One of the core problems we have in software engineering is the longstanding philosophical problem around creation of cohesive, consistent, objective mental models of inherently subjective concepts like identifying a person, place, etc. Look at the endless lists of falsehoods programmers (tend to) believe about any topic.
You’re right that LLMs specifically have no guarantees about accuracy nor veracity of the text they generate but I posit that that’s the same with people, especially when filtered through the socialization process. The difference is in the kind of errors machines make compared to ones that humans make.
It’s frustrating we’re using anthropomorphic concepts like hallucinations when describing LLM behaviors when the fundamental units of computation and thus failures of computation are so different at every level.
> The difference is in the kind of errors machines make compared to ones that humans make.
There's another difference, and that is that other humans can learn and study that mental model (which is why "readable code" is a goal — the code is a physical manifestation of the model that you, the programmer, has to learn), and then the model can be tweaked and taught back to the original programmer, who can then think of that tweak in the future. Programming is inherently (in most cases) a collaborative art, because you're working with people to collectively develop a mental model and refine it, smoothing it down until (as Christopher Alexander said) there are no misfits between the model and the domain.
This is what amorality means to me in the context of socioeconomics. It operates in an area of reduced dimensionality to economic value because no other value can be agreed upon in trade between cultures. It doesn’t care if a piece of art, nature or human invention is genuinely novel, rare, irreplaceable, invaluable, etc. unless it can be converted into materializable economic value that is itself subjective and present oriented so that we can plan for our future selves about resources as a proxy.
Both ideas can be true. It’s not on their radar because despite their popularity in consumer space they can’t find a business purpose that aligns with their self interests that require such user information. If I’m running a free podcast, in contrast, I might be happy anyone’s even bothering to visit and listen to what I have to say compared to who they are and whether I can assign a monetary value to their attention because spending money on something without clear, intentional, measurable ROI is anathema to our predominant modus operandi in business
I think it’s unfair to say that it’s “lazy.” Neo Marxists understand the way those factors affect impact economics but because they’re very difficult to quantify in a heavily quantitative focused academic environment you’ll see less focus upon it, even when analyzing standard fare free market capitalism. I consider it a double standard to put burdens of analysis more upon one ideology than another. Sociology and religion are already highly qualitative disciplines by nature of the limits of science and our known physical reality (eg. we can’t time travel let alone reliably) and the kind of advantages systems like feudalism, monarchism, mercantilism, etc. work with authoritarian centralized systems is consistent across many societies including why some cultures tend toward syncretism and why others reject certain tenets and customs. The Nordics weren’t Christian, after all, and they absorbed it much more than American indigenous while Black Americans have very different relationships to Christianity than African blacks such as in Somalia and Ethiopia - this isn’t the wheelhouse of economists typically in academics but various humanities departments unrelated to business. Those influences and trends usually get labeled under the standard generic realities of imperialism given so much orchestrated power influences societies en masse while most bottom up movements against these structures tend to come from humanities focused areas away from economic interests like the arts, but this is why the Soviet Union and even China suppressed these freedoms because of the tendency to cause discord and dissent in a precarious society. As such, most leftist literature, especially outside academia (already an institution that must exist within the confines of a funding society), sounds predominantly like they’re criticizing basically every dominant human social construct, which is where the ideological position is cornered. This isn’t to say that I agree with this kind of discourse either because it doesn’t convince people beholden to these dominant constructs.
If they did, why are they constantly surprised, science gives you prediction power- scientific analysis, should make you the one who surprises others. "Eclipse.. now!" is science, not "Why is it suddenly dark! Must be xyzs fault!"
Why do people focus on things that have never worked? Studying Marxism is analogous to studying astrology. It's the most failed economic system ever tried.
When you study Marxism you're studying an economic point of view, not a form of government. IIRC most of his work is observational and there's just not much to disagree with once you break essays down to a bullet-pointed outline.
The only thing you should be upset about with Marx is his prose.
