I was a long time Scott Adams fan with the Dilbert Principle being one of my favorite books.
What I found most interesting about him was around the time Trump was running for president the first time, Adams was one of the first people to point out that Trump was, to use Adams' terms, a "master persuader". No one else at the time seemed to be talking about this and it was fascinating to see a humorist have this take/insight.
I think many who dislike Trump (myself included) don't really want to think of him as having skills of any sort.
But I think it's more so that he does absolutely have certain skills such as persuasion or, some argue, charisma. He just doesn't have any of that pesky morality or sense of responsibility to the greater good, the entire citizenship, etc. that often gets in the way of such ambitions.
So we're left with a master manipulator who will hurt a great number of people, maybe benefit a few if necessary, but ultimately a subset of people think he's genius and a net positive. And I can't help but think that the only ones who think he's a "net positive" are either personally benefiting, or have been persuaded to believe it, despite reality painting a different picture.
I dunno if the "just paraphrasing [...] Fox" works as an explanation for success. It sounds like you believe he just keeps unaccountably stumbling into piles of cash and power?
An ageing Biden and Dubya have also occupied that office and they don't exactly strike me as "master persuader" types either.
Nobody is accusing Trump of lacking ambition or charisma, and there's also no doubt the party machine that backed him is pretty sophisticated in the arts of political campaigning. But there's a difference between being a "master persuader" able to convince almost anyone of almost anything and being a shameless braggart in front of an electorate that's unusually impressed by a celebrity's overconfidence and wealth, and also being a lot less shameless about appealing to their chauvinistic attitudes than predecessors.
Trump was born rich to a father who taught him cruelty and insulated him from consequences. It was a golden ticket.
He still managed to go bankrupt 6 times, and couldn't get financing. He had to resort to selling his name or getting money from one of the most corrupt banks in the world.
He's rumored to have been despised in the NY social scene since his youth and up to the present.
He's been accused of rape by his own ex-wife and SA by more than 20 others. He bought pageants so beautiful women would have to interact with him. His longest relationship is with an illegal migrant (possibly trafficked) escort whose visa he had to pay for.
He gained no following during his time at the head of the Reform party.
Since 2015 his political base, like Nixon's, is largely built on white grievance and fear. It's incapable of building much once in power.
Now the Trump family accumulates money by selling power, hot air, and fleecing fools.
For those looking for a "successor theory" to the Dilbert Principle, I highly suggest Venkatesh Rao's Gervais Principle [0].
To use Dilbert terms: Adams would say that PHB is dumb and he is promoted into management as that's where he can do the least damage.
Rao would say that PHB is actually put there by upper management to be a combination of:
- fall guy/lightning rod to take blame for failed projects
- dumb subordinates are less likely to try to take your job (dumb doesn't mean unintelligent. Rather, Rao uses the term "clueless" to highlight smart people who are not political)
The Gervais Principle is much more accurate in my experience. One of the important reasons middle management has to be "clueless" to drink the kool-aid and take on more responsibility for minimal extra compensation. The checked out employees of the world know their work is meaningless, but the clueless ascribe to it some greater meaning which makes them trustworthy.
No. A first level manager cannot be middle management. A small company might not have middle management but the first level manager is bottom management.
We are teaching the sand to think and working on 3d printing organs and peering at the beginning of time with super-telescopes and landing rockets.
Then look at our leadership class. Look at the leaders of the most powerful countries. Look at the most powerful leaders in finance and business.
Look at that contrast. It’s very clear where the actually smart people are.
But those actually smart people keep putting leaders like that in power. It’s not a conspiracy. We do it. We need them for some reason.
I have two hypotheses.
One is familiar: they are sacrificial lightning rods. Sacrifice the king when things don’t go well.
The other is what I call the dopamine donor hypothesis. Compared to the speed and complexity of the modern world, most human beings are essentially catatonic. Our dopamine systems are not calibrated for this. So we sit there and do nothing by default, or we play and invent but lack the intrinsic motivation to do the hardest parts.
So we find these freaks: narcissists, delusional manic prophets, psychopaths. They’re deeply dysfunctional people but we use them. We use the fact that they have tireless non stop motivation. Dopamine always on. Go go go.
We place them in positions of authority and let them drive us, even to the point of abuse, as a hack to get around the fact that our central nervous systems don’t natively do this.
Then of course if things go wrong, it’s back to their other purpose: sacrificial scapegoats.
So in a sense we are both victims of these people and exploiters of them. It’s a dysfunctional relationship.
