logs archiveIRC Archive / Freenode / #emacs / 2015 / October / 10 / 1
Fuco
corpus: emacs is not an object, treat it with respect
wgreenhouse
colorless green ideas do *not* sleep furiously
Fuco
what was the original argument?
I used to be pretty hardcore syntax ueber alles person :D
then I read some of chomsky and figured it is batsh*t insane
thomasfuston
corpus: emacs is absolute there is no such thing as "sexier emacs" or "better emacs" :P
ryouma
insipid environmental concepts do, however, sometimes languish to the great annoyance of their proponents
nzst
,unvers-creed
fsbot
I think unvers-creed is This is my emacs. There are many like it, but this one is mine. My emacs is my best friend. It is my life. I must master it as I master my life. My emacs, without me, is useless. Without my emacs, I am useless
wgreenhouse
Fuco: I said that Chomsky was a bad although heavily accredited linguist, but a decent unaccredited philosopher
Fuco: and that generative grammar actively harmed the field, by driving it away from being science (i.e. falsifiable hypotheses)
corpus
so no one will take a snapshot of their emacs? :(
         

nzst
http://emacsthemes.com/
mrmyers
Aside from any hypothesis about what relation a generative grammar for a language has to the mental processes that cause that grammar to be approximately accurate, a grammar is a useful tool for describing compactly the kind of transformations that frequently occur. The mathematical methods associated with them, divorced from any presuppositions as to what conclusions we are to draw from them, are a useful
addition to the field.
thomasfuston
corpus: what do you expect to see?
Fuco
wgreenhouse: that's pretty hard to judge, it harmed the field but also liberated it in a way... someone would come up with that sooner or later
wgreenhouse: just like descartes came up with eather waves which newton admired and then ripped to shreds
corpus
thomasfuston: screenshots of peoples emacs
didi
corpus: Like rms?
thomasfuston
corpus: pretty sure is most time just a random emacs theme + modline :)
Fuco
corpus: http://i.imgur.com/a0pV8mP.png
wgreenhouse
mrmyers: sure, hence its usefulness to non-linguistics, like writing compilers or describing properties of formal logics
didi
Let's increase our civilization level.
wgreenhouse
mrmyers: it's also proved noatbly bad as a basis for machine translation, though, so it's not a way to understsnd humans, either in their development or their customary usage
*notably, understanf
*understand
it3ration
i just stumbled upon this: http://emacs-fu.blogspot.com/2011/08/customizing-mode-line.html
horray
wgreenhouse
gah must stop phone-sshing
Fuco
With an average sentence having some million derivations pure syntax approach to interpretation can become pretty bothersome
I believe logical analysis is where the future lies
thomasfuston
Fuco
with hither monadic logics or intensional logics etc
higher*
didi
And statistics is where the present lies.
ryouma
why would exotic logics be helpful for natural languages?
corpus
thomasfuston: that one looks nice
didi
Moar data! Laws is dead.
         

Fuco
ryouma: because they model it pretty damn well?
ryouma
i had no idea
thomasfuston
corpus: but as you saw on fuco its basicly text with modline, and different themes, dont expect to see to much fancy stuff
Fuco
ryouma: some logics were designed to model natural language, unlike predicate logic... for example transparent intensional logic
wgreenhouse
ryouma: I think cog sci is a very mixed bag. there's a lot of scope for "cuteness"
Fuco
ryouma: they are designed to capture language, intensions and extensions (sinn und bedeutung in Fregian terminology)
and what you want to arive is meaning not a parse tree
ryouma
wgreenhouse: i have been to a few conferences. they seem to me to be "let's draw circles and arrows, claim that this models human thinking, or thinking of some kind at least, and not test it, becasuet hat is how cogsci works"
lacrymology
I've been using an overgrown emacs.d from a friend, but I decided to start over.. but tbh I've no clue where to start. I mean I make an okay use of emacs, but I don't know how to get started.. what package manager to install, how to configure it, how to get packages
is there some kind of "intermediate tutorial"?
