that the depressurization theory is more (or less) probable than the copilot suicide theory.  But I am saying that until we know more – in particular, until we get the flight data recorder ‘black box’ decoded, we shouldn’t rush to judgment and vilify the memory of the copilot.

An Equally Ill-Considered Rush Also For ‘Solutions’

By the same reasoning, nor should we think that by instituting a ‘two person always in the cockpit’ rule (as has been hurriedly introduced by most airlines around the world), the problem is now solved.  We need to first be certain what the problem was, and if the problem is the copilot’s mendacious act, then how do we solve that problem.  Does putting a flight attendant in the cockpit while one of the two pilots is out and possibly some minutes away (remember that 15 second thing with depressurization) guarantee this type of disaster (manmade or not) won’t occur again in the future?

If this is such a good idea, one also wonders why it was so selectively observed in only some jurisdictions and by some airlines prior to now.  As shown above, pilots suiciding by crashing their planes is nothing new.

Do you really think that a pilot couldn’t sabotage a plane in under 30 seconds, right in front of a flight attendant, and in a way that the flight attendant couldn’t rectify before it was ‘too late’?  (Possibly including the incapacitation of the flight attendant – remember many pilots carry firearms.)  I could certainly do that, and most other pilots, if they wished to think it through, could do so too.  And as for the pilot somewhere else in the plane?  I could make certain that by the time he possibly returned to the cockpit, it would be way too late.

Oh, there’s also the matter of the cockpit door itself.  Why have a ‘clever’ high tech electronic lock on it if it can be abused by the people in the cockpit?  Why not have a good old fashioned keyed lock instead?  (Answer – because terrorists would probably get keys to the locks before too long!)  But by making the cockpit semi-inviolable, it seems we’ve substituted one problem (terrorist hijacking) for possibly another (pilot hijacking).

Other Things at Fault

Let’s also think about the situation where the autopilot was as stupid and singleminded as to agree to fly the plane smack bang into the side of a mountain that it knew was there – even while other parts of the plane were shrieking out warnings of its pending doom.  Heck, we are now seeing cars that are clever enough to drive themselves from coast to coast, complete with traffic and crazy drivers around them – avoiding a mountain is child’s play compared to the complexity of that.

Can’t we make a plane clever enough not to fly into the side of a mountain that it either ‘knows’ (from a GPS terrain map) or can see (from radar, radio altimeters, or even cameras) is in the plane’s path?

This also brings up another question that it is timely to revisit.  Who do you really trust to most safely fly your plane?  A computer, or a pilot?  Computers can and do fly most of most flights already, and it is possible to have a plane under computer control from the start of the take-off roll at the departure airport to the end of the landing roll at the arrival airport.  I’m not sure if the plane can be computer controlled on the ground, but really, that’s the easiest part of the puzzle and could be resolved very quickly if necessary, too.

My perception is that (these days) many more accidents are caused than prevented by the presence of pilots on board planes.  That’s the complete opposite of what most people perceive to be the case, and it’s not what pilots would proudly boast to you, but if we look at recent accident and near accident statistics, it is easy to see many cases where pilots either caused and/or didn’t solve the problem, and difficult to see many where pilots saved the day in a manner which a fully intelligent system couldn’t do as well or better.

Is it Time to Give Up on Pilots – For Our Safety’s Sake?

We’re often told ‘passengers would never agree to be flown in a plane without pilots’.  Now that statement was undoubtedly true 50 years ago.  But is it still true, today?  And, more to the point, will it remain true in five years time – you know, when our car does all our driving for us, and the same too for the bus and train?

One compromise would be to have remote piloted planes, just like the military and their remote piloted drones.  I know that historically the military have had fairly high loss rates with their remote piloted drones, but I’m not thinking about what happened several generations ago (technologically speaking) and in any event, the military engineer their products to a different set of parameters than civilian planes.  We could have planes that basically fly themselves, but in an emergency, could be remote piloted from a central facility.  It isn’t as though a pilot these days has to manually actually ‘do’ anything by hand – all they do is activate levers and controls that send electrical/electronic commands to the actual engines and control surfaces of the plane.