If we could find ways to tweak our systems like amphetamine but without the side effects, we could perhaps replace this system with a pill.
It would be more compassionate for the freaks too. They’re not happy people. If we stopped using them this way they might get help and be happier.
Governance creates markets -> markets create innovation. These things have feedback loops into governance, but the tail ultimately does not wag the dog.
Engineers-- especially in the Bay where discussion of such is written off as mental illness-- often dismiss politics and governance as nonsense subjects that lack rules and are run by the mob/emotions. The reality however, is that these societal constructs have their own "physics" and operate like a (very complex and challenging to study) system just like everything else in the natural world.
The attitude itself is of course something has been designed and implemented into engineering culture by precisely the leaders you contend are scape goats to society. POSIWID.
> The attitude itself is of course something has been designed and implemented into engineering culture by precisely the leaders you contend are scape goats to society. POSIWID.
I don’t know if this particular statement is true or not, but the number of smart people I know who thinks they’re not affected by propaganda is wild. We’re all affected by propaganda.
I am not sure this is necessarily the case, at least historically. We have good evidence of long distance trade from the Stone Age, and even some Neanderthal sites contain stones whose origin can be traced to distant regions (over 100 km, IIRC, which is far away in a primordial roadless countryside).
I would agree that markets cannot grow beyond a certain size without a government, though.
Those games operate far more probablistically and high dimensionally than programming and I suspect engineers would rather dismiss them as “dumb” than accept they are simply inferior players in those games.
Primary multi agent multi dimension probabilistic resolution problems model human and crowd interaction better than “code do this every time”.
I’ve spent a long time in the valley and I’ve come to the personal conclusion that engineers are often the dumbest (and most narrowly useful) in the room not the smartest. And the rest of them let them think they are very smart (tm) so they do what we say.
How very Dilbertian. If one were to compress the above post into a comic, it would star Dilbert wondering why people with towering intellects like Dilbert weren't running the world in the first panel and then humorously demonstrating in subsequent panels Dilbert's disastrous and irreparable lack of understanding of messy human interrelationships and motivations that have to be navigated to not implode as a leader.
Well observed. And seen in tragic relief as the piles of dead in Russia and China during their most technocratic periods run by engineers.
Which wasn’t just about refusal to interact with humanity but to acknowledge that complex multi factor problems can’t be solved as top down heuristics.
The piles of bodies in China came from Mao and his cultural revolution, and he can hardly be called an engineer. The recent success of China has come when it was run by engineers. And when was Russia ever run by engineers? So I think you have it backwards here.
"Russia was never run by engineers?" That's a massive oversight of 20th-century history. The Soviet Union was the world’s first and most committed technocracy. The GOSPLAN (State Planning Committee) was a literal attempt to run a continent-sized economy as a deterministic engineering problem. By the 70s, the Soviet leadership was more densely packed with engineers than any administration in US history.
They failed because they tried to 'refactor' nature. Stalin’s 'Great Plan for the Transformation of Nature' and Mao’s 'Great Leap Forward' (which applied industrial throughput logic to biology/close-planting) are the ultimate warnings of what happens when you treat complex, probabilistic systems (ecology and humanity) like a closed-loop machine.
Mao wasn’t an engineer by degree, but he was a High Modernist by practice. He believed society could be 'debugged' and 'optimized' through central planning. The result wasn't a more efficient system; it was a total system crash that cost tens of millions of lives.
Current China is a perfect example of 'Success by Engineering'—high-speed rail and ghost cities built on a demographic 'memory leak' (the One Child Policy) that is now crashing the entire stack. This is exactly my point: Engineers optimize for the metric they can see, while ignoring the high-dimensional chaos that actually sustains life.
> We are teaching the sand to think and working on 3d printing organs and peering at the beginning of time with super-telescopes and landing rockets.
There are a lot of smart and skilled people involved in making a cutting edge chip fab. It's not one ubermensch in a basement inventing a new TSMC process by thinking really hard. There's technicians, scientists, researchers in multiple disciplines. All of those people have to be organized.
I don't know where you think the "smart" people are, but maybe meditate on the fact that "smartness" is not a single variable that dictates a person's value or success. Someone who is an expert at researching extreme UV patterning isn't going to necessarily run a great chip manufacturer.
It's pretty simple: those people are the absolute experts in their field, similar to those top chemists or whatever. That field is societal power systems.
Of course someone who dedicated his time to climbing and understanding power systems will have more power than someone who doesn't.
