Interesting, and well written, and presumably breaking my record for the longest ever thread header. I enjoyed it and see the point, but I must take issue with one comment:
The DfE became the department that knows next to nothing about education.
That is of course bullshit.
The words 'next to' have somehow inserted themselves to suggest, wrongly, that the DfE know something about education.
The owner of the dogs that the lady in Jaywick yesterday made a rap (grime?) song in tribute to his ex when she died. At the end it might be the lady who died yesterday with him in the church
Not particularly my cup of tea musically, but here it is
"Well, if we reduce the amount of paperwork social workers have to do by a quarter, we might *double* the time they spend actually doing social work"
I am, unfortunately, a cynic.
Something that is far easier and produces a great soundbite is "we have reduced the number of administrators by a quarter!"
Always popular. Get rid of the beancounters and paper-shufflers. Easy savings. Tea and medals and the adulation of an adoring public.
Unfortunately the amount of administration is either unchanged or increased. So instead of admin assistants, the social workers, surgeons, teachers, and so on end up having to do it and we end up with them spending even less time at their primary job (and taking longer to do the paperwork than a specialist and doing it worse).
We see this, call for change, and someone reacts by reducing the number of remaining administrators by a quarter...
Actually addressing the process as you call for would be the solution. But take longer and require investing time and resources to analyse the processes and then pay to have them changed. Far easier to simply fire a few existing administrators and load up the social workers and doctors with a bit more of the paperwork. After all - who sees that bit beyond the front-page savings in reduced admin staff?
Process is a good thing. It gives you consistency, efficiency hopefully, transparency if you do it right. It also normally measurable so you can learn from what works and what doesn't and get an improvement that is otherwise difficult to achieve.
You do need to be the master of the process and not its slave however.
This is a good header because it is a counterpoint to the many articles criticising situations where due process isn't followed and the consequential problems, failings, injustices, whatever.
When there is a pressing problem process just gets set aside. Russia built the Kerch bridge in 2 years, 18 miles long, the biggest bridge in Europe, and it still stands despite the various attempts to blow it up in the current war. We cold do with adopting a similar attitude to things like climate change and housing.
The owner of the dogs that the lady in Jaywick yesterday made a rap (grime?) song in tribute to his ex when she died. At the end it might be the lady who died yesterday with him in the church
Not particularly my cup of tea musically, but here it is
That poor little boy in the video is presumably his son. Who lost his mother 2 years ago. And now, because his father is an inadequate dickless moron who wants big dogs to make him feel tough, has now witnessed his grandmother being mauled to death
Nice header - agree that processes often substitute for thinking. Not easy to solve though. As society gets more complex and things interact more, it naturally gets harder to do things. Some countries seem to manage it a bit better though.
Where UK does better IMO is online... many many things can be done online much more easily than when I have to interact with other govts.
Nice header - agree that processes often substitute for thinking. Not easy to solve though. As society gets more complex and things interact more, it naturally gets harder to do things. Some countries seem to manage it a bit better though.
Where UK does better IMO is online... many many things can be done online much more easily than when I have to interact with other govts.
Yes, gov.Uk has come on hugely over recent years, both in terms of the clarity and user-friendliness of the guidance, to all the various things you can arrange entirely online. I know a lot of pet travellers use it for advice on pet travel, as the US equivalents are clunky and often out of date. The key to a useful website is to ensure it’s always up to date, and they must have a fairly big website team to achieve that. And the integration between different databases - for example the car tax stuff, that pulls through from the MOT database and the car insurance one, is impressive.
Doubtless some Germans are blaming Olaf Scholz for their local market struggling or because a single minimum wage doesn't provide affluent living for a family of six.
And similar pieces could be done about all developed countries.
But moaning about things or blaming governments for not providing everything demanded isn't going to achieve anything anywhere.
The way to improvement is through work not whining, through increasing productivity not increasing demands, through taking the opportunities on offer not expecting others to provide more.
Interesting, and well written, and presumably breaking my record for the longest ever thread header...
Nope, Malmesbury's opus is 1680-ish words. The "Ydoethur Limit" is still 1,800 words. The "OGH Limit" was 0-800 words, which gave way to the "Cyclefree Limit" of 800-1,200 words, which in turn gave way to the "Ydoethur Limit" of 1,200-1,800 words.
Think of them like electron shells. Each layer expands but it becomes progressively more difficult to jump to the next one. This third "Ydoethur" phase in PB Cosmic Evolution is analogous to the d-block: the articles are rusty, metallic, and go "clang" when you strike them. The articles will be bigger on the inside than on the PB outside, which is only logical when you consider the name.
Good piece. My bugbear is "targets" rather than "process" - they are toxic - but there's no doubt process can become counterproductive and if left to grow unchecked a monster. Where it comes into its own (imo) is where you're trying to master something as an individual. There it's advisable to forget about the desired outcome of getting better and concentrate purely on the "process" of getting better. So, sport, music, cooking, these type of areas.
A related issue to all this is "measurement". I think we could usefully do a bit less of this. Eg you should not normally spend more time/money measuring how big of a problem something is than you do in resolving it. Yet we often do. Also the very act of measuring something can change it and so you are likely to get a false reading.
This is a good header because it is a counterpoint to the many articles criticising situations where due process isn't followed and the consequential problems, failings, injustices, whatever.
When there is a pressing problem process just gets set aside. Russia built the Kerch bridge in 2 years, 18 miles long, the biggest bridge in Europe, and it still stands despite the various attempts to blow it up in the current war. We cold do with adopting a similar attitude to things like climate change and housing.
To put it rather horribly and simply, we need to do less with less. Absent a war or a really nasty f*** in charge, process takes over. I looked as the Ajax reports as the delay mounted, and I swear to goodness I would have hired the Russians to kill them. They are very good at doing things. Just not the necessary things.
Good piece. My bugbear is "targets" rather than "process" - they are toxic - but there's no doubt process can become counterproductive and if left to grow unchecked a monster. Where it comes into its own (imo) is where you're trying to master something as an individual. There it's advisable to forget about the desired outcome of getting better and concentrate purely on the "process" of getting better. So, sport, music, cooking, these type of areas.
A related issue to all this is "measurement". I think we could usefully do a bit less of this. Eg you should not normally spend more time/money measuring how big of a problem something is than you do in resolving it. Yet we often do. Also the very act of measuring something can change it and so you are likely to get a false reading.
Ahem. Given that my job involves measuring things, may I put the contrary view? To do it correctly, you fit the metric to the task. If you are spending too much time measuring, the cure is to measure something else, not "no measurement". Classical statistics was created to measure a big thing quickly using a small thing. That's (one of the reasons why) I think Big Data is not a universal panacea.
Nice header - agree that processes often substitute for thinking. Not easy to solve though. As society gets more complex and things interact more, it naturally gets harder to do things. Some countries seem to manage it a bit better though.
Where UK does better IMO is online... many many things can be done online much more easily than when I have to interact with other govts.
Britain does online very well but mostly in a way that enriches American firms because heaven forefend we look after our own companies like, say, America.
One small quibble on the 737 issue. The process looks to be sensible and appropriate. The problem is it wasn't followed.
The examples I know something about from your list - Boeing, Horizon etc - the root cause is poor culture in those organisations. In general organisations with good culture have effective processes, but not necessarily fewer of them. I can think of processes the Post Office would have benefitted from, including a proper escalation process for people raising concerns.
Change becomes increasingly difficult as organisations grow larger and older.
Yes the issue is almost always having change-minded management. Many large companies, and the public sector, are terrible at this.
Yes and no.
Sometimes change for change's sake becomes a fetish. Perhaps tinkering would be a better word, but having Changed Something is pretty much a prerequisite for promotion in schools. Either that, or being the only full time permanent body left standing. And a lot of those changes may individually be desirable- adding something nice or closing off the pathway to something bad happening. But the overall effect is something like adding another bit of luggage to the donkey in Buckaroo, it's fine, it's fine and it stays that way until the donkey throws everything off.