ubuntu__
Hey guys! I'm looking for a bit of help to install flycheck-rust. I really can not get it to work
benaiah
lacrymology: not really, at the intermediate level people's wants diverge so much you can't really do a generic tutorial
ryouma
defaults
better-defaults
benaiah
there are usually guides on each specific thing you want, though
ubuntu__
I have flycheck working. When I edit my init.el file, flycheck shows advice
benaiah
lacrymology: that said what you mentioned is pretty basic and any recentish emacs guide should touch on most of it
wgreenhouse
ryouma: right, by cuteness I meant mechanism without empirical basis or alternately empirical basis with many asterisks but really silly sounding mechanism
ubuntu__
however, when I try to edit a .rs file, it wont' show anything
I am using flycheck-rust-setup
and it doesn't seem to do anything
wgreenhouse
ryouma: and relatively few people worried about bridging those
ryouma
wgreenhouse: ic
benaiah
ubuntu__: do you have the rust linter installed?
mrmyers
As a tool, they are agnostic to their usage. Regular languages and context free grammars both fail to capture crucial aspects of programming languages, that doesn't invalidate their usage within restricted contexts.
benaiah
ubuntu__: is it on the $PATH that's being used by emacs?
those are the two most likely culprits of flycheck breakage
thomasfuston
lacrymology: this could be a good starting point, http://y.tsutsumi.io/emacs-from-scratch-part-1-extending-emacs-basics.html
if you dont know where to start at all
other than that it is just depending on what you need, in your config add it to it etc.
ryouma
wgreenhouse: i guess the issue, presuming testing, is the complexity of the model. is that your objection to chomsky also?
one issue*
benaiah
ryouma: "presuming testing" I thought wgreenhouse's whole point is that you can't presume testing, or if you can the model tends to be overfitted to the data
lacrymology
thomasfuston: thanks
wgreenhouse
ryouma: chonsky is on the mechanism-devoid-of-empirivs side. some cog stuff is on that side of nonscience, some is on the asterisks-without-proposed-mechanism side (e.g. most fMRI experiments)
*emoirics
empirics
ryouma
benaiah: if you mean by overfitting "too many alternate explanations" then that is what i was saying
wgreenhouse
goddamn
also medicine classically has that problem
$drug is correlated with something but we have no guess why
ship it!
mrmyers
The fact that a particular model approximately works is a useful scientific observation, that any complete theory of that domain has to eventually explain.
benaiah
fu*k this heisenbug
I've read the entire diff like six times and it still makes no damn sense
ryouma
/if/ it approximately works and if people do not get so drawn into it that they lose track of what they are doing
wgreenhouse
imo you need both reproducible observations and some proposed, parsimonious, testable mechanism to call it science
much brain stuff lacks one of those
tali713
wgreenhouse: well, hard science vs soft science.
ryouma
or perpetuate zombie (non-)science or corrupt and bias everything they can get their hands on etc.
Malabarb`
Anyone know of any function to brighten/darken colors
benaiah
"soft science" is just a nice way to say "not science but please take us seriously" most of the time
Malabarb`
Like, something that takes a color string like 33aa22, and returns a color string 10% brighter
wgreenhouse
tali713: econ at this point is better on covering both empirics and mechanism than much neuro/cog
mrmyers
"hard science" is just an easy way to say "easy science"
s/easy way/nice way/
benaiah
unscientific approaches can be useful, but it's dumb and muddies the terminology to call them "soft science" instead
wgreenhouse
benaiah: nobody does except people making a calumny against $unwanted_field
tali713
wgreenhouse: well, depends on the question being asked. "Why do people do normal people thing?" who the fu*k knows. "how do people with severed right left lobes perceive the world" bit closer.
wgreenhouse
benaiah: it's a purely political term
with the implication that soft => stab it and drive it out of the academy
« prev 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 next »