Sure, but then my question is why we need them. What service do they provide? That’s what I was speculating about. I don’t buy the conspiracy theory that they’re pure parasites, since hosts without parasites would then be stronger and would ultimately outcompete.
We have all the skills to do all the things without these power systems so what are they for?
I don’t mean policing and courts. Those are administrative and managerial functions. I mean power of the sort that makes large numbers of people do stuff. I mean gurus and aggrandizers, basically. The people who con and goad us into doing hard things.
My hypothesis is that we can’t self generate that due to neurological limitations rooted in our evolutionary history in a much slower world that rarely changed.
Amphetamine could work too but it has ugly side effects. Social pressure is less hazardous and scales better.
Managers are here to accommodate the need for cooperation, while compensating for lack of telepathy.
Put two people with a lot of expertise in different domain. Require them to come up with a solution to a problem you have.
That's three people. You'll get at the very least four opinions about each and every step.
Scale the complexity of the problems and the number of people.
You end up with full time jobs consisting purely in routing information from brain A to brain Z.
Unfortunately, the skills to do this job are never properly taught, but learnt in the job. (MBA don't teach management - they either teach the mechanism of some administration, or ways to get rich consulting.)
Problems occur because we conflate management, supervision, decision making, strategy setting, etc...
P.H.B. is an antipattern, a caricature, a stereotype like all other : it's funny cause there is truth to it. But we are by no mean condemned to fulfill our stereotypes (should I remind all engineers here about the stigmas attached to nerd in the real world ?)
> I don’t mean policing and courts. Those are administrative and managerial functions.
Middle management is also an admininstrative and managerial function. Even in a best-case scenario, coördinating work among a huge amount of people within enterprises that are mostly run via command-and-control mechanisms and inside politics (as opposed to any self-regulating "market") obviously takes a whole lot of effort. That's really the natural job description for PHB's.
The WIDGET model of "working geniuses" is one possible answer, it does explain a lot of team dynamics in my experience.
Since no one has all six working geniuses, and you're only a genius at two, it takes a collection of people, proportional to the work that needs to be done, of each type.
You got it backwards. We (which we?) don't need them, they need us. They can't play the games they like without massive resource extraction. If someone continually catches the flu, it doesn't mean they need the flu.
We don’t just use these people we create them. Since ancient Egypt the priest class of every society is employed to apply ritual trauma to psychologically prepare princes for their vocation of restless leadership.
I have a niche Instagram account that goes out to find content and then "reposts" it. There were several fun aspects of this e.g. finding good content, writing my own little algorithm to prioritize contents from older posts on smaller accounts etc.
Lately, much as others have said, you are seeing entire accounts of AI generated images that are high quality, near photo realistic and consistent e.g. it looks like the same person in different scenes/times of day etc
You sometimes hear the quote about "pre-war steel" that hadn't been hit by radiation and that's EXACTLY what it feels like looking for an account with posts from before ~2022.
I wonder if the above means that people are going to spend less time online and prioritize "in real life" events or if the slop is just going to get more addicted.
Probably a mix of both in the same way that Tough Mudder/Spartan Races became popular while at the same time the number of other people NOT leaving their houses went up.
I think CEO of Instagram publicly admitted that the Social Media aspect of Instagram peaked around 2020 and has been declining since - he claimed it was being replaced by more group chats, private stories.
I was the head of the men's group for my town's Newcomers Club and when it comes to having a blueprint for organizing a small, quick gathering I cannot say enough about Nick Gray's [0] book The Two Hour Cocktail Party [1]
It has tons of small but useful tips:
- host it Monday or Tuesday from 7-9pm. People are usually free those nights and make sure it ends at 9pm for the folks who have to wake up early
- don't send an evite with "0 of 60 guest have responded". Start by having your core group accept and then send the invite directly to each new person
- have name tags. but make sure YOU fill out the name tags or you will have "Batman" and "Superman" at your party
- introduce people and have "get to know you games"
Now, I'm sure someone will say "this is so formulaic and doesn't feel natural!". That's kind of the point. You need to give folks some structure to be able to interact. The name tags for example remove the "oh, I met this person before but I can't remember their name so I just won't talk to them" etc.
In a general sense, I think two things about this suggestion are universally useful in solving the epidemic: structured activities with clear expectations, and people who take the initiative to design and execute them.
For those people who I'm personally trying to reach, those who sit at home alone all day, and you only see them on the way to the grocery store and back, who desperately want to interact with people but don't know how to begin, they need someone to initiate the interaction, and they need to know the rules of the interaction, since this is mainly the reason they don't feel comfortable with freeform interactions with strangers and so avoid them.