As to the answer to the problem @Malmesbury describes so well... I'm sure that milages vary. But I suspect the issue is that remote micromanagement is a possibility now in a way that it wasn't before. Partly because of electronic documentation of process, but also because of the speed of electronic communication. It's probably still a bad idea, because an organisation, let alone a nation, shouldn't fit inside one brain, however enormous.
It's an attractive idea that everything can be run from one centre issuing commands. It appeals to fairness (no postcode lottery) and apparent efficiency (fewer politicians) and vanity (I can command everything). It's just that, even if the technology now exists (unlike 1970s Chile or 1960s Soviet Union), life isn't like that.
Good piece. My bugbear is "targets" rather than "process" - they are toxic - but there's no doubt process can become counterproductive and if left to grow unchecked a monster. Where it comes into its own (imo) is where you're trying to master something as an individual. There it's advisable to forget about the desired outcome of getting better and concentrate purely on the "process" of getting better. So, sport, music, cooking, these type of areas.
A related issue to all this is "measurement". I think we could usefully do a bit less of this. Eg you should not normally spend more time/money measuring how big of a problem something is than you do in resolving it. Yet we often do. Also the very act of measuring something can change it and so you are likely to get a false reading.
Ahem. Given that my job involves measuring things, may I put the contrary view? To do it correctly, you fit the metric to the task. If you are spending too much time measuring, the cure is to measure something else, not "no measurement". Classical statistics was created to measure a big thing quickly using a small thing. That's (one of the reasons why) I think Big Data is not a universal panacea.
Thing is, if you move from trying to measure everyone to doing a sample to get reasonably robust statistics, you lose the ability to crown winners and punish losers. (Even if both the winners and losers aren't statistically robust).
And for many, the crowning and punishing are the point of the exercise, even if the real information is what's happening around the average.
One small quibble on the 737 issue. The process looks to be sensible and appropriate. The problem is it wasn't followed.
The examples I know something about from your list - Boeing, Horizon etc - the root cause is poor culture in those organisations. In general organisations with good culture have effective processes, but not necessarily fewer of them. I can think of processes the Post Office would have benefitted from, including a proper escalation process for people raising concerns.
From my understanding (which may be wrong), the process was flawed. Boeing allowed employees from subcontractor Spirit to fix snags. Which is fine. But the meaning of certain terms, such as door opened, was different between the two companies, and on their tracking systems. Hence when Spirit 'opened' (removed) a door, Boeing people further down the line did not realise what extra work needed to be done.
The devil will be in the details, but an obvious question that comes int my mind is why no-one further down the production line realised the flaw. Is it because they are so tuned to the process that they're not *meant* to notice issues? A human version of 'the computer says no'...
One small quibble on the 737 issue. The process looks to be sensible and appropriate. The problem is it wasn't followed.
The examples I know something about from your list - Boeing, Horizon etc - the root cause is poor culture in those organisations. In general organisations with good culture have effective processes, but not necessarily fewer of them. I can think of processes the Post Office would have benefitted from, including a proper escalation process for people raising concerns.
From my understanding (which may be wrong), the process was flawed. Boeing allowed employees from subcontractor Spirit to fix snags. Which is fine. But the meaning of certain terms, such as door opened, was different between the two companies, and on their tracking systems. Hence when Spirit 'opened' (removed) a door, Boeing people further down the line did not realise what extra work needed to be done.
The devil will be in the details, but an obvious question that comes int my mind is why no-one further down the production line realised the flaw. Is it because they are so tuned to the process that they're not *meant* to notice issues? A human version of 'the computer says no'...
Yes, there’s no job card required for opening or closing a door - but what about when the door isn’t really a door? That’s how a plane left the factory with a hole in the side secured by no bolts.
Good piece. My bugbear is "targets" rather than "process" - they are toxic - but there's no doubt process can become counterproductive and if left to grow unchecked a monster. Where it comes into its own (imo) is where you're trying to master something as an individual. There it's advisable to forget about the desired outcome of getting better and concentrate purely on the "process" of getting better. So, sport, music, cooking, these type of areas.
A related issue to all this is "measurement". I think we could usefully do a bit less of this. Eg you should not normally spend more time/money measuring how big of a problem something is than you do in resolving it. Yet we often do. Also the very act of measuring something can change it and so you are likely to get a false reading.
Ahem. Given that my job involves measuring things, may I put the contrary view? To do it correctly, you fit the metric to the task. If you are spending too much time measuring, the cure is to measure something else, not "no measurement". Classical statistics was created to measure a big thing quickly using a small thing. That's (one of the reasons why) I think Big Data is not a universal panacea.
99 times out of a 100 Big Data is collecting stuff you will never need for the sake of it...
I made a few changes to PB in the last 24 hours to solve a few issues people were having:
(1) Stopped using external caching, because it was throwing errors. (I'll reimplement something closer to election night, but for now page load times are sufficiently low that it's fine.)
(2) Implemented proper forwarding for people who went to http rather than https
(3) Solved the bug where sometimes people would see the default Apache2 page rather than pb
If there are any other issues, please let me know. My email address is simply my username at gmail.com
I made a few changes to PB in the last 24 hours to solve a few issues people were having:
(1) Stopped using external caching, because it was throwing errors. (I'll reimplement something closer to election night, but for now page load times are sufficiently low that it's fine.)
(2) Implemented proper forwarding for people who went to http rather than https
(3) Solved the bug where sometimes people would see the default Apache2 page rather than pb
If there are any other issues, please let me know. My email address is simply my username at gmail.com
Hi Robert,
You’ve forgotten to put up the donation link for the new server fund.
One small quibble on the 737 issue. The process looks to be sensible and appropriate. The problem is it wasn't followed.
The examples I know something about from your list - Boeing, Horizon etc - the root cause is poor culture in those organisations. In general organisations with good culture have effective processes, but not necessarily fewer of them. I can think of processes the Post Office would have benefitted from, including a proper escalation process for people raising concerns.
From my understanding (which may be wrong), the process was flawed. Boeing allowed employees from subcontractor Spirit to fix snags. Which is fine. But the meaning of certain terms, such as door opened, was different between the two companies, and on their tracking systems. Hence when Spirit 'opened' (removed) a door, Boeing people further down the line did not realise what extra work needed to be done.
The devil will be in the details, but an obvious question that comes int my mind is why no-one further down the production line realised the flaw. Is it because they are so tuned to the process that they're not *meant* to notice issues? A human version of 'the computer says no'...
On my understanding, Boeing staff required its contractor Spirit to make a QA fix that necessitated the removal of the door plug, but they didn't record the removal in Boeing's system of record as they should have done under the policy. This meant reinstatement of the door wasn't on anyone's job queue, once the fix was made. Basic stuff, if true.
I do have a related angle on this, though. How did we end up with so many rules and regulations?
Two answers are stupidity and malice. You have a loose set of rules and regulations, but someone does not do step 1 properly and causes a disaster - a fallen bridge, or a child being abused. Either because they made a mistake, or because they could not be bothered to do their job properly, perhaps because they gained from it.
There are many possible 'fixes' for this. Two are: 1) Better training and employees. 2) Add another later of process in, so the results of step 1 are checked.
Either. or both, of these can work. Or several others, depending on the application.
Let's look at something like railway signalling. A new signal can cost hundreds of thousands of pounds to install. That sounds stupid. But if you look at the reasons for that cost, then you can see incidents in the past that have led to the rules and regulations that are now in place, and lead to the cost. And those incidents often cost lives. Do we want to save money and put more risk onto rail workers' (or passengers)?
I don't know what the answer is. In the case of the £250k bridge in Stiffkey, I'd guess it's because the NT don't want it and have elevated costs as much as possible. Or perhaps not.
The owner of the dogs that the lady in Jaywick yesterday made a rap (grime?) song in tribute to his ex when she died. At the end it might be the lady who died yesterday with him in the church
Not particularly my cup of tea musically, but here it is
Not my taste either, but as far as I can tell it’s actually a decent effort. And fascinating to see a grime video major on the architectural details of an Anglican church.
