/ˈbɑːltəkʊteɪ/. Knows some chemistry and piping stuff. TeXmacs user.

Website: reboil.com

Mastodon: [email protected]

  • 1 Post
  • 49 Comments
Joined 3 years ago
cake
Cake day: June 12th, 2023

help-circle



  • efficiency is just high functioning laziness

    Reminds me of discussions about how economists working for capitalist oligarchs, define efficiency in terms of value extracted per dollar invested without taking into amount negative externalities like environmental destruction or worker well-being. Such economists are “lazy” about those last two points which, for billionaires that hire them to get their next corporate merger approved, is a feature, not a bug.

    Your comment reminds me that every efficiency is tightly coupled to a specific goal that benefits a particular group of people that may not necessarily include myself.

    Chapter 40 of The Ministry for the Future by Kim Stanley Robinson

    Jevons Paradox proposes that increases in efficiency in the use of a resource lead to an overall increase in the use of that resource, not a decrease. William Stanley Jevons, writing in 1865, was referring to the history of the use of coal; once the Watt engine was introduced, which greatly increased the efficiency of coal burning as energy creation, the use of coal grew far beyond the initial reduction in the amount needed for the activity that existed before the time of the improvement.

    The rebound effect of this paradox can be mitigated only by adding other factors to the uptake of the more efficient method, such as requirements for reinvestment, taxes, and regulations. So they say in economics texts.

    The paradox is visible in the history of technological improvements of all kinds. Better car miles per gallon, more miles driven. Faster computer times, more time spent on computers. And so on ad infinitum. At this point it is naïve to expect that technological improvements alone will slow the impacts of growth and reduce the burden on the biosphere. And yet many still exhibit this naiveté.

    Associated with this lacuna in current thought, perhaps a generalization of its particular focus, is the assumption that efficiency is always good. Of course efficiency as a measure has been constructed to describe outcomes considered in advance to be good, so it’s almost a tautology, but the two can still be destranded, as they are not quite the same. Examination of the historical record, and simple exercises in reductio ad absurdum like Jonathan Swift’s “A Modest Proposal”, should make it obvious that efficiency can become a bad thing for humans. Jevons Paradox applies here too, but economics has normally not been flexible enough to take on this obvious truth, and it is very common to see writing in economics refer to efficiency as good by definition, and inefficient as simply a synonym for bad or poorly done. But the evidence shows that there is good efficiency and bad efficiency, good inefficiency and bad inefficiency. Examples of all four can easily be provided, though here we leave this as an exercise for the reader, with just these sample pointers to stimulate reflection: preventative health care saves enormous amounts in medical costs later, and is a good efficiency. Eating your extra children (this is Swift’s character’s “modest proposal”) would be a bad efficiency. Any harm to people for profit is likewise bad, no matter how efficient. Using an over-sized vehicle to get from point A to point B is a bad inefficiency, and there are many more like it; but oxbows in a river, defining a large flood plain, is a good inefficiency. On and on it goes like this; all four categories need further consideration if the analysis of the larger situation is to be helpful.

    The orienting principle that could guide all such thinking is often left out, but surely it should be included and made explicit: we should be doing everything needed to avoid a mass extinction event. This suggests a general operating principle similar to the Leopoldian land ethic, often summarized as “what’s good is what’s good for the land.”[cmt 2] In our current situation, the phrase can be usefully reworded as “what’s good is what’s good for the biosphere.” In light of that principle, many efficiencies are quickly seen to be profoundly destructive, and many inefficiencies can now be understood as unintentionally salvational. Robustness and resilience are in general inefficient; but they are robust, they are resilient. And we need that by design.

    The whole field and discipline of economics, by which we plan and justify what we do as a society, is simply riddled with absences, contradictions, logical flaws, and most important of all, false axioms and false goals. We must fix that if we can. It would require going deep and restructuring that entire field of thought. If economics is a method for optimizing various objective functions subject to constraints, then the focus of change would need to look again at those “objective functions”. Not profit, but biosphere health, should be the function solved for; and this would change many things. It means moving the inquiry from economics to political economy, but that would be the necessary step to get the economics right. Why do we do things? What do we want? What would be fair? How can we best arrange our lives together on this planet?

    Our current economics has not yet answered any of these questions. But why should it? Do you ask your calculator what to do with your life? No. You have to figure that out for yourself.



  • Where’s the Terminator film that takes the POV of a billionaire’s child who learns about the securing of natural resource extraction sites and trade routes against piracy (actually perfectly normal people fighting for survival against the 0.1%) through the use of automated drones? It would be interesting to see how ultra wealthy schools screen for intelligent children for the dual purpose of identifying candidate leaders to keep the idiot dynastic majority-minority shareholders (they hold majority shares but make up a minority of the population) in line as well as identifying potential rebel sympathizers that almost always arise among potential leaders. I feel like I’m echoing yet another 2010 young adult dystopian plot…







    • The Terraformers (2023) by Annalee Newitz. In lieu of a single simple reset, there is a continuous bargaining between capitalistic rentiers and enslaved residents who fight over hundreds of years throughout the final stages of a planet’s terraforming. Homelessness, mass transit, wealth inequality, and racism all are running themes.
    • Hieroglyph: Stories and Visions for a Better Future (2018) by Ed Finn (ed.) and Kathryn Cramer (ed.) is anthology of several science fiction authors who submitted stories with the purpose of specifically counteracting the trend at the time of (post-)apocalyptic stories anticipating a bleak future for humanity. One particularly aspirational story that persists in my memory is Girl in Wave : Wave in Girl by Kathleen Ann Goonan; it is a story about the sociopolitical impact of universal literacy, achieved thanks to the release of an inexpensive medicine that enhances human cognitive abilities, allowing anyone to overcome learning disabilities such as dyslexia. Other stories in the anthology are optimistic for other reasons, but this story comes to mind when I read your post.




  • In a separate study of 20 LLMs, Omar found that LLMs are more prone to hallucinate and elaborate on misinformation when the text they’re processing looks professionally medical — formatted like a hospital discharge note or clinical paper — than when it comes from social-media posts (M. Omar et al. Lancet Digit. Health 8, 100949; 2026). “When the text looks professional and written as a doctor writes, there’s an increase in the hallucination rates,” says Omar.

    You can just make an Overleaf account (or install GNU TeXmacs) and start outputting academic-like papers for fun and profit. I would have thought LLM developers would have at least highlighted PageRank-like citation metadata as very important when training on academic publications; papers with no citations clearly aren’t reputable.