Awareness of fallibility is an important driver of reliability.
I met my surgeon on a gurney just before a shoulder operation, and joked with him that we had considered writing "not this one" on the good shoulder. He smiled, and then handed my then-wife a pen, saying "Write that. Things like that happen all the time." And I felt that much more confident in him, because I think people who know they can do something stupid are so much less likely to actually do them.
I don't know how widespread this is, but when I went to get surgery on my foot (NHS in the UK) they literally drew an arrow on my leg with a marker pen to avoid operating on the wrong foot haha.
I had surgery to have a tumor removed from my back earlier this year and right before the surgeon drew arrows on my back and then initialed it. "No one goes into the operating room without my initials" Made me feel more confident.
Many people in a lot of fields no longer admit their mistakes. I think a lot of the problem is that if you admit to a mistake you often become legally liable, to a not inconsiderable sum. So, the easiest thing's to pretend that the mistake never happened.
It's better for the individual who makes the mistake, for all that it's worse for everyone else.
my dad was a nurse his whole career, worked in a variety of areas including surgery. After his returement he went in for surgery and wrote things like that on his body so they didnt fuck it up
I don't think entrepreneurs or developers can look back on their mistakes with the same regret as surgeons or medical people. Watching a startup fail or a complete failure of a payment system is probably nothing compared to the opportunity cost involved when its someones life.
I barely have the emotional maturity to handle a breakup -- never mind a mistake that cost someone their life or even a finger. Certain structural engineers or medical device engineers might face such decisions but the vast majority of devs and tech people don't.
Developers are less likely to be directly responsible for the sort of things that surgeons are, unless you write mission-critical software. However, if you write any kind of software that a lot (millions) of people depend on, there is a very good chance that you've ruined people's lives or even caused death.
I dread to think how many people's relationships have been destroyed because the developers of Facebook, Whatsapp, Telegram et al decided to implement read notifications/"last seen" on their instant messaging apps.
There are some cases in England of post office managers going to jail because the Post Office software audit trail made it look like those managers were defrauding the post office. It turns out there might be severe bugs causing these errors
Suicide is complex and I don't want to ascribe a simplistic reason to suicide, but at least one post office manager died by suicide after a wrongful accusation.
A bungled software update left many people in the UK without access to their money for weeks. At least one person was imprisoned for that - they had to pay a fine by a certain date; they tried to pay but couldn't; even though this was totally outside their control the system wasn't flexible enough to cope.
British Gas allows some people to make weekly payments as part of a payment plan scheme. But you have to pay exactly that amount on exactly the agreed day or their software does not recognise the payment as part of the plan. EG you agree to pay £10 every Friday. On Wednesday you start by paying £20, then on Wednesday you pay £12, and you continue to pay £9 or £12 every Wednesday. You're always actually paid more than your plan agreement; and your actual balance is in credit; but you'll be getting aggressive "red" letters from British Gas telling you that you have not kept up to date with your plan and that your account is in arrears. British Gas, and I say this as politely as I can, are fucking scum.
Sadly, the RBS debacle led only to knighthoods, bail-outs and taxpayer money being spent on bonuses so they could "retain world-class talent". Apparently, firing their asses would have been a Bad Thing due to their stellar performance to date (and since).
The very idea of prison time for RBS re-cycling the Enron ideas of debt vehicles is laughable.
Enron people went to jail and the banks (like RBS) then had regulators permit exactly the same debt-swap bullshit that killed Enron. Banks went on to trade in debt - an eventual cause of the financial meltdown. RBS was a major culpable party but their chairman, Fred Goodwin, managed to get away with returning the knighthood and a fraction of his pension, while protected by a super-injunction making it illegal to reveal to the public he was a banker[1]. He continues to enjoy 'Royal Family' level police protection, has his properties removed from Street View [2] and is granted a level of establishment protection no-one else could dream of if they had been responsible for any catastrophe of even a fractional proportion.
Not long after he left RBS I was crossing a zebra crossing on George Street Edinburgh and I noticed a chap in a convertible Ferrari waiting for me to cross and I thought the driver looked rather familiar.
Took me a few moments to realise it was our very own "Fred the Shred" out for a quick spin....
> However, if you write any kind of software that a lot (millions) of people depend on, there is a very good chance that you've ruined people's lives or even caused death.
I've written code that computes performance bonuses for thousands of employees, and I'm confident that it's about 85% accurate. Took me a while to come to peace with that, since the workers my code is evaluating have a high ratio of bonus to base pay. But eh, I can only do my best.