Yes, that does appear to be the grandmother at the end.
Horrifying little vignette of English life. I wonder how the mother died.
Good piece. My bugbear is "targets" rather than "process" - they are toxic - but there's no doubt process can become counterproductive and if left to grow unchecked a monster. Where it comes into its own (imo) is where you're trying to master something as an individual. There it's advisable to forget about the desired outcome of getting better and concentrate purely on the "process" of getting better. So, sport, music, cooking, these type of areas.
A related issue to all this is "measurement". I think we could usefully do a bit less of this. Eg you should not normally spend more time/money measuring how big of a problem something is than you do in resolving it. Yet we often do. Also the very act of measuring something can change it and so you are likely to get a false reading.
Ahem. Given that my job involves measuring things, may I put the contrary view? To do it correctly, you fit the metric to the task. If you are spending too much time measuring, the cure is to measure something else, not "no measurement". Classical statistics was created to measure a big thing quickly using a small thing. That's (one of the reasons why) I think Big Data is not a universal panacea.
Yes of course. It all depends. But an example of what I mean from my City days - involving traders and accountants both of which I was at various times.
So you'd have traders and every day at close of business each one would be presented with their "Daily P/L". This told them how much they had made or lost that day on their activities.
It typically took a whole bunch of qualified accountants (overpaid compared to everybody other than the traders) to produce this thing and they'd use a system linked to but separate from the main ledger. At month end the two systems never agreed so there'd be another team of professionals tasked with reconciling them. What a palaver.
And for what? Why did a trader have to know their profit or loss before they knocked off? They didn't. It was a nice to have at best. It's risk you need to monitor constantly not how much money you're making. So why did a whole operation-within-the-operation, with serious headcount and IT spend, exist to provide this info? Nobody could ever tell me.
One small quibble on the 737 issue. The process looks to be sensible and appropriate. The problem is it wasn't followed.
The examples I know something about from your list - Boeing, Horizon etc - the root cause is poor culture in those organisations. In general organisations with good culture have effective processes, but not necessarily fewer of them. I can think of processes the Post Office would have benefitted from, including a proper escalation process for people raising concerns.
From my understanding (which may be wrong), the process was flawed. Boeing allowed employees from subcontractor Spirit to fix snags. Which is fine. But the meaning of certain terms, such as door opened, was different between the two companies, and on their tracking systems. Hence when Spirit 'opened' (removed) a door, Boeing people further down the line did not realise what extra work needed to be done.
The devil will be in the details, but an obvious question that comes int my mind is why no-one further down the production line realised the flaw. Is it because they are so tuned to the process that they're not *meant* to notice issues? A human version of 'the computer says no'...
On my understanding, Boeing staff required its contractor Spirit to make a QA fix that necessitated the removal of the door plug, but they didn't record the removal in Boeing's system of record as they should have done under the policy. This meant reinstatement of the door wasn't on anyone's job queue, once the fix was made. Basic stuff, if true.
From what I read, the Spirit employees recorded the change on both their own system and Boeing's, but the meaning of 'door opened' varied on both. In the case of Spirit, it meant the door had been removed (as seems sensible for doors that may not have closing mechanisms. For Boeing, it just meant the door had been opened.
Good piece. My bugbear is "targets" rather than "process" - they are toxic - but there's no doubt process can become counterproductive and if left to grow unchecked a monster. Where it comes into its own (imo) is where you're trying to master something as an individual. There it's advisable to forget about the desired outcome of getting better and concentrate purely on the "process" of getting better. So, sport, music, cooking, these type of areas.
A related issue to all this is "measurement". I think we could usefully do a bit less of this. Eg you should not normally spend more time/money measuring how big of a problem something is than you do in resolving it. Yet we often do. Also the very act of measuring something can change it and so you are likely to get a false reading.
Ahem. Given that my job involves measuring things, may I put the contrary view? To do it correctly, you fit the metric to the task. If you are spending too much time measuring, the cure is to measure something else, not "no measurement". Classical statistics was created to measure a big thing quickly using a small thing. That's (one of the reasons why) I think Big Data is not a universal panacea.
Yes of course. It all depends. But an example of what I mean from my City days - involving traders and accountants both of which I was at various times.
So you'd have traders and every day at close of business each one would be presented with their "Daily P/L". This told them how much they had made or lost that day on their activities.
It typically took a whole bunch of qualified accountants (overpaid compared to everybody other than the traders) to produce this thing and they'd use a system linked to but separate from the main ledger. At month end the two systems never agreed so there'd be another team of professionals tasked with reconciling them. What a palaver.
And for what? Why did a trader have to know their profit or loss before they knocked off? They didn't. It was a nice to have at best. It's risk you need to monitor constantly not how much money you're making. So why did a whole operation-within-the-operation, with serious headcount and IT spend, exist to provide this info? Nobody could ever tell me.
Not my field, but I refer the Honourable Gentlemen to my earlier answer;
you lose the ability to crown winners and punish losers. (Even if both the winners and losers aren't statistically robust).
And for many, the crowning and punishing are the point of the exercise, even if the real information is what's happening around the average.
(I think it's a personal advantage to me that that part of my psyche remains massively underdeveloped, certainly compared with the average City trader. Society needs herbivores and carnivores, and being a carnivore just looks so damn stressful.)
You need to know what each piece of process is gaining you, but some parts are necessary.
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
One small quibble on the 737 issue. The process looks to be sensible and appropriate. The problem is it wasn't followed.
The examples I know something about from your list - Boeing, Horizon etc - the root cause is poor culture in those organisations. In general organisations with good culture have effective processes, but not necessarily fewer of them. I can think of processes the Post Office would have benefitted from, including a proper escalation process for people raising concerns.
From my understanding (which may be wrong), the process was flawed. Boeing allowed employees from subcontractor Spirit to fix snags. Which is fine. But the meaning of certain terms, such as door opened, was different between the two companies, and on their tracking systems. Hence when Spirit 'opened' (removed) a door, Boeing people further down the line did not realise what extra work needed to be done.
The devil will be in the details, but an obvious question that comes int my mind is why no-one further down the production line realised the flaw. Is it because they are so tuned to the process that they're not *meant* to notice issues? A human version of 'the computer says no'...
On my understanding, Boeing staff required its contractor Spirit to make a QA fix that necessitated the removal of the door plug, but they didn't record the removal in Boeing's system of record as they should have done under the policy. This meant reinstatement of the door wasn't on anyone's job queue, once the fix was made. Basic stuff, if true.
From what I read, the Spirit employees recorded the change on both their own system and Boeing's, but the meaning of 'door opened' varied on both. In the case of Spirit, it meant the door had been removed (as seems sensible for doors that may not have closing mechanisms. For Boeing, it just meant the door had been opened.
We'll know more when the reports come out.
I should have added: both these definitions may make sense. Neither Spirit or Boeing may be wrong; the problem was the interface: the different definitions between their processes. But my question is (*) why no-one further down the process noticed the issue - which seems to have occurred on several planes. Do the processes Boeing lineworkers work under prevent them from noticing and/or reporting such issues (they are trained to only follow the Holy Process), or do they have a Japanese-style anyone-can-stop-the-production-line system?
(*) If my understanding of what happened is correct...
You need to know what each piece of process is gaining you, but some parts are necessary.
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
That’s why attempts for planes to have only one pilot are doomed to fail. Almost everything a pilot does, is cross-checked with the other pilot, and these cross checks pick up a whole load of human errors. The chances of two humans making a mistake that could kill them, is orders of magnitude more than a single human making the same decision.
One small quibble on the 737 issue. The process looks to be sensible and appropriate. The problem is it wasn't followed.
The examples I know something about from your list - Boeing, Horizon etc - the root cause is poor culture in those organisations. In general organisations with good culture have effective processes, but not necessarily fewer of them. I can think of processes the Post Office would have benefitted from, including a proper escalation process for people raising concerns.
