- The big squiggly mess in the article is filled with people. I think Deming’s deepest concept was giving workers on the production line simple tools to improve processes on their own. His books are filled with exhortations to trust the workers. This is what American managers could never bring themselves to do.
Even in manufacturing, the application of statistical process control was never entrusted to the workers, but became a department of its own, with bureaucracy, OKRs, and elaborate software.
- > for thermostat B there are many more outliers. We’d say that [...] thermostat B is not [under statistical process control]. (In practice, you’d draw a control chart to identify whether the system is under statistical control).
I did draw the control chart, and thermostat B is definitely under statistical process control: https://xkqr.org/info/xmr.html?baseline=33,97,41,65,72,71,64...
by anonymousiam
6 subcomments
- The main point that I did not see mentioned in this piece is that Deming should only be applied to MANUFACTURING environments, because things like engineering are too chaotic to identify processes or trends in the engineering itself, and trying to control those engineering processes with SPC doesn't really improve the quality of the engineering, it just adds stress, makes things take longer, and probably lowers the quality of the thing that is being engineered.
Obviously, if a quality issue is detected in manufacturing, there may be some steps that engineering could take to improve the manufacturing process and make things stable enough to obtain meaningful statistics. This is part of the Deming feedback process, and part of the System Engineering Life Cycle.
by whatever1
3 subcomments
- Fundamentally stock markets won the world of business, so everything has a horizon of a financial quarter.
Hence, every action of a company needs to be measured against the upcoming quarterly results.
OKRs et al are great at that.
Who cares about quality/sustainabily. We just want the stock go wheeeeee and get our bonuses.
- This is a very trivial treatment of Deming and I’m surprised how it makes its way to the top of HN.
The arc from Walter Shewhart to W.E. Deming is a bedrock foundation in an Industrial Engineering curriculum. These men paved the manufacturing process quality principles of modern industrialization. Drucker was about management science, truly an apples to oranges comparison.
- It is also worth noting that US management is notoriously bad at the actual management. Toyota v. US car manufacturers did not look like a fair fight when Deming was in the ascendant, and it is hard to tell given the scales involved but it looks a lot like the US has been outmanoeuvred in all aspects of industry by the Asians.
US companies are generally a better bet though, because despite the handicap of being run by Americans, they are hosted in a country that generally believes in freedom and rule of law which means they have an unfair advantage even if they do a sub-par job of making the most of what they have.
Exceptions abound in the details.
- Maybe it’s my limited intellect but I found Drucker to be a lot easier to understand.
Where Deming reads like a science paper, Drucker reads like an installation guide.
by ontouchstart
0 subcomment
- This topic is very relevant in the age of agentic AI when every decision is a statistical next token prediction “trained” on some loss function. AGENT.md, SOUL.md etc are just smoke and mirrors of The Wizard of the Oz.
Eventually manager as a profession will be replaced by tools, just like computer as a profession, editor as a profession.
The evolution of computer science will be manager science. There is more than loss function and KPI.
by foobarbecue
2 subcomments
- For the millionth time, would it kill ya to spell out the abbreviation the first time you use it? My googling suggests we're talking about https://en.wikipedia.org/wiki/Objectives_and_key_results , but my googling isn't always right.
- The analogy I used with the team was that, set the goal, present the map, and figure how to make a better map. Drucker was about the goal with a given map. It is not uncommon for people receiving the OKR not resonating with it. Sometimes they actually have insight into making a better map, but if OKR is OKR, one just have to follow, people swallow their thoughts
by iamflimflam1
2 subcomments
- This made me spit out my coffee…
> One of the virtues of OKRs is that they are straightforward for managers to apply.
by ineedasername
0 subcomment
- * Eliminate management by numbers, numerical goals. Substitute leadership.*
That’s a lot harder, and more expensive in the short term. It’s also something that benefits from having long term employees that both understand the organization and don’t view the organization as hostile to their employee interests whenever some other metric of short term profitability is a mutually exclusive choice.
Without this, you’re limited to people who may have more natural or intuitive leadership capabilities instead of those who can learn, given the opportunity and example.
In short, corporate practices have systematically eliminated the circumstances under which there would grow a sufficient number of people with leadership skills.
by lowbloodsugar
1 subcomments
- “ Drucker makes a manager’s life easier, Deming makes it harder”
This is why companies can’t do Agile, especially Scrum: Scrum requires the most of the people in power, who typically can’t be bothered and who get to dictate process.
- OKRs attempts to impose top down centralized command and control. What ends up happening is an executive who is accountable notices an OKR trends the wrong way and when he asks why, the best bullshitters deflect blame. And nothing is resolved.
The Toyota Way attempts bottom up process improvement. Teams generally organize themselves. Team leads take responsibility for quality and report it up the chain. Instead of deflecting blame, they often work themselves to exhaustion. Which is not an ideal result either.
- Optimize for mediocrity, you get mediocrity. Optimize for quality, you get quality.
Choose wisely.
- I think this article is a great opportunity to mention two under-used statistical techniques: Deming regression [0] and the Theil-Sen estimator [1].
They both fit straight lines to noisy data.
Deming regression is an errors-in-variables model that tries to fit the line of best fit when you have errors in both x and y and they are both known and in general _different_; the Theil-Sen estimator is based on medians and is particularly robust if you have an error process that fails more "one way" than the other. Simple linear regression is everywhere in our lives and yet remarkably not robust to errors that are not IID normal, particularly with a small number of data points: a process that can only fail in one direction if it breaks is likely to completely and utterly bugger up the line that you fit. Both approaches have their place and I wish were more widely used, particularly by people who like fitting linear models to complex phenomena because they are easily understood.
[0] https://en.wikipedia.org/wiki/Deming_regression
[1] https://en.wikipedia.org/wiki/Theil%E2%80%93Sen_estimator
- yea that system was a lot more than he was prepared for :)
- [dead]
- [dead]
- Or you can just abolish the fiat soft money system, let the corporations go out of business, let efficient small companies take their place and then you'd have founders who actually care about long term results in charge and they could manage the company however they want. If they do a bad job, they'd go out of business. If they do a good job, they'd stay afloat or maybe even grow a little. And over time, all the companies with incompetent leaders would be wiped out, everyone would get a fair shot at being a leader and everyone would end up in their rightful place and capitalism would function as it was designed.
Hacker News