Further proof that Comp Sci is nothing but fashionable nonsense
Howdy, Stranger!
It looks like you're new here. If you want to get involved, click one of these buttons!
It looks like you're new here. If you want to get involved, click one of these buttons!
Comments
this is about publishers being unscrupulous
it's like an onion wrapped in an enigma tied up in a box of pixie sticks or whatever i have no idea where this analogy is going anymore
much easier to deal with
fuck sarcasm on the internet
it's all over the fucking place and it's poisoning our discourse in the name of some idiots being smug jerks who want to feel superior by social posing using mockery
fuck mockery
want to fight about something, get DOWN to its level.
fuck "get on my level". your level sucks and you should feel bad because you are making an arbitrarily high level just to show off how oh-so-socially-adept you are. fuck your level and everything your level stands for.
no one else go to that guy's level. pelt him with rotten tomatoes or something. i'll provide the tomatoes. and popcorn.
Fashionable Nonsense is the title of a book by Alan Sokal. Alan Sokal is a professor of mathematics publically best known for the "Sokal Affair", in which he submitted a deliberately nonsensical article to a (non-peer reviewed) journal of critical theory called Social Text. Fashionable Nonsense claims, roughly, that postmodernism in academia and fields that get inundated with it (such as critical theory) are nonsense, and that's why Sokal can submit gibberish to them and get accepted.
The news article linked concerns a similar situation to the Sokal affair. In particular, a researcher mechanically generated a number of papers and submitted them to (I think, again, non-peer-reviewed) conferences where they were accepted. The difference of course is that these papers concerned computer science, rather than critical theory.
The intended joke here, I believe, is an assertion that by Sokal's reasoning this means that computer science is nonsense, and Sokal probably does not want to think that computer science is nonsense. Furthermore, it is likely that Myrmidon does not think that computer science is nonsense, or probably even that critical theory is not nonsense, and therefore asserts that the entire Sokalian methodology is invalid.
It is possibly worth mentioning that similar Sokalian attempts have been successful. For example, back in October of 2013, a biologist Bohannon submitted a nonsense paper to a number of open access journals, in a criticism of open access policies/journals as a whole. As you can see from the regurgitation I have linked, Sokal the innovator is well known within the communities of people who actually care about journals.
It is also possibly notable that responses such as questioning the particular journals' integrity have been common in response to the original Sokal as well as imitators, as in this thread. Of course, IEEE and Springer are very much not minor publishers; and neither is Duke University, the publishers of Social Text.
Finally, remember that we have no shortage of pins. Don't be afraid to use as many pins as you need or want to hold the skin and fascia over frog's body cavity open. It is likely that without the body cavity being fully exposed, you will be unable to effectively investigate the animal's interior.
Although I'd find that kind of claim to sort of be missing the point.
we don't consider him an authority on psychology, if that's what you're asking
his dream analyses are still read as an alternative method for decoding imagery, but direct Freudian psychoanalyses of texts are considered dated
there is a field of theory based on psychoanalysis, but that's more of a post-structuralist thing and doesn't have a lot to do with Freud himself (and is also very controversial even within theory)
that is to say, a lot of 'Freudian' writing is by people who strongly disagree with him, but they're still Freudian because they're using his terminology and responding to his ideas
and i'm just sitting here like oh my god if you're going to fetishize some fucking european nutjob couldn't you go for someone cool like vygotsky or someone who generally iddn't just make shit up about wolf trees
what's the deal with wolf trees
anyway i don't know anything about critical theory
i do know some hilarious CS shit tho
i don't ge twhy you'd use him for non-psychology anyway, like what the fuck? i'm not going to talk about my field with fourier's descriptions of heat equations
it's a subject that tends to produce rather polarizing opinions, though
unless you want us to throw out everything that refers to the unconscious on the basis that Freud popularized the term, a founder of discursivity he will remain
dream interpretation is actually kind of interesting because it provides an alternative way of looking at certain forms of common imagery, but very few people consider it anything more than that.