From my understanding (which may be wrong), the process was flawed. Boeing allowed employees from subcontractor Spirit to fix snags. Which is fine. But the meaning of certain terms, such as door opened, was different between the two companies, and on their tracking systems. Hence when Spirit 'opened' (removed) a door, Boeing people further down the line did not realise what extra work needed to be done.
The devil will be in the details, but an obvious question that comes int my mind is why no-one further down the production line realised the flaw. Is it because they are so tuned to the process that they're not *meant* to notice issues? A human version of 'the computer says no'...
They were trying to avoid reporting problem officially. Hence the second process/system.
They were also trying to use delivery snagging as a substitute for QA. Instead of building planes right, just chuck them together. And believe that the pre delivery checks would find the problems.
So you had process interference multiplied by process abuse. Plus straight up incompetence.
Or, as Nicola put it: "Former senior civil servants, men and women trained in the art of diplomacy, recall with a shudder her foul-mouthed phone calls. “I don’t f**ing care about your f**ing process, I just want it done,” was typical of how Sturgeon conducted government business, according to one official I spoke to recently."
You need to know what each piece of process is gaining you, but some parts are necessary.
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
That’s why attempts for planes to have only one pilot are doomed to fail. Almost everything a pilot does, is cross-checked with the other pilot, and these cross checks pick up a whole load of human errors. The chances of two humans making a mistake that could kill them, is orders of magnitude more than a single human making the same decision.
One small quibble on the 737 issue. The process looks to be sensible and appropriate. The problem is it wasn't followed.
The examples I know something about from your list - Boeing, Horizon etc - the root cause is poor culture in those organisations. In general organisations with good culture have effective processes, but not necessarily fewer of them. I can think of processes the Post Office would have benefitted from, including a proper escalation process for people raising concerns.
From my understanding (which may be wrong), the process was flawed. Boeing allowed employees from subcontractor Spirit to fix snags. Which is fine. But the meaning of certain terms, such as door opened, was different between the two companies, and on their tracking systems. Hence when Spirit 'opened' (removed) a door, Boeing people further down the line did not realise what extra work needed to be done.
The devil will be in the details, but an obvious question that comes int my mind is why no-one further down the production line realised the flaw. Is it because they are so tuned to the process that they're not *meant* to notice issues? A human version of 'the computer says no'...
They were trying to avoid reporting problem officially. Hence the second process/system.
They were also trying to use delivery snagging as a substitute for QA. Instead of building planes right, just chuck them together. And believe that the pre delivery checks would find the problems.
So you had process interference multiplied by process abuse. Plus straight up incompetence.
On your first line: that's not what I understood. The second process was because Spirit people were working on Boeing's production line. Hence they needed to mark what they were doing on both the Spirit systems and the Boeing ones.
Potentially disagree with your second line. Minor problems will always occur, and I bet Airbus has people from (say) Broughton fixing minor snags on the main production lines in France. It's probably better than having people trying to fix problems Spirit created, for a variety of reasons.
You need to know what each piece of process is gaining you, but some parts are necessary.
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
That’s why attempts for planes to have only one pilot are doomed to fail. Almost everything a pilot does, is cross-checked with the other pilot, and these cross checks pick up a whole load of human errors. The chances of two humans making a mistake that could kill them, is orders of magnitude more than a single human making the same decision.
Assuming the culture allows a junior second pilot to question their senior captain.
You need to know what each piece of process is gaining you, but some parts are necessary.
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
That’s why attempts for planes to have only one pilot are doomed to fail. Almost everything a pilot does, is cross-checked with the other pilot, and these cross checks pick up a whole load of human errors. The chances of two humans making a mistake that could kill them, is orders of magnitude more than a single human making the same decision.
Assuming the culture allows a junior second pilot to question their senior captain.
A culture that’s been assiduously enforced, at least at Western airlines, since the Teneriffe crash of 1977. Yes, a junior pilot can call out a senior Captain.
You need to know what each piece of process is gaining you, but some parts are necessary.
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
That’s why attempts for planes to have only one pilot are doomed to fail. Almost everything a pilot does, is cross-checked with the other pilot, and these cross checks pick up a whole load of human errors. The chances of two humans making a mistake that could kill them, is orders of magnitude more than a single human making the same decision.
Assuming the culture allows a junior second pilot to question their senior captain.
AIUI that's been an acknowledged issue for decades; including the Staines crash. It's why Crew Resource Management has become such a thing. Much more recently, ISTR a Korean crash was down to similar CRM issues.
"An hour and a half before the departure of Flight 548, its rostered captain, Stanley Key, a former Royal Air Force pilot who had served during the Second World War, was involved in a quarrel in the crew room at Heathrow's Queens Building with a first officer named Flavell. The subject was the threatened strike, which Flavell supported and Key opposed. Both of Key's flight deck crew on Flight 548 witnessed the altercation, and another bystander described Key's outburst as "the most violent argument he had ever heard".[6] Shortly afterward Key apologised to Flavell, and the matter seemed closed."
You need to know what each piece of process is gaining you, but some parts are necessary.
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
That’s why attempts for planes to have only one pilot are doomed to fail. Almost everything a pilot does, is cross-checked with the other pilot, and these cross checks pick up a whole load of human errors. The chances of two humans making a mistake that could kill them, is orders of magnitude more than a single human making the same decision.
Assuming the culture allows a junior second pilot to question their senior captain.
A culture that’s been assiduously enforced, at least at Western airlines, since the Teneriffe crash of 1977. Yes, a junior pilot can call out a senior Captain.
Or, as Nicola put it: "Former senior civil servants, men and women trained in the art of diplomacy, recall with a shudder her foul-mouthed phone calls. “I don’t f**ing care about your f**ing process, I just want it done,” was typical of how Sturgeon conducted government business, according to one official I spoke to recently."
One small quibble on the 737 issue. The process looks to be sensible and appropriate. The problem is it wasn't followed.
The examples I know something about from your list - Boeing, Horizon etc - the root cause is poor culture in those organisations. In general organisations with good culture have effective processes, but not necessarily fewer of them. I can think of processes the Post Office would have benefitted from, including a proper escalation process for people raising concerns.
From my understanding (which may be wrong), the process was flawed. Boeing allowed employees from subcontractor Spirit to fix snags. Which is fine. But the meaning of certain terms, such as door opened, was different between the two companies, and on their tracking systems. Hence when Spirit 'opened' (removed) a door, Boeing people further down the line did not realise what extra work needed to be done.
The devil will be in the details, but an obvious question that comes int my mind is why no-one further down the production line realised the flaw. Is it because they are so tuned to the process that they're not *meant* to notice issues? A human version of 'the computer says no'...
They were trying to avoid reporting problem officially. Hence the second process/system.
They were also trying to use delivery snagging as a substitute for QA. Instead of building planes right, just chuck them together. And believe that the pre delivery checks would find the problems.
So you had process interference multiplied by process abuse. Plus straight up incompetence.
On your first line: that's not what I understood. The second process was because Spirit people were working on Boeing's production line. Hence they needed to mark what they were doing on both the Spirit systems and the Boeing ones.
Potentially disagree with your second line. Minor problems will always occur, and I bet Airbus has people from (say) Broughton fixing minor snags on the main production lines in France. It's probably better than having people trying to fix problems Spirit created, for a variety of reasons.
Read the account of various whistleblowers.
Boeing is (and was) delivering broken planes and trying to fix them pre delivery. That’s process abuse. They shouldn’t be fixing that many issues that late in the game.
Spirit is a bullshit way of putting a chunk of Boeing at arms length. So the bean counters only need to count beans. It’s really a part of Boeing - just semi-disconnected.
I made a few changes to PB in the last 24 hours to solve a few issues people were having:
(1) Stopped using external caching, because it was throwing errors. (I'll reimplement something closer to election night, but for now page load times are sufficiently low that it's fine.)
(2) Implemented proper forwarding for people who went to http rather than https
(3) Solved the bug where sometimes people would see the default Apache2 page rather than pb
If there are any other issues, please let me know. My email address is simply my username at gmail.com
Could you have a go at the BBC website next? Recently, every time i go on there each page is far too small to read so that they can stick the maximum number of headlines on it. Incredibly irritating.