I’ll even concede that maybe there’s a one in a million situation where a human pilot, on the plane, could do something that either the plane’s computer systems or a remote piloting station could not do.  But, at present, without these capabilities, we’re accepting perhaps ten or one hundred in a million situations where the pilots make things worse, not better, by being present and part of the control equation.

Accepting ten errors in return for one fix is not a good way to manage ‘safety’.  Like it or not, the time has come to strip away our prejudices about automation and our dated notions about human pilots having superhuman abilities, and instead, to accept and enhance the best route to safety in the new future – a slightly scary one, where computers are better than humans.

Sure, I like the thought of seeing two smartly uniformed smiling pilots busy in the cockpit when I board my flight, but most of all, I like the feeling of safely deplaning at the end of the journey, and if that means I no longer get to see a hint of humans up front while boarding, that’s just fine by me.

Okay, a Couple of Other Things

Okay, you insist.  A couple of extra things.  Why not a China article – I continue to maintain that none of us truly comprehend the enormity of the waking giant that is China and its economy, but perhaps this article, detailing how in three short years, China used more cement in construction projects than the US did in the entire 20th century, helps to shine another light on this absolutely stunning phenomenon.

The flipside of that article is that, well, we’re no longer building much of much in the US, are we.  Here’s an article that contrasts our inability to successfully initiate and complete and more ‘think big’ projects such as defined us in the last century, even though other countries (not only China) do so regularly.

Talking about other countries, TripAdvisor just published their list of the top 25 travel destinations in the world for 2015.  As always, there are – well, let’s be polite – some surprises on the list.  The methodology for this, and all other lists, is always a bit suspect, but as an idea for future Travel Insider tours, maybe there are some pointers to be taken from it!

Talking about TripAdvisor, the problem of maliciously dishonest reviews is one which plagues review sites such as TripAdvisor and Yelp.  More to the point, it can cause huge problems to small establishments who rely on these review sites for much of their business.

Here’s an interesting story of how a Chinese restaurant in San Francisco responded to a malicious review.  Definitely a case of a first world solution to a first world problem!

Talking about first world problems and solutions, on Thursday Amazon announced a new online/cloud storage product that is compellingly appealing.  You can now get unlimited online storage, for $60/year.  Depending on how many GB of storage you might end up wanting/using, that could be hundreds of dollars less than other companies charge for the same service.

It is probable the other major online storage providers – Microsoft, Google and Apple – will grudgingly come close to matching, perhaps with ‘strings attached’ such as requiring one to be a member of some of their other services.  And, as something close to a loss-leader, they probably can afford to.  But how will companies like Dropbox respond?  It will be difficult for them.

Amazon’s product is less well integrated into most of our workflows, so I’ll wait to see how Google and Microsoft respond, but if nothing good is forthcoming, I know where I’ll be doing all my off-site backup in the future.

And Lastly This Week….

As you know, I like to put ridiculous and amusing things at the end of the newsletter, and there’s really no better place to put this item, reporting how the t-shirted nerdy-types at Apple stores – people we love dearly and appreciate when needing help with our iPhones and other Apple devices – are now going to be giving fashion advice and are being taught how to make facile insincere statements about customers and their ‘styles’, all in an attempt to up-sell overpriced Apple watches.

My favorite of the coached conversations these people are being taught is the one about how these watches, with their technology that is sure to be obsolete within a year or two, is something they expect they’ll have for years to come.  Yeah, sure, right.  But only unloved, in their bottom drawer, gathering dust.

Truly lastly this week, the mystery of the missing umbrella.  It turns out it isn’t only North Korea that likes to photoshop pictures.

Until next week, please enjoy safe travels.

David.