3 ms·

I think one issue, underneath this, is that threats are anti-inductive. Whatever you learn eg., about hacking-X today, will be obsolete tomorrow because X has

I think one issue, underneath this, is that threats are anti-inductive. Whatever you learn eg., about hacking-X today, will be obsolete tomorrow because X has been patched. You can't do induction (here, I supposed called Bayesianism) nor can you deduce anything since the problem space is intractable. (The system underneath this, of course, is the human animal). There's a new epistemic (, ontological) framework lurking here that has to do with complex systems, anti-induction, the failure of the scientific method (eg., methodological reductionism), and the limits of human knowledge. "Antifragility", complexity theory, chaos etc. get you 70% of the way there. But as this article demonstrates, all that talk still doesn't give us a clear language for, eg., infosec. In my view we're 95% of the way along the S-curve called "scientific knowledge of the reducible world", and nearing an end of the success of the scientific method. Since all open domains now have the characteristics above, and simply, there is no method which produces knowledge in the face of them.

_glass 4y ago
This is actually why we have in the social sciences the concept of abduction. Finding out something that defies the rules, where new rules have to be created. In deduction, you have a type, and infer the tokens. For induction, you have tokes and infer a type. But for abduction, you have a radically new token, where the type needs to be created. I am pretty sure that the techy infosec community finds out that it's closer to the social sciences.
- mjburgess 4y ago
  Well all explanation, imv, is abductive. But in scientific abduction, the "types" are properties which obtain invariantly across "the space in which they're located". Eg., water heats the same regardless of where it is, assuming identical other conditions. Whereas if we think of human society (, psychology, threats, etc.) as a space, then properties arent invariant (eg., there's path dependence). Hence, I dont think there is such a thing as a "social science". I think we'd be better of ditching the term science entirely, and scrapping much of the scientific method here. We need a new term, we arent building explanations, but rather ways of coping not having them. A kind of "empirical-speculative coping".
  - _glass 4y ago
    I might be misunderstanding you, exactly because I am a social scientist, and we have different kind of vocabulary. Maybe a definition by group instead of an understanding of reality that differs. Your approach looks to me more like what the social sciences critique as a paradigm, the positivist approach, where everything has to be law-like, "nomothetic". From there a stack of ontological/epistemological revelations are following, (the postmodern is one of them). But I think our backgrounds are very different, so that would need a few cups of coffee or tea :)
    - mjburgess 4y ago
      Sure, imv, "science" as a term is "positivist" project. There is no such thing as a "scientific critique" of a "paradigm". There can be a philosophical critique of a paradigm. The very phrase, "a social science critique of a paradigm" shows that "social science" isn't a science -- which is no bad thing. My point is there can't be a social science. Since "society" is an object the scientific method fails on.
      - _glass 4y ago
        I am German, maybe this makes the difference. We define science as an object of study, with the basic distinction between natural sciences, which can be explained, and the human sciences that have to be understood. The social sciences enhance in the way also the natural sciences, as everything is a social construct, even physical terms, what is a desert, what is climate, or even what is a man or a woman. Philosophy once included natural philosophy where we would now locate the scientific method. But hermeneutics is equally a scientific method. The reduction of science to the explanation of "certain" physical phenomena is exactly why we then fail to understand the world.
        mjburgess 4y ago
        Sure, https://en.wikipedia.org/wiki/Wissenschaft#Compared_to_the_term_%22science%22 https://en.wikipedia.org/wiki/Wissenschaft#Compared_to_the_t... "hermeneutics" isnt a scientific method in the english sense: https://plato.stanford.edu/entries/scientific-method/ https://plato.stanford.edu/entries/scientific-method/
AstixAndBelix 4y ago
Being a hacker is just like being a scientist. You find yourself in front of an unknown device or mechanism, you study its inputs and outputs and try to take it apart as much as possible. Then you formulate a model and see how much it is able to predict, until you reach an accurate enough one that allows you to deduce the real inner workings of the machine. The fact that the system can get patched does not change anything, you just start again.

_glass 4y ago

This is actually why we have in the social sciences the concept of abduction. Finding out something that defies the rules, where new rules have to be created. In deduction, you have a type, and infer the tokens. For induction, you have tokes and infer a type. But for abduction, you have a radically new token, where the type needs to be created. I am pretty sure that the techy infosec community finds out that it's closer to the social sciences.

mjburgess 4y ago
Well all explanation, imv, is abductive. But in scientific abduction, the "types" are properties which obtain invariantly across "the space in which they're located". Eg., water heats the same regardless of where it is, assuming identical other conditions. Whereas if we think of human society (, psychology, threats, etc.) as a space, then properties arent invariant (eg., there's path dependence). Hence, I dont think there is such a thing as a "social science". I think we'd be better of ditching the term science entirely, and scrapping much of the scientific method here. We need a new term, we arent building explanations, but rather ways of coping not having them. A kind of "empirical-speculative coping".
- _glass 4y ago
  I might be misunderstanding you, exactly because I am a social scientist, and we have different kind of vocabulary. Maybe a definition by group instead of an understanding of reality that differs. Your approach looks to me more like what the social sciences critique as a paradigm, the positivist approach, where everything has to be law-like, "nomothetic". From there a stack of ontological/epistemological revelations are following, (the postmodern is one of them). But I think our backgrounds are very different, so that would need a few cups of coffee or tea :)
  - mjburgess 4y ago
    Sure, imv, "science" as a term is "positivist" project. There is no such thing as a "scientific critique" of a "paradigm". There can be a philosophical critique of a paradigm. The very phrase, "a social science critique of a paradigm" shows that "social science" isn't a science -- which is no bad thing. My point is there can't be a social science. Since "society" is an object the scientific method fails on.
    - _glass 4y ago
      I am German, maybe this makes the difference. We define science as an object of study, with the basic distinction between natural sciences, which can be explained, and the human sciences that have to be understood. The social sciences enhance in the way also the natural sciences, as everything is a social construct, even physical terms, what is a desert, what is climate, or even what is a man or a woman. Philosophy once included natural philosophy where we would now locate the scientific method. But hermeneutics is equally a scientific method. The reduction of science to the explanation of "certain" physical phenomena is exactly why we then fail to understand the world.
      - mjburgess 4y ago
        Sure, https://en.wikipedia.org/wiki/Wissenschaft#Compared_to_the_term_%22science%22 https://en.wikipedia.org/wiki/Wissenschaft#Compared_to_the_t... "hermeneutics" isnt a scientific method in the english sense: https://plato.stanford.edu/entries/scientific-method/ https://plato.stanford.edu/entries/scientific-method/

AstixAndBelix 4y ago

Being a hacker is just like being a scientist. You find yourself in front of an unknown device or mechanism, you study its inputs and outputs and try to take it apart as much as possible. Then you formulate a model and see how much it is able to predict, until you reach an accurate enough one that allows you to deduce the real inner workings of the machine. The fact that the system can get patched does not change anything, you just start again.