Or, as Nicola put it: "Former senior civil servants, men and women trained in the art of diplomacy, recall with a shudder her foul-mouthed phone calls. “I don’t f**ing care about your f**ing process, I just want it done,” was typical of how Sturgeon conducted government business, according to one official I spoke to recently."
Or, as Nicola put it: "Former senior civil servants, men and women trained in the art of diplomacy, recall with a shudder her foul-mouthed phone calls. “I don’t f**ing care about your f**ing process, I just want it done,” was typical of how Sturgeon conducted government business, according to one official I spoke to recently."
You need to know what each piece of process is gaining you, but some parts are necessary.
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
That’s why attempts for planes to have only one pilot are doomed to fail. Almost everything a pilot does, is cross-checked with the other pilot, and these cross checks pick up a whole load of human errors. The chances of two humans making a mistake that could kill them, is orders of magnitude more than a single human making the same decision.
Assuming the culture allows a junior second pilot to question their senior captain.
A culture that’s been assiduously enforced, at least at Western airlines, since the Teneriffe crash of 1977. Yes, a junior pilot can call out a senior Captain.
Or, as Nicola put it: "Former senior civil servants, men and women trained in the art of diplomacy, recall with a shudder her foul-mouthed phone calls. “I don’t f**ing care about your f**ing process, I just want it done,” was typical of how Sturgeon conducted government business, according to one official I spoke to recently."
You need to know what each piece of process is gaining you, but some parts are necessary.
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
That’s why attempts for planes to have only one pilot are doomed to fail. Almost everything a pilot does, is cross-checked with the other pilot, and these cross checks pick up a whole load of human errors. The chances of two humans making a mistake that could kill them, is orders of magnitude more than a single human making the same decision.
Assuming the culture allows a junior second pilot to question their senior captain.
A culture that’s been assiduously enforced, at least at Western airlines, since the Teneriffe crash of 1977. Yes, a junior pilot can call out a senior Captain.
All well documented and it’s hoped ‘lessons have been learned’! I think Korean Airlines now require all cockpit conversations to be in English.
I’m trying to find fascinating report I found (from the NTSB) that argued that using a second language had a serious cultural effect - so speaking English would shift mindsets.
Or, as Nicola put it: "Former senior civil servants, men and women trained in the art of diplomacy, recall with a shudder her foul-mouthed phone calls. “I don’t f**ing care about your f**ing process, I just want it done,” was typical of how Sturgeon conducted government business, according to one official I spoke to recently."
Not just Nicola. Remember the incessant rants from Dominic Cummings about process slowing change. Or Liz Truss sidelining civil servants in department after department if they were not to her liking. Or Tony Blair's dozens of SpAds and sofa government.
Right, so just abolish process and trust a strong leader unconstrained by rules who will Get Things Done.
Good to see you didn’t read to the end.
Congratulations, you win the prize for what I expected.
I read to the end explicitly to see if there were any examples of what processes to follow instead. There were a few dozen bad processes, and no alternative provided, of course. Just a demand for more human discretion - literally, "just trust the experts".
Or, as Nicola put it: "Former senior civil servants, men and women trained in the art of diplomacy, recall with a shudder her foul-mouthed phone calls. “I don’t f**ing care about your f**ing process, I just want it done,” was typical of how Sturgeon conducted government business, according to one official I spoke to recently."
Personally I think it's easy to wander around a society powered by internet/electric/financial networks that work 99.99% of the time and only spot the bad cases.
Completely agree with Malmesbury about the process problem. An issue about why it is hard to resolve.
The world is full of process designers. There have to be some. There have to be rules and protocols and standards. And the world, on the whole, divides into doers and admin people, including process designers. So there are people who collectively build and maintain dwellings (eg Grenfell), and people who are part of ancillary processes.
Each group wishes to avoid blame; they are human.
The process designers can only avoid blame by covering all bases, even though they know their millions of pages will be mostly unread, ignored or not recalled.
The actual doers can't possibly cover all the bases, because it is impossible and impractical to know it all. So they hide behind blaming each other, apologising early and often, ceasing trading, and the force of inertia.
You need to know what each piece of process is gaining you, but some parts are necessary.
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
That’s why attempts for planes to have only one pilot are doomed to fail. Almost everything a pilot does, is cross-checked with the other pilot, and these cross checks pick up a whole load of human errors. The chances of two humans making a mistake that could kill them, is orders of magnitude more than a single human making the same decision.
Assuming the culture allows a junior second pilot to question their senior captain.
A culture that’s been assiduously enforced, at least at Western airlines, since the Teneriffe crash of 1977. Yes, a junior pilot can call out a senior Captain.
All well documented and it’s hoped ‘lessons have been learned’! I think Korean Airlines now require all cockpit conversations to be in English.
I’m trying to find fascinating report I found (from the NTSB) that argued that using a second language had a serious cultural effect - so speaking English would shift mindsets.
You need to know what each piece of process is gaining you, but some parts are necessary.
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
That’s why attempts for planes to have only one pilot are doomed to fail. Almost everything a pilot does, is cross-checked with the other pilot, and these cross checks pick up a whole load of human errors. The chances of two humans making a mistake that could kill them, is orders of magnitude more than a single human making the same decision.
Assuming the culture allows a junior second pilot to question their senior captain.
A culture that’s been assiduously enforced, at least at Western airlines, since the Teneriffe crash of 1977. Yes, a junior pilot can call out a senior Captain.
All well documented and it’s hoped ‘lessons have been learned’! I think Korean Airlines now require all cockpit conversations to be in English.
I’m trying to find fascinating report I found (from the NTSB) that argued that using a second language had a serious cultural effect - so speaking English would shift mindsets.
Look at French accident reports for that.
They were having a 400 paragraph argument about which cheese as they crashed?
You need to know what each piece of process is gaining you, but some parts are necessary.
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
That’s why attempts for planes to have only one pilot are doomed to fail. Almost everything a pilot does, is cross-checked with the other pilot, and these cross checks pick up a whole load of human errors. The chances of two humans making a mistake that could kill them, is orders of magnitude more than a single human making the same decision.
Assuming the culture allows a junior second pilot to question their senior captain.
A culture that’s been assiduously enforced, at least at Western airlines, since the Teneriffe crash of 1977. Yes, a junior pilot can call out a senior Captain.
All well documented and it’s hoped ‘lessons have been learned’! I think Korean Airlines now require all cockpit conversations to be in English.
I’m trying to find fascinating report I found (from the NTSB) that argued that using a second language had a serious cultural effect - so speaking English would shift mindsets.
Look at French accident reports for that.
They were having a 400 paragraph argument about which cheese as they crashed?
If you get the wrong cheese is there any point in going on?
You need to know what each piece of process is gaining you, but some parts are necessary.
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
That’s why attempts for planes to have only one pilot are doomed to fail. Almost everything a pilot does, is cross-checked with the other pilot, and these cross checks pick up a whole load of human errors. The chances of two humans making a mistake that could kill them, is orders of magnitude more than a single human making the same decision.
Assuming the culture allows a junior second pilot to question their senior captain.
A culture that’s been assiduously enforced, at least at Western airlines, since the Teneriffe crash of 1977. Yes, a junior pilot can call out a senior Captain.
All well documented and it’s hoped ‘lessons have been learned’! I think Korean Airlines now require all cockpit conversations to be in English.
I’m trying to find fascinating report I found (from the NTSB) that argued that using a second language had a serious cultural effect - so speaking English would shift mindsets.
Look at French accident reports for that.
They were having a 400 paragraph argument about which cheese as they crashed?
They might as well have been, for all that anyone else on the airfield could understand what they were saying.
Measuring to the nth degree benefits three groups. Non specialist managers, politicians and the media. They don’t understand what is being measured, but can compare result x to target y, and use it to suit their own purposes; which probably won’t be the purposes originally intended.
You need to know what each piece of process is gaining you, but some parts are necessary.