I'm 100% confident that it's 85% accurate. The problem is that our sampling rate was too low. I can only make so many statistical interpolations before it starts to look like guesswork. Yes, management was aware of the issue. Luckily, the people whose bonuses were affected could not possibly have enough information to dispute my results. :-P
One of the interesting things about developing medical devices is the formal recognition of this process. During development, you must perform and document risk analysis - some of this is pretty dry, but part of it is a session of what-if where you need to look at the worst possible outcomes for you system, used properly or not. Often this is quite eye opening.
If there is an adverse event, or interaction reveals that one is possible, you'll do corrective-action/preventative-action analysis and depending on the severity you may need to inform you entire user base.
At first this sort of thing can feel process heavy and unwieldy, but I quickly came to appreciate the way in improved the actual engineering and made you think about system design in particular, useful ways.
In 2008 I left the structural engineering field. I have 0 deaths due to my actions. The decisions weigh on you, but not like medicine. Almost always your structure will survive and when it doesn't it will almost always give warning of its failure.
I suspect a lot of this would be more difficult now. I had minor surgery not long ago and (while I was still conscious) everyone in the room (including me) had to agree what was about to happen, sound off about their part in it, and confirm that they had everything they needed for their part, before continuing.
It did make me feel more comfortable about he process.
There's a fair bit of evidence that -- regardless of how they might seem at first -- these types of "checklist" procedures really do reduce error rates. Much of it is cross-pollination from aviation safety, where use of checklists has been standard procedure for a long time.
This New Yorker article talks a bit about the early days of aviation checklists, and some attempts to apply them in medicine, including a hospital which tried checklists as a method of avoiding common causes of line infections with IVs. The result:
These steps are no-brainers; they have been known and taught for years. So it seemed silly to make a checklist just for them. Still, Pronovost asked the nurses in his I.C.U. to observe the doctors for a month as they put lines into patients, and record how often they completed each step. In more than a third of patients, they skipped at least one.
The next month, he and his team persuaded the hospital administration to authorize nurses to stop doctors if they saw them skipping a step on the checklist; nurses were also to ask them each day whether any lines ought to be removed, so as not to leave them in longer than necessary. This was revolutionary. Nurses have always had their ways of nudging a doctor into doing the right thing, ranging from the gentle reminder (“Um, did you forget to put on your mask, doctor?”) to more forceful methods (I’ve had a nurse bodycheck me when she thought I hadn’t put enough drapes on a patient). But many nurses aren’t sure whether this is their place, or whether a given step is worth a confrontation. (Does it really matter whether a patient’s legs are draped for a line going into the chest?) The new rule made it clear: if doctors didn’t follow every step on the checklist, the nurses would have backup from the administration to intervene.
Pronovost and his colleagues monitored what happened for a year afterward. The results were so dramatic that they weren’t sure whether to believe them: the ten-day line-infection rate went from eleven per cent to zero. So they followed patients for fifteen more months. Only two line infections occurred during the entire period. They calculated that, in this one hospital, the checklist had prevented forty-three infections and eight deaths, and saved two million dollars in costs.
A dentist once pulled the wrong tooth. I could not really afford a crown, so I was going to have it taken out (long time ago). Thankfully it was a wisdom tooth (I have room for them).
The look on his face after the procedure was one of shock. He gave me a free root canal and crown for the original bad tooth.
Here is another attempt to try and get health care systems to learn how to deal with fallibility. In this case a woman died unnecessarily during a minor operation, however her husband was an airline pilot and he is now trying to teach the UK National Health Service the lessons that airlines learned long ago: http://www.newstatesman.com/2014/05/how-mistakes-can-save-li...
Thank you for posting this...surgery is one of the few professions that I think about, and then think, "Nope, no way, not even in the wildest dreams in an alternative universe, that I could ever be a surgeon"...but I get the sense that so much pride is tied up into the skill of a surgeon, such that mechanical ability and intuition are almost inextricable. I have no problem with the assertion that a human expert is needed to make decisions as the cuts are being made. But it's astonishing to me that so much of the success of surgery relies on a near-superhuman ability of muscle control and focus, an ability that surely declines with age, even as the surgeon's wisdom and experience grows. I would love to read more about "autopilot"-type advances in medicine (though too much of the coverage focuses on, "Will robots replace your doctor?" rather than man-and-machine working together)
I liked where he had to track down one patient because he had left the keys to the summer condo inside after they fell in during the operation!
Makes you think...
My name is not Richard, or any abbreviations of it!
But yes, I do actually know people who have have to have subsequent ops, where material was found from an earlier op. Not as uncommon as you would think, sadly.
I met my surgeon on a gurney just before a shoulder operation, and joked with him that we had considered writing "not this one" on the good shoulder. He smiled, and then handed my then-wife a pen, saying "Write that. Things like that happen all the time." And I felt that much more confident in him, because I think people who know they can do something stupid are so much less likely to actually do them.