Its more aching to theology, as the economic model is very simplified, assuming all constraints can be easily overcome - once you sparsed out the oversimplified sociological model. They have no concept for a limitation of resources, no concept for organizational complexity, no concept for investment pooling- beyond "the party of benevolent revolutionaries" controls all. Any toddler could come up with something more coherent..
The Labor Theory of Value is a basic component of Marxism and is demonstrably false.
The other tenet is looking at everything through the lens of class struggle. The modern incarnation of that is interpreting all of history in terms of race. Social mobility in free countries like the US shows that the class struggle theory is woefully inadequate.
Time would be better spent studying free markets, the law of supply and demand, etc.
The double standard I see across all political and cultural lines is people that demand empathy for them while they demonstrate none for an arbitrary outgroup or one specific to their personal lived experience. This is basically using emotion to drive thoughts rather than emotions to inform thoughts. I have doubts this will go away anytime soon given it takes an incredible amount of effort for people to critically examine themselves. I liken it to debugging and reverse engineering your brain and nervous system.
This seems fairly consistent with deep, accomplished experts in any field or craft - their competencies in one area don't necessarily translate well into validity anywhere else laterally or even vertically. This seems extremely obvious even for typically "smart" people such as doctors, lawyers, engineers because among many folks I know scamming these white collar professionals out of money by feeding into their egotism is basically how they make their living. While I don't think we should fence people into professional castes or anything like that but in the modern age of AI and charisma-based validity / authority healthy skepticism seems like a requisite to not be suckered into modern infomercial quackery.
> This seems fairly consistent with deep, accomplished experts in any field or craft - their competencies in one area don't necessarily translate well into validity anywhere else laterally or even vertically.
With this person, his core area of pseudo-research has been the race/IQ generalizations. He has a long history of eugenics content and his Reddit posts engage in Nazi stuff.
He jumps into popular topics like autism statistics to ride those trends on Twitter and expand his reach. Once you follow him, you realize it’s a steady drip of the lightly disguised eugenics stuff that has been his core focus for a long time.
This whole process happening is exactly what happens in a quest in Cyberpunk 2077. There’s an e-mail chain where a gang tried to extort a corporation and gave up after being unable to reach a person.
I sincerely hope that the game doesn’t become prophetic in the manner Idiocracy has.
Switzerland has a slightly high suicide rate (not the best inverse metric of happiness but a correlation on unhappiness at least) for a country with such high standards of living, so if we look into suicide rates over time we also see a conundrum that over the past 20+ years the suicide rate is overall decreasing but has mostly flattened out. But from what I've observed anecdotally it still has problems like other developed countries with legacy industries declining (see: watchmakers and other artisanal crafts trades rather than mining) where boomers in the country are pretty miserable and that will probably be noticed in macro level statistics.
The irony of the fabled Seattle complaints of the “freeze” is it’s not unique to the region whatsoever and due to so many transplants in the past 15-ish years along with so many locals forcibly relocating out of the city people are more likely than ever to be interacting with those that moved as adults / other transplants.
For me as someone that grew up in the region the people, nature, and weather are sufficient enough for me. Having lived in several other large metro areas in the US I’ve pretty much felt like an alien species even though it’s not like I don’t feel welcome. In the PNW being weird and unconventional is kind of celebrated regardless of socioeconomic castes historically, but that’s certainly eroded as the problems of hyper growth have strained everyone.
That would be an HOA moreso than the features of a condo, although a condo tends to imply an HOA in the US. The irony of my experiences with an HOA is that its actions tended to suppress my property value rather than preserve or grow it.
I don't know how a condo building could operate without some kind of shared ownership structure for the common areas and shared infrastructure. What are the alternatives to an HOA in a condo building?
You’re right that LLMs specifically have no guarantees about accuracy nor veracity of the text they generate but I posit that that’s the same with people, especially when filtered through the socialization process. The difference is in the kind of errors machines make compared to ones that humans make.
It’s frustrating we’re using anthropomorphic concepts like hallucinations when describing LLM behaviors when the fundamental units of computation and thus failures of computation are so different at every level.
reply