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
That’s why attempts for planes to have only one pilot are doomed to fail. Almost everything a pilot does, is cross-checked with the other pilot, and these cross checks pick up a whole load of human errors. The chances of two humans making a mistake that could kill them, is orders of magnitude more than a single human making the same decision.
Assuming the culture allows a junior second pilot to question their senior captain.
A culture that’s been assiduously enforced, at least at Western airlines, since the Teneriffe crash of 1977. Yes, a junior pilot can call out a senior Captain.
All well documented and it’s hoped ‘lessons have been learned’! I think Korean Airlines now require all cockpit conversations to be in English.
I’m trying to find fascinating report I found (from the NTSB) that argued that using a second language had a serious cultural effect - so speaking English would shift mindsets.
Look at French accident reports for that.
They were having a 400 paragraph argument about which cheese as they crashed?
They might as well have been, for all that anyone else on the airfield could understand what they were saying.
Good piece. My bugbear is "targets" rather than "process" - they are toxic - but there's no doubt process can become counterproductive and if left to grow unchecked a monster. Where it comes into its own (imo) is where you're trying to master something as an individual. There it's advisable to forget about the desired outcome of getting better and concentrate purely on the "process" of getting better. So, sport, music, cooking, these type of areas.
A related issue to all this is "measurement". I think we could usefully do a bit less of this. Eg you should not normally spend more time/money measuring how big of a problem something is than you do in resolving it. Yet we often do. Also the very act of measuring something can change it and so you are likely to get a false reading.
Ahem. Given that my job involves measuring things, may I put the contrary view? To do it correctly, you fit the metric to the task. If you are spending too much time measuring, the cure is to measure something else, not "no measurement". Classical statistics was created to measure a big thing quickly using a small thing. That's (one of the reasons why) I think Big Data is not a universal panacea.
Yes of course. It all depends. But an example of what I mean from my City days - involving traders and accountants both of which I was at various times.
So you'd have traders and every day at close of business each one would be presented with their "Daily P/L". This told them how much they had made or lost that day on their activities.
It typically took a whole bunch of qualified accountants (overpaid compared to everybody other than the traders) to produce this thing and they'd use a system linked to but separate from the main ledger. At month end the two systems never agreed so there'd be another team of professionals tasked with reconciling them. What a palaver.
And for what? Why did a trader have to know their profit or loss before they knocked off? They didn't. It was a nice to have at best. It's risk you need to monitor constantly not how much money you're making. So why did a whole operation-within-the-operation, with serious headcount and IT spend, exist to provide this info? Nobody could ever tell me.
Not my field, but I refer the Honourable Gentlemen to my earlier answer;
you lose the ability to crown winners and punish losers. (Even if both the winners and losers aren't statistically robust).
And for many, the crowning and punishing are the point of the exercise, even if the real information is what's happening around the average.
(I think it's a personal advantage to me that that part of my psyche remains massively underdeveloped, certainly compared with the average City trader. Society needs herbivores and carnivores, and being a carnivore just looks so damn stressful.)
Yes I wouldn't try to change that if I were you. My conclusion was this particular measurement exercise - this process - was a pander to traders and the resource devoted to it escaped scrutiny because there was so much money sloshing around in the sector.
I think on measurement exercises generally (which does account for a lot of process) you ought to detach yourself and ask, does the benefit of knowing this thing justify the effort/cost of finding out? If that's not a clear yes, forget it.
And then there’s another issue (even with a yes) where the benefit is to one set of people but the effort comes from another. But let's not get onto capitalism. Replacing that is above my paygrade.
Or, as Nicola put it: "Former senior civil servants, men and women trained in the art of diplomacy, recall with a shudder her foul-mouthed phone calls. “I don’t f**ing care about your f**ing process, I just want it done,” was typical of how Sturgeon conducted government business, according to one official I spoke to recently."
Good piece. My bugbear is "targets" rather than "process" - they are toxic - but there's no doubt process can become counterproductive and if left to grow unchecked a monster. Where it comes into its own (imo) is where you're trying to master something as an individual. There it's advisable to forget about the desired outcome of getting better and concentrate purely on the "process" of getting better. So, sport, music, cooking, these type of areas.
A related issue to all this is "measurement". I think we could usefully do a bit less of this. Eg you should not normally spend more time/money measuring how big of a problem something is than you do in resolving it. Yet we often do. Also the very act of measuring something can change it and so you are likely to get a false reading.
Ahem. Given that my job involves measuring things, may I put the contrary view? To do it correctly, you fit the metric to the task. If you are spending too much time measuring, the cure is to measure something else, not "no measurement". Classical statistics was created to measure a big thing quickly using a small thing. That's (one of the reasons why) I think Big Data is not a universal panacea.
Yes of course. It all depends. But an example of what I mean from my City days - involving traders and accountants both of which I was at various times.
So you'd have traders and every day at close of business each one would be presented with their "Daily P/L". This told them how much they had made or lost that day on their activities.
It typically took a whole bunch of qualified accountants (overpaid compared to everybody other than the traders) to produce this thing and they'd use a system linked to but separate from the main ledger. At month end the two systems never agreed so there'd be another team of professionals tasked with reconciling them. What a palaver.
And for what? Why did a trader have to know their profit or loss before they knocked off? They didn't. It was a nice to have at best. It's risk you need to monitor constantly not how much money you're making. So why did a whole operation-within-the-operation, with serious headcount and IT spend, exist to provide this info? Nobody could ever tell me.
Not my field, but I refer the Honourable Gentlemen to my earlier answer;
you lose the ability to crown winners and punish losers. (Even if both the winners and losers aren't statistically robust).
And for many, the crowning and punishing are the point of the exercise, even if the real information is what's happening around the average.
(I think it's a personal advantage to me that that part of my psyche remains massively underdeveloped, certainly compared with the average City trader. Society needs herbivores and carnivores, and being a carnivore just looks so damn stressful.)
Yes I wouldn't try to change that if I were you. My conclusion was this particular measurement exercise - this process - was a pander to traders and the resource devoted to it escaped scrutiny because there was so much money sloshing around in the sector.
I think on measurement exercises generally (which does account for a lot of process) you ought to detach yourself and ask, does the benefit of knowing this thing justify the effort/cost of finding out? If that's not a clear yes, forget it.
And then there’s another issue (even with a yes) where the benefit is to one set of people but the effort comes from another. But let's not get onto capitalism. Replacing that is above my paygrade.
Why do you think that's anything to do with capitalism ?
Its rather a case of management/executive capture with, if its of negative financial return, the shareholders (ie the capitalists) losing out.
It will also happen in non-capitalist sectors as well.
Good piece. My bugbear is "targets" rather than "process" - they are toxic - but there's no doubt process can become counterproductive and if left to grow unchecked a monster. Where it comes into its own (imo) is where you're trying to master something as an individual. There it's advisable to forget about the desired outcome of getting better and concentrate purely on the "process" of getting better. So, sport, music, cooking, these type of areas.
A related issue to all this is "measurement". I think we could usefully do a bit less of this. Eg you should not normally spend more time/money measuring how big of a problem something is than you do in resolving it. Yet we often do. Also the very act of measuring something can change it and so you are likely to get a false reading.
Ahem. Given that my job involves measuring things, may I put the contrary view? To do it correctly, you fit the metric to the task. If you are spending too much time measuring, the cure is to measure something else, not "no measurement". Classical statistics was created to measure a big thing quickly using a small thing. That's (one of the reasons why) I think Big Data is not a universal panacea.
Yes of course. It all depends. But an example of what I mean from my City days - involving traders and accountants both of which I was at various times.
So you'd have traders and every day at close of business each one would be presented with their "Daily P/L". This told them how much they had made or lost that day on their activities.
It typically took a whole bunch of qualified accountants (overpaid compared to everybody other than the traders) to produce this thing and they'd use a system linked to but separate from the main ledger. At month end the two systems never agreed so there'd be another team of professionals tasked with reconciling them. What a palaver.
