egg|nomz|egg changed the topic of #kspacademia to: https://gist.github.com/pdn4kd/164b9b85435d87afbec0c3a7e69d3e6d | Dogs are cats. Spiders are cat interferometers. | Космизм сегодня! | Document well, for tomorrow you may get mauled by a ネコバス. | <UmbralRaptor> egg|nomz|egg: generally if your eyes are dewing over, that's not the weather. | <ferram4> I shall beat my problems to death with an engineer.
<kmath>
<diodelass> Something about the version of LibreOffice installed on my toughbook is extremely fucked up to the point of being b… https://t.co/RapgdTbjHT
<UmbralRaptop>
Should I attempt to corrupt someone to the LaTeX way?
<kmath>
<NicolaSpaldin> An alternative to watching Switzerland play Serbia in the World Cup this evening: Orchestra Accento Musical Zürich… https://t.co/YG84WCHEOl
<kmath>
<bofh453> THIS IS SO *PAINFULLY* TRUE AND AUGH: ⏎ ⏎ <@eggleroy> physicists like reinterpret_casting things in coordinates ⏎ <… https://t.co/uwwvjFa8wi
<bofh>
egg|work|egg: I mean it's a depressingly good way of putting it. :P
* awang_
🔪 fax2tiff
<awang_>
WHAT DO YOU MEAN I HAVE AN EMPTY INPUT FILE
<awang_>
IT'S 200KB
<UmbralRaptop>
ACTION suggests additional 🔪 if a 🖷
<kmath>
<vancura> While we’re here: a friendly reminder that you can write good typography in your source code. The rule of thumb: if… https://t.co/ssUr13appj
<egg|work|egg>
!u %﹪％٪
<Qboid>
U+0025 PERCENT SIGN (%)
<Qboid>
U+FE6A SMALL PERCENT SIGN (﹪)
<Qboid>
U+FF05 FULLWIDTH PERCENT SIGN (％)
<Qboid>
U+066A ARABIC PERCENT SIGN (٪)
egg|phone|egg has joined #kspacademia
egg|cell|egg has joined #kspacademia
egg|phone|egg has quit [Read error: Connection reset by peer]
egg|mobile|egg has quit [Read error: Connection reset by peer]
egg|phone|egg has joined #kspacademia
egg|cell|egg has quit [Ping timeout: 202 seconds]
tawny has joined #kspacademia
APlayer has quit [Ping timeout: 182 seconds]
tawny has quit [Ping timeout: 198 seconds]
egg|cell|egg has joined #kspacademia
egg|phone|egg has quit [Read error: -0x1: UNKNOWN ERROR CODE (0001)]
<egg|zzz|egg>
bofh: those dimensionful algebraic structures are fun
<egg|zzz|egg>
bofh: e.g. in vector spaces with dimensionful scalars, the concepts of angles and inner products become a bit more distinct
<bofh>
egg|zzz|egg: which ones?
<egg|zzz|egg>
bofh: the ones I use in Principia (or the ones in that Tao article which serves as an inspiration for the Principia libs)
<bofh>
so like, first off what do you mean by "angle" since the only sane way to define that for arbitrary inner product spaces is simply in terms of the inner product (or perhaps the arccos of the inner product)
<egg|zzz|egg>
bofh: yes, but what I mean is that the inner product doesn't give you the starting scalar
tawny has joined #kspacademia
<egg|zzz|egg>
otoh angles remain dimensionless in all dimensionalities
<egg|zzz|egg>
so you can't say that a dimensionful linear map preserves the inner product
<egg|zzz|egg>
but you can say it preserves angles
<egg|zzz|egg>
bofh: e.g. consider A = 1 m * identity, ⟨Av, Aw⟩ is incommensurable with ⟨v, w⟩, you are looking for ⟨Av, Aw⟩/(⟨Av, Av⟩, ⟨Aw, Aw⟩) = ⟨v, w⟩/(⟨v, v⟩, ⟨w, w⟩) instead
awang has quit [Ping timeout: 198 seconds]
awang has joined #kspacademia
<egg|zzz|egg>
bofh: i.e. you have a generalization of ℝ⁺*×O(3), not of O(3)
<bofh>
hrm.
<bofh>
I *think* you're right.
<egg|zzz|egg>
bofh: I mean within the same dimension, you have O(3) of course
<egg|zzz|egg>
but you do have "angle-preserving" as a property where "inner-product-preserving" is devoid of meaning
awang has quit [Ping timeout: 194 seconds]
awang has joined #kspacademia
<awang>
!slap people who send images embedded in PDFs instead of actual data
* Qboid
[LOG 06/22/2018 16:30:18]: Exception Was Recorded: Object reference not set to an instance of an object. people who send images embedded in PDFs instead of actual data was slapped so hard that is now read only.
<awang>
Um
<awang>
Oops?
<UmbralRaptop>
Hah
<UmbralRaptop>
awang: especially if you wanted a CSV?
<kmath>
<whitequark> . - * - . getting a CS degree, a flowchart . - * - . ⏎ do it ←(yes)— do you want to? —(no)→ don't do it
<egg|zzz|egg>
though I guess there's much more of a weird tendency for people to study CS for the purpose of doing software engineering, rather than, say, studying CS, and wtf
<egg|zzz|egg>
I am endlessly baffled by how many of my colleagues studied CS
<iximeow>
for a few years i used to hear "yeah i studied computer science" and thought "cool! can we talk about cool comp sci stuff? opinions on graph algorithms and compilers!! numerical computation!!"
<bofh>
egg|zzz|egg: I mean likewise
<iximeow>
eventually realized that's not what "studied computer science" means
<egg|zzz|egg>
iximeow: also CS curricula are horribly lacking in numerics
<bofh>
^
<bofh>
ASTOUNDINGLY so, in fact.
<iximeow>
yes!
<egg|zzz|egg>
just go for maths if you want that
<iximeow>
my single favorite CS homework question was numerics-related and only extra credit
<egg|zzz|egg>
(maybe go for *applied* but then by the time you get to specialize away from pure you will hopefully have been bitten into doing something less applied :-p)
<UmbralRaptop>
Math for people doing science and engineering is oddly lacking in numerics.
<iximeow>
(it was proving some function is not computable correctly under floating point arithmetic)
<iximeow>
well, more specifically, proving when the function is not computable under floating point arithmetic
<egg|zzz|egg>
bofh: also in linear algebra, which for some inane reason means a bunch of my colleagues are learning it again with silly terminology :-p
<bofh>
egg|zzz|egg: augh
<egg|zzz|egg>
bofh: I did try telling one of them about the universal property of the tensor product :-p
<egg|zzz|egg>
I'm confused by https://twitter.com/whitequark/status/1010244269968326657, my degree programme seemed fairly sensible from the "the point is to turn you into a mathematician" perspective (it would fit that description if you were somehow using that as a means to get an industry job, but wtf)
<kmath>
<whitequark> since people are taking this thread seriously... ⏎ it's true that completing a MSc in a modern university leaves you… https://t.co/X2Lsj6oKQW
<egg|zzz|egg>
!wpn whitequark
* Qboid
gives whitequark an indium tombstone
<bofh>
egg|zzz|egg: I mean I think math degrees are *slightly* different, both in that there really isn't an "industry job" there so they can't be commoditized *and* math is difficult enough that self-study is a lot harder than self-study of most other subjects (doable, but still).
<egg|zzz|egg>
bofh: well, banks would be the closest to a directly-related industry job
<egg|zzz|egg>
now of course if you're into abstract algebra rather than stats that's not really a closer fit than software engineering :-p
<bofh>
egg|zzz|egg: ehh, that's only for a very narrow subset of math.
<egg|zzz|egg>
yeah
<bofh>
(and I wouldn't say so much banks as finance)
<egg|zzz|egg>
yeah, but banks employ mathematicians
<bofh>
true, but banks are a special case of the financial industry :P
<egg|zzz|egg>
yeah, I tend to think of them because, well, I'm in Zurich
<bofh>
true.
<bofh>
also I like how I keep forgetting Zürich exists since all of my trips to Switzerland have been to the French part of it.
<egg|zzz|egg>
(if you're in pure mathematics, I guess you could make up derivatives on proofs of the abc conjecture)
<bofh>
(ROFL)
<bofh>
(who the heck would pay one for that???)
* egg|zzz|egg
shorts Θ±ellNF-Hodge theaters
<bofh>
rofl
<bofh>
are those the ones in good places or in bad places?
Qboid was kicked from #kspacademia by *status [You have been disconnected from the IRC server]
Qboid is now known as Qboid_
Qboid_ has joined #kspacademia
* UmbralRaptop
has start date questions if an MSc takes 5 years instead of 1-2,depending on the program. (Maybe 3 if there are delays)
<UmbralRaptop>
(Is a BSc also being counted?)
Qboid_ is now known as Qboid
<egg|zzz|egg>
UmbralRaptop: here I'd see the BSc-MSc as part of the same thing really
<egg|zzz|egg>
UmbralRaptop: see a good number of previous discussions about the separation being [BS MSc] [PhD] here and more like [BS] [MSc PhD] where you are
<egg|zzz|egg>
(and about the words "graduate student" and "graduation" being confusing for me)
<egg|zzz|egg>
whitequark: is it similar in RU?
<UmbralRaptop>
(Although the median time for getting a BSc might be be 5 years in the US now)
<egg|zzz|egg>
okay now that's odd
<egg|zzz|egg>
!wpn
* Qboid
gives egg|zzz|egg a salted mace
<bofh>
egg|zzz|egg: actually it's more like [BS] (MSc) [PhD]
<whitequark>
MSc is 5 years in RU
<whitequark>
approximately
<whitequark>
and most everyone gets MSc
<egg|zzz|egg>
whitequark: from having a BSc, or from getting out of school
<whitequark>
well, there used to just be BSc+MSc programs, they split them in BSc and MSc for compatibility with Europe
<whitequark>
from getting out of school
<egg|zzz|egg>
yeah so same as here
<whitequark>
in the program I was in you couldn't get a BSc alone I think
<egg|zzz|egg>
technically I finished my BS the year before my MSc :D
<whitequark>
yea something like that
<egg|zzz|egg>
and I assigned courses from the 5 years randomly between the two :-p
<UmbralRaptop>
hm
<egg|zzz|egg>
(the electives that is, obviously linalg I isn't in the MSc :-p)
<egg|zzz|egg>
!wpn whitequark
* Qboid
gives whitequark a maritime saker
<egg|zzz|egg>
!wpn коте
* Qboid
gives коте a stern heptode/ramipril hybrid
<UmbralRaptop>
egg|zzz|egg: were your courses assigned with an LCG, Mersenne twistor, or something else?
<bofh>
UmbralRaptop: xorshift64* in my case
<UmbralRaptop>
Hm
* UmbralRaptop
wonders if anyone has used MCMC.
<egg|zzz|egg>
whitequark: yeah here you can get a BSc alone (the usual european structure) but nobody does that
<egg|zzz|egg>
and there are very few hard dependencies on the order in which you do the bits of the BS and MSc, I think the only constraint is that you need to have the BS before you can start the MS thesis and you need to have passed the block of basic courses to enrol in the MSc
<egg|zzz|egg>
(which is how I ended up getting the BS and MS within one year of each other)
<egg|zzz|egg>
!u M
<Qboid>
U+004D LATIN CAPITAL LETTER M (M)
<egg|zzz|egg>
!meow whitequark
<Qboid>
(1,2): error CS1660: Cannot convert `lambda expression' to non-delegate type `object'
<Qboid>
System.Func`1[System.String]
<Qboid>
(1,2): error CS0103: The name `bofhtime' does not exist in the current context
* Qboid
meows at whitequark
<egg|zzz|egg>
um
<bofh>
lmao
<egg|zzz|egg>
bofh: so what are physical quantities anyway, they sound like the homogeneous elements of a graded algebra? but there's more structure to that
<egg|zzz|egg>
in particular within each dimension they are totally ordered and there is a definition of positive
<bofh>
define physical quantities. observables?
<egg|zzz|egg>
things with a dimension
<bofh>
conserved quantities?
<bofh>
hm
<egg|zzz|egg>
quantities in the SI brochure sense
<bofh>
AH.
<bofh>
that's a *good* question and let me remember what a graded algebra is again.
<egg|zzz|egg>
wait why did I say "9 mol" when I could have said "5 mol" (cc whitequark)
<bofh>
okay, wait, why does it being graded matter at all? I mean it's almost tautologically true they're homogeneous elements of some algebra, that much I'll give you
<egg|zzz|egg>
yes but you have the multiplicative groupoid!
<bofh>
(presumably to get a notion of an exterior algebra?)
<egg|zzz|egg>
bofh: Tao describes that as a tensor product of 1d spaces, with duals for the inverse
<bofh>
ahh, I see where you're coming from.
<egg|zzz|egg>
so 1 m / 5 s is an element of the lengths space tensored with an element of the dual of the times space
<bofh>
(also I think you can go as far as to say we're working in a Clifford algebra, I *think*)
<egg|zzz|egg>
yielding canonically an element of the speeds space
<egg|zzz|egg>
bofh: that's the thing, we're not using the whole algebra
<egg|zzz|egg>
because we purposefully want to ignore things like 1 m + 1 s
<bofh>
oh so you want to project out degrees of freedom/subspaces that we don't care about for a given quantity?
<egg|zzz|egg>
in particular because then you lose things such as "any quantity has a sign"
<awang>
bofh: I'm getting flashbacks and I'm not even done fighting with these PDFs :(
<egg|zzz|egg>
bofh: basically I wonder whether there's a nice name of description of whatever that is
<bofh>
egg|zzz|egg: I'm... not sure. I don't think I know enough abstract algebra to say for certain.
<bofh>
I see where you're coming from, at least.
<egg|zzz|egg>
"the homogeneous elements of the tensor algebra of a 7 1d tensor spaces" is OK, but it feels wrong to build up an algebra only to discard everything inhomogeneous
<egg|zzz|egg>
bofh: basically I'm looking at the fairly sound type system I've built up in principia and thinking how I'd explain that to a bourgeois pedantic mathematician
<egg|zzz|egg>
("bourgeois pedantic mathematician" is a phrase I got from Trubowitz :-p)
<bofh>
...I figured as much. :P
<egg|zzz|egg>
(in fact looking it up yields one result)
<egg|zzz|egg>
(gods those lecture notes are so very quotable)
<egg|zzz|egg>
"You don't need to know how to differentiate determinants. You just differentiate this one, like a physicist---or a not so bourgeois and pedantic mathematician. A mathematician, at least one with a certain kind of education, would stop dead here. They'd develop a whole new theory of differentiating matrices, with their own journal, and they'd have conjectures, and prove them, and go on to prove things nobody wants
<egg|zzz|egg>
to hear about."
<bofh>
(and soon it will likely yield another, in the form of the Thomas-whitequark logs for this channel :P)
<bofh>
I mean most of the math I did was analysis, so I guess same? :P
<egg|zzz|egg>
bofh: otoh I have shorthand for all greek letters because they actually come up often and writing out every letter makes note taking annoying
<egg|zzz|egg>
bofh: one potential issue with calling it the homogeneous elements of the tensor algebra is that you end up identifying 0, 0 m, 0 s, etc. which feels odd