> they need someone to initiate the interaction, and they need to know the rules of the interaction, since this is mainly the reason they don't feel comfortable with freeform interactions with strangers and so avoid them.
You hit the nail on the head with the above.
It's why things like square dancing where so popular back in the day.
It gave men and women an easy to learn, simple to follow set of rules for interacting with the opposite sex with as little ambiguity as possible. e.g someone was literally calling out what to do next so you could enjoy the moment rather than thinking "oh no, I'm not sure what to do next!"
One of my favorite stories from the dotcom bust is when people, after the bust, said something along the lines of: "Take Pets.com. Who the hell would buy 40lb dogfood bags over the internet? And what business would offer that?? It doesn't make sense at all economically! No wonder they went out of business."
Yet here we are, 20 years later, routinely ordering FURNITURE on the internent and often delivered "free".
My point being, sure, there is a lot of hype around AI but that doesn't mean that there aren't nuggets of very useful projects happening.
Fun fact: apart from the main office SW1 they're alphabetised by area, from SW2 Brixton to SW19 Wimbledon. All of the London postcode areas are like this.
SW2 to SW9 are in alphabetical order: Brixton, Chelsea, Clapham, Earls Court, Fulham, South Ken, South Lambeth, Stockwell.
But then it starts again and you have to squint a bit for SW10-SW20: Brompton, Battersea, Balham, Barnes, Mortlake, Putney, Streatham, Tooting, Wandsworth, Wimbledon, West Wimbledon.
Looking at a few others (SE, etc) I see that the first chunk of them are in alphabetical order, but then they've added some extra ones later that break the ordering (e.g. SE19 onwards) but they have tried to add the extra ones in mostly alphabetical order too.
Yeah, they've become a bit muddled over the years but generally alphabetical in the batches they're added. E was nice and clean before the Olympics, then they added E20 for Stratford after E18 Woodford.
Most people assume it's relative to how far out the area is from the centre
And a bit further to SW20 in Raynes Park (a.k.a. “West Wimbledon” in Estate Agent vernacular).
I’ve lived somewhere in SW18/SW15/SW19 for the last 30 years. Having not grown up in London I can’t imagine living anywhere else. Apparently many other bits of London (North, East, central, etc) are good too but I’m not ready for change.
> It’s why the best engineers try to write as little code as possible. They understand that each additional line is another thing that must be understood, remembered, and maintained.
As a long time SRE/DevOps, this is the key point that I believe people are overlooking in the "AI will let anyone write code!!"
Support does not scale linearly with additional software. If anything, it grows faster due to the "two systems == one connection, three == three connections, four == six connections" etc.
Sure, you could argue that LLMs could ALSO do the support but the reason you have people around to begin with is for when the technical systems break in weird and interesting ways. Similar to how planes have pilots even in the age of autopilots (the humans are, of course, still fallible).
Another argument might be: well, LLMs will write better software. My counter is that there weren't really excuses for not writing better software before but we still got buggy code with poor documentation. If anything, LLMs will make it easier to crank out barely maintainable and supportable code at even faster rates.
I was once watching an old school survivalist talk about Native American/First Peoples legends.
These legends often had a bumbling main character who would usually cause some kind of problem b/c he forgot to do the key thing required for survival. For example, he would pick wet wood that wouldn't work for making a fire etc while his smart friend would pick the dry wood or the wood with lots of oil in it. Let's say bumbling dude is name "Chintatook" (made up name).
Now, when someone is starting to do the wrong thing or not think things through, you can say "Hey, don't be a Chintatook!" and everyone knows what you are talking about.
I switched to using Pinboard [0] for all bookmarking and never looked back.
The real unlocks were:
- using the bookmarklet that pops open a small browser window with the page title, suggested tags
- doing the same on my iphone
- have a couple in browser bookmarks that point to the tags for important things
It's so good I even used it to track all of my LinkedIn connections tagged by location, job function etc (inspired by Derek Sivers post on having a database [1])
Another, arguably even more powerful, alternative is Rhino + Grasshopper. Grasshopper is often used for generative designs, but can include arbitrary Python nodes and can even be used for "parametrically" designed functional parts.
Grasshopper can also output gcode directly [1], enabling pretty wild things like [2].
What I found most interesting about him was around the time Trump was running for president the first time, Adams was one of the first people to point out that Trump was, to use Adams' terms, a "master persuader". No one else at the time seemed to be talking about this and it was fascinating to see a humorist have this take/insight.
reply