And for what? Why did a trader have to know their profit or loss before they knocked off? They didn't. It was a nice to have at best. It's risk you need to monitor constantly not how much money you're making. So why did a whole operation-within-the-operation, with serious headcount and IT spend, exist to provide this info? Nobody could ever tell me.
Not my field, but I refer the Honourable Gentlemen to my earlier answer;
you lose the ability to crown winners and punish losers. (Even if both the winners and losers aren't statistically robust).
And for many, the crowning and punishing are the point of the exercise, even if the real information is what's happening around the average.
(I think it's a personal advantage to me that that part of my psyche remains massively underdeveloped, certainly compared with the average City trader. Society needs herbivores and carnivores, and being a carnivore just looks so damn stressful.)
Yes I wouldn't try to change that if I were you. My conclusion was this particular measurement exercise - this process - was a pander to traders and the resource devoted to it escaped scrutiny because there was so much money sloshing around in the sector.
I think on measurement exercises generally (which does account for a lot of process) you ought to detach yourself and ask, does the benefit of knowing this thing justify the effort/cost of finding out? If that's not a clear yes, forget it.
And then there’s another issue (even with a yes) where the benefit is to one set of people but the effort comes from another. But let's not get onto capitalism. Replacing that is above my paygrade.
Why do you think that's anything to do with capitalism ?
Its rather a case of management/executive capture with, if its of negative financial return, the shareholders (ie the capitalists) losing out.
It will also happen in non-capitalist sectors as well.
Read the The Mitrokhin Archive - the KGB died of pointless process in the middle 1970s. Meetings about the meetings to decide the biscuits for the meeting…. For something.
You need to know what each piece of process is gaining you, but some parts are necessary.
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
That’s why attempts for planes to have only one pilot are doomed to fail. Almost everything a pilot does, is cross-checked with the other pilot, and these cross checks pick up a whole load of human errors. The chances of two humans making a mistake that could kill them, is orders of magnitude more than a single human making the same decision.
Assuming the culture allows a junior second pilot to question their senior captain.
A culture that’s been assiduously enforced, at least at Western airlines, since the Teneriffe crash of 1977. Yes, a junior pilot can call out a senior Captain.
All well documented and it’s hoped ‘lessons have been learned’! I think Korean Airlines now require all cockpit conversations to be in English.
I’m trying to find fascinating report I found (from the NTSB) that argued that using a second language had a serious cultural effect - so speaking English would shift mindsets.
Sounds reasonable. I would have to speak German or French a very great deal than I do to make a lifesaving point.
JPR and Barry John both dead within a month. I don't often feel so strongly that we are never going to see their like again, nor perhaps see any sport played the way they played it.
Right, so just abolish process and trust a strong leader unconstrained by rules who will Get Things Done.
Thankfully, we should be grateful that one underappreciated side effect of the British obsession with class is that our class system is perfected to inhibit any attempt at domestic fascism.
For our home-grown aspirant far-right leaders, such as Home Counties born privately-educated Farage or Home Counties born privately educated Tice, have their natural environment as the saloon bar of a Surrey pub, regailing their stockbroker friends with their ‘fnarr fnarr’ anecdotes on a Sunday afternoon….whilst their prospective electorate is out in places like Jaywick, trying to keep their fighting dogs away from the authorities. The appeal to the likes of Farage or Tice of spending their spare time knocking on mobile home doors in Jaywick and the like quickly pales.
Comments
The DfE became the department that knows next to nothing about education.
That is of course bullshit.
The words 'next to' have somehow inserted themselves to suggest, wrongly, that the DfE know something about education.
Not particularly my cup of tea musically, but here it is
https://www.youtube.com/watch?v=zWa8mhYLFt4
Edit: nope, 4th. Have to LOL at Chelski 1-3 Wolves.
I am, unfortunately, a cynic.
Something that is far easier and produces a great soundbite is "we have reduced the number of administrators by a quarter!"
Always popular. Get rid of the beancounters and paper-shufflers. Easy savings. Tea and medals and the adulation of an adoring public.
Unfortunately the amount of administration is either unchanged or increased. So instead of admin assistants, the social workers, surgeons, teachers, and so on end up having to do it and we end up with them spending even less time at their primary job (and taking longer to do the paperwork than a specialist and doing it worse).
We see this, call for change, and someone reacts by reducing the number of remaining administrators by a quarter...
Actually addressing the process as you call for would be the solution. But take longer and require investing time and resources to analyse the processes and then pay to have them changed. Far easier to simply fire a few existing administrators and load up the social workers and doctors with a bit more of the paperwork. After all - who sees that bit beyond the front-page savings in reduced admin staff?
You do need to be the master of the process and not its slave however.
When there is a pressing problem process just gets set aside. Russia built the Kerch bridge in 2 years, 18 miles long, the biggest bridge in Europe, and it still stands despite the various attempts to blow it up in the current war. We cold do with adopting a similar attitude to things like climate change and housing.
Bleakness upon bleakness
Where UK does better IMO is online... many many things can be done online much more easily than when I have to interact with other govts.
Change becomes increasingly difficult as organisations grow larger and older.
But this time its about Germany.
https://www.bbc.co.uk/news/world-europe-68181799
Doubtless some Germans are blaming Olaf Scholz for their local market struggling or because a single minimum wage doesn't provide affluent living for a family of six.
And similar pieces could be done about all developed countries.
But moaning about things or blaming governments for not providing everything demanded isn't going to achieve anything anywhere.
The way to improvement is through work not whining, through increasing productivity not increasing demands, through taking the opportunities on offer not expecting others to provide more.
Successful investment is also difficult and usually requires the same things.
Much easier to keep doing the same things, hope problems disappear and pay yourself another bonus.
Think of them like electron shells. Each layer expands but it becomes progressively more difficult to jump to the next one. This third "Ydoethur" phase in PB Cosmic Evolution is analogous to the d-block: the articles are rusty, metallic, and go "clang" when you strike them. The articles will be bigger on the inside than on the PB outside, which is only logical when you consider the name.
A related issue to all this is "measurement". I think we could usefully do a bit less of this. Eg you should not normally spend more time/money measuring how big of a problem something is than you do in resolving it. Yet we often do. Also the very act of measuring something can change it and so you are likely to get a false reading.
The examples I know something about from your list - Boeing, Horizon etc - the root cause is poor culture in those organisations. In general organisations with good culture have effective processes, but not necessarily fewer of them. I can think of processes the Post Office would have benefitted from, including a proper escalation process for people raising concerns.
Sometimes change for change's sake becomes a fetish. Perhaps tinkering would be a better word, but having Changed Something is pretty much a prerequisite for promotion in schools. Either that, or being the only full time permanent body left standing. And a lot of those changes may individually be desirable- adding something nice or closing off the pathway to something bad happening. But the overall effect is something like adding another bit of luggage to the donkey in Buckaroo, it's fine, it's fine and it stays that way until the donkey throws everything off.
As to the answer to the problem @Malmesbury describes so well... I'm sure that milages vary. But I suspect the issue is that remote micromanagement is a possibility now in a way that it wasn't before. Partly because of electronic documentation of process, but also because of the speed of electronic communication. It's probably still a bad idea, because an organisation, let alone a nation, shouldn't fit inside one brain, however enormous.
It's an attractive idea that everything can be run from one centre issuing commands. It appeals to fairness (no postcode lottery) and apparent efficiency (fewer politicians) and vanity (I can command everything). It's just that, even if the technology now exists (unlike 1970s Chile or 1960s Soviet Union), life isn't like that.
And for many, the crowning and punishing are the point of the exercise, even if the real information is what's happening around the average.
The devil will be in the details, but an obvious question that comes int my mind is why no-one further down the production line realised the flaw. Is it because they are so tuned to the process that they're not *meant* to notice issues? A human version of 'the computer says no'...
I made a few changes to PB in the last 24 hours to solve a few issues people were having:
(1) Stopped using external caching, because it was throwing errors. (I'll reimplement something closer to election night, but for now page load times are sufficiently low that it's fine.)