actual psychoanalysis (as practiced by Freud) is dated and is not really practiced anymore, and the various derived approaches are, as i say, controversial.
i am not, i should add, a Freudian, nor am i particularly versed in psychoanalysis, so 'it's controversial' is the closest i'm going to come to defending it
i can try to be charitable for once. i like jung. not as a psychologist, i can't recall anything good he did there, unlike freud. freud i can say was a pioneer of developmental psychology, or something. but jung's crazy ideas about psychology don't affect me in the same way. and archetypes are nice for what they are. probably his aren't /so/ useful in analysis of literature but it's some kind of start? maybe it's analogous for freud?? mods??? just ugh trauma
so does anyone want to hear about how kolmogorov complexity solves hume,
but i guess they don't lend themselves so well to poststructuralist approaches? and theory still hasn't quite gotten over its collective hard-on for poststructuralism
i dunno what kolmogorov complexity is but go for it
(teehee)
kolmogorov complexity: what if we characterized generic "complexity" of objects by seeing what the length of the shortest algorithm that can describe them is? oops it's impossible to get any actual numbers for that because [basic halting problem reduction]. well, let's do it anyway
now let's apply this to some weird shit. let's say induction. induction is where bla bla basic logic class here. so we have some data and we want to figure out what hypothesis is best. what if:
specifically we use bayesian inference to go from some data to a hypothesis. now the usual bayesian problem is that to pick a prior you basically just make some shit up. however we can get around that by using the 'solomonoff distribution' named after some smart dude named solomonoff. the solomonoff distribution to oversimplify even more lol treats parameters with low kolmogorov complexity as more probable. so we use that as our initial prior and go crazy.
cite: li, vitányi 2008. good book but lol
The philosopher D. Hume (1711-1776) argued that true induction is impossible because we can reach conclusions only by using known data and methods. Therefore, the conclusion is logically already contained in the start configuration. Consequently, the only form of induction possible is deduction. Philosophers have tried to find a way out of this deterministic conundrum by appealing to probabilistic reasoning such as using Bayes's rule. One problem with this is where the prior probability one uses has to come from. Unsatisfactory solutions have been proposed by philosophers such as R. Carnap (1891-1970) and K.R. Popper.
However, R.J. Solomonoff's inductive method[...] may give a rigorous and satisfactory solution to this old problem in philosophy.
Essentially, combining the ideas of Epicurus, Ockham, Bayes, and modern computability theory, Solomonoff has successfully invented a perfect theory of induction. It incorporates Epicurus's multiple explanations idea, since no hypothesis that is still consistent with the data will be eliminated. It incorporates Ockham's simplest explanation idea, since the hypotheses with low Kolmogorov complexity are more probable. The inductive reasoning is performed by means of the mathematically sound rule of Bayes.
did i just blow your mind?? <-- not part of the book
advanced readers may note a few problems such as
so
does that inextricability thing mean that an analysis is going to be bound to its author as well as to the thing being analyzed
btw what is critical theory i honestly have no idea. i kind of don't care just because it doesn't seem like it'll be relevant to me ever but like what's even going on there. i literally know more about water bears than about lacan. well ok i know he did that matheme thing ha ha ufukcing ha
It doesn't really have any shared goal or method, other than a shared body of reference.
Also Marxism, usually.
i mean i guess it could still be an interesting and potentially useful exercise, but it's still nothing to do with Hume or the fundamental problem of induction
i mean unless you envision the sceptic as some kind of philosophical troll who thinks all science should be thrown out the window, but that's not Hume at all
it's like trying to refute Descartes' demon argument using string theory
it's completely the wrong tool for the task at hand
this is stating the obvious, maybe
It winds up being computationally incapable of real Bayes, but more flexible than a computer when dealing with abstract matters because that's when the spaghetti-at-wall approach to association actually works.