raptop 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. | We can haz pdf
<kmath>
<_juliannemarie> Google: Bach Doodle! Happy Birthday! 😄 ⏎ ⏎ don't say it ⏎ don't say it ⏎ don't say it ⏎ don't say it ⏎ don't say it ⏎ don't say… https://t.co/BVN1vA4uz8
<oeuf>
bofh: apparently last time they had issues where the police had been deployed in the western suburbs because facebook rumours, and nothing happened there; this time they apparently flew quadcopters to have a centralized idea of where shit was happening
<bofh>
So I'm pretty sure with a bit of added training rulesets you can fix a lot of the issues in that thread.
<oeuf>
yeah, it seems half-arsed
<bofh>
Like, very minimal added rulesets. But you have to have a very good knowledge of counterpoint for that, and I suspect the makers didn't.
<bofh>
Indeed.
<oeuf>
then again, even if you just hardcode in the rules of counterpoint you'll get boring
<bofh>
also rofl amazing. by Western suburbs do you mean like, Boulogne? Nanterre? Versailles?
<oeuf>
not sure
<oeuf>
bofh: with the thing that passes for organization in that movement, it's sort of hard to predict where things will happen
<bofh>
I mean part of *that* issue is that "music theory rules" really *ARE* more like guidelines, and treating them like hard rules gets boring fast. At the same time, part of the art is knowing *how* to break them.
<bofh>
i.e. Bach has done parallel fifths a few times! Except they don't sound objectionable, b/c of where in the piece they are and what the surrounding voices are doing.
<bofh>
And yeah, point. Which is why I'm a bit surprised they didn't try the video quadcopter thing earlier.
<bofh>
Particularly given that a key property of hi-vis jackets is that they're, erm, highly reflective.
<bofh>
(Or, fuck, just send someone up to the observation deck on the Tour Montparnasse with a pair of binoculars).
<oeuf>
bofh: from a tower you only really see into streets that are aligned with it
<oeuf>
bofh: and the traditional technique of people everywhere radioing reports is probably a bit messy esp. if you're short on personel
<oeuf>
!wpn whitequark
* galois
gives whitequark a mercury americium polynomial with a pterodactyl attachment
<oeuf>
!wpn bofh
* galois
gives bofh a mace
<oeuf>
!wpn UmbralRaptop
* galois
gives UmbralRaptop a robin with a pterodactyl attachment
<kmath>
<✔vvantighem> #GiletsJaunes #ChampsElysees : Ce lundi, on est devant la chambre 24-1 du TGI de Paris pour suivre les comparution… https://t.co/oVHMJ4iwz4
<oeuf>
UmbralRaptop: also are the planetenweg radiation burns getting better
<oeuf>
bofh: so apparently they've done the thing again of using sentinelle troops to guard buildings that would be guarded by police, so as to be able to have more police doing riot control
<oeuf>
bofh: and then melenchon decided to make a speech telling them that they shouldn't shoot the crowds, because melenchon
<kmath>
<Michel_Goya> 21 mars 2019 : la France découvre que les individus en tenue camouflée qui patrouillent dans les rues depuis 24 ans… https://t.co/E2ft0UODGh
<oeuf>
!wpn bofh
* galois
gives bofh a adobe pick
<oeuf>
hm
<oeuf>
i guess you can dig into adobe pretty easily
<oeuf>
bofh: oh, TIL: The word adobe [...] can be traced from the Middle Egyptian (c. 2000 BC) word ɟbt "mud brick".
<bofh>
yep, heck I've heard "adobe bricks" being used as a term in even fairly recent books
<bofh>
oeuf: yeah, I followed @vvantighem for that reason :p
<oeuf>
bofh: ah, so you hear about the regularly scheduled trials right after the regularly scheduled riots
<bofh>
yep
<bofh>
I can tell that my French reading ability is getting better since my rate of facepalms upon reading the regularly scheduled trials is increasing pretty much linearly.
ferram4_ has quit [Read error: Connection reset by peer]
ferram4_ has joined #kspacademia
<oeuf>
bofh: oh, remember how they used to prepare for saturdays with planks on every window
<kmath>
<✔YahooActuFR> Une sorte de sarcophage de métal est installé autour du Fouquet's à la veille de l'#Acte19 #acteXIX #GiletsJaunes https://t.co/cgi4tDV9CO
<kmath>
<cjccommander> New app idea: Tinder, but for PhD students who need to find a post-doc. Pairs you with compatible P.I.s in your are… https://t.co/KlhzeeoV37
<kmath>
<Tom_Ruen> Here are the position of some 18 black hole candidates in the Milky Way galaxy, overview and edge-on view, all but… https://t.co/yiRjUPEIsU
<ZHRraptor>
It feels like airports are standardizing on 3 security checkpoints?
ZHRraptor has quit [Ping timeout: 202 seconds]
UmbralRaptop has joined #kspacademia
<UmbralRaptop>
Between using US and CH transit systems, I refuse to believe that they're simple and low stress.
<egg>
bofh: if i points to the the middle row, it seems fairly obvious that you want to compute symmetric differences of the fⱼ, coefficient that, and sum from the outside, but what about the other rows
<egg>
bofh: meow
* egg
stares at a type in KSP called TypeData
<bofh>
egg: meow
<bofh>
uhh, sec.
<egg|cell|egg>
Bofh: sec
<egg|cell|egg>
Bofhsec
<egg>
!wpn bofh
* galois
gives bofh a titanium glaive
<bofh>
egg|cell|egg: need to first obtain caffeine, bad headache. almost done the fetchquest tho.
<bofh>
turns out 1g caffeine/day means you run out of pills fast >_<
<bofh>
egg|cell|egg: honestly I'd say the same