(2) Implemented proper forwarding for people who went to http rather than https
(3) Solved the bug where sometimes people would see the default Apache2 page rather than pb
If there are any other issues, please let me know. My email address is simply my username at gmail.com
You’ve forgotten to put up the donation link for the new server fund.
Kind regards
Sandpit.
There's a discussion here: https://leehamnews.com/2024/01/15/unplanned-removal-installation-inspection-procedure-at-boeing/#comment-509962
I do have a related angle on this, though. How did we end up with so many rules and regulations?
Two answers are stupidity and malice. You have a loose set of rules and regulations, but someone does not do step 1 properly and causes a disaster - a fallen bridge, or a child being abused. Either because they made a mistake, or because they could not be bothered to do their job properly, perhaps because they gained from it.
There are many possible 'fixes' for this. Two are:
1) Better training and employees.
2) Add another later of process in, so the results of step 1 are checked.
Either. or both, of these can work. Or several others, depending on the application.
Let's look at something like railway signalling. A new signal can cost hundreds of thousands of pounds to install. That sounds stupid. But if you look at the reasons for that cost, then you can see incidents in the past that have led to the rules and regulations that are now in place, and lead to the cost. And those incidents often cost lives. Do we want to save money and put more risk onto rail workers' (or passengers)?
I don't know what the answer is. In the case of the £250k bridge in Stiffkey, I'd guess it's because the NT don't want it and have elevated costs as much as possible. Or perhaps not.
Yes, that does appear to be the grandmother at the end.
Horrifying little vignette of English life. I wonder how the mother died.
So is Rishi Sunak.
No wonder the country is buggered.
So you'd have traders and every day at close of business each one would be presented with their "Daily P/L". This told them how much they had made or lost that day on their activities.
It typically took a whole bunch of qualified accountants (overpaid compared to everybody other than the traders) to produce this thing and they'd use a system linked to but separate from the main ledger. At month end the two systems never agreed so there'd be another team of professionals tasked with reconciling them. What a palaver.
And for what? Why did a trader have to know their profit or loss before they knocked off? They didn't. It was a nice to have at best. It's risk you need to monitor constantly not how much money you're making. So why did a whole operation-within-the-operation, with serious headcount and IT spend, exist to provide this info? Nobody could ever tell me.
We'll know more when the reports come out.
you lose the ability to crown winners and punish losers. (Even if both the winners and losers aren't statistically robust).
And for many, the crowning and punishing are the point of the exercise, even if the real information is what's happening around the average.
(I think it's a personal advantage to me that that part of my psyche remains massively underdeveloped, certainly compared with the average City trader. Society needs herbivores and carnivores, and being a carnivore just looks so damn stressful.)
For example, we know from back in the days when banks were run by clerks, that even simple human tasks have an error rate of about 1 in 500. To get a lower error rate than that, which is essential in many tasks, you need cross checking of some kind. But if people don't understand why this is, then 2 people checking turns into 0 people because they each thought the other had checked.
The best kind of cross checking is where the second check is where you have some kind of telltale which works on larger batches, eg 'I have checked that each pole is vertical, and now I can see they are all lined up'.
Often this kind of report gets written in the passive voice. Your teachers probably told you in school that scientific reports should be written in the passive voice. But did they tell you why? Probably not. The reason is, that science, which is intended to produce universal truths, is not supposed to depend on who is doing the process. In other areas it may be essential that the sheep was personally drowned by Damian Hirst, the cigar was rolled between virgin's thighs, or that the elixir was compounded by the only alchemist with sufficient skill. In science, this is not good practice. But most work is not science. If a process is supposed to enforce that things were done properly, eliding who did it just disclaims responsibility.
Where I partly disagree is that while there is often too much process at a lower level, there is often too little at the top. Leaders and high executives are allowed to get away with a narrative of success rather than having well defined responsibilities. The culture of vague aspirational claims is bad enough in consumer marketting, it's downright dangerous elsewhere.
(*) If my understanding of what happened is correct...
They were also trying to use delivery snagging as a substitute for QA. Instead of building planes right, just chuck them together. And believe that the pre delivery checks would find the problems.
So you had process interference multiplied by process abuse. Plus straight up incompetence.
https://www.instagram.com/p/Cpm6j8tIgsY/?igsh=OHNqcHpzdTNpYnJn
"Former senior civil servants, men and women trained in the art of diplomacy, recall with a shudder her foul-mouthed phone calls. “I don’t f**ing care about your f**ing process, I just want it done,” was typical of how Sturgeon conducted government business, according to one official I spoke to recently."
https://www.msn.com/en-gb/news/other/nicola-sturgeon-s-behaviour-suggests-she-is-a-politician-who-never-grew-up-susan-dalgety/ar-BB1hHpAX
Potentially disagree with your second line. Minor problems will always occur, and I bet Airbus has people from (say) Broughton fixing minor snags on the main production lines in France. It's probably better than having people trying to fix problems Spirit created, for a variety of reasons.
https://en.m.wikipedia.org/wiki/Crew_resource_management
(Certain Asian airlines have struggled with this, for well-documented cultural reasons).
https://en.m.wikipedia.org/wiki/Asiana_Airlines_Flight_214
https://en.wikipedia.org/wiki/British_European_Airways_Flight_548
"An hour and a half before the departure of Flight 548, its rostered captain, Stanley Key, a former Royal Air Force pilot who had served during the Second World War, was involved in a quarrel in the crew room at Heathrow's Queens Building with a first officer named Flavell. The subject was the threatened strike, which Flavell supported and Key opposed. Both of Key's flight deck crew on Flight 548 witnessed the altercation, and another bystander described Key's outburst as "the most violent argument he had ever heard".[6] Shortly afterward Key apologised to Flavell, and the matter seemed closed."
Boeing is (and was) delivering broken planes and trying to fix them pre delivery. That’s process abuse. They shouldn’t be fixing that many issues that late in the game.
Spirit is a bullshit way of putting a chunk of Boeing at arms length. So the bean counters only need to count beans. It’s really a part of Boeing - just semi-disconnected.
Congratulations, you win the prize for what I expected.
Perfectly lined up with everything else that Sturgeon fucked up.
https://www.youtube.com/watch?v=rWFtoqxj93U
I think Korean Airlines now require all cockpit conversations to be in English.
The skill is to use extreme politeness and sympathy as you reduce them to 10-12mm in height.
The world is full of process designers. There have to be some. There have to be rules and protocols and standards. And the world, on the whole, divides into doers and admin people, including process designers. So there are people who collectively build and maintain dwellings (eg Grenfell), and people who are part of ancillary processes.
Each group wishes to avoid blame; they are human.
The process designers can only avoid blame by covering all bases, even though they know their millions of pages will be mostly unread, ignored or not recalled.
The actual doers can't possibly cover all the bases, because it is impossible and impractical to know it all. So they hide behind blaming each other, apologising early and often, ceasing trading, and the force of inertia.
Non specialist managers, politicians and the media. They don’t understand what is being measured, but can compare result x to target y, and use it to suit their own purposes; which probably won’t be the purposes originally intended.
I think on measurement exercises generally (which does account for a lot of process) you ought to detach yourself and ask, does the benefit of knowing this thing justify the effort/cost of finding out? If that's not a clear yes, forget it.
And then there’s another issue (even with a yes) where the benefit is to one set of people but the effort comes from another. But let's not get onto capitalism. Replacing that is above my paygrade.
Its rather a case of management/executive capture with, if its of negative financial return, the shareholders (ie the capitalists) losing out.
It will also happen in non-capitalist sectors as well.
And Haaland and De Bruyne back.
Allardyce?
For our home-grown aspirant far-right leaders, such as Home Counties born privately-educated Farage or Home Counties born privately educated Tice, have their natural environment as the saloon bar of a Surrey pub, regailing their stockbroker friends with their ‘fnarr fnarr’ anecdotes on a Sunday afternoon….whilst their prospective electorate is out in places like Jaywick, trying to keep their fighting dogs away from the authorities. The appeal to the likes of Farage or Tice of spending their spare time knocking on mobile home doors in Jaywick and the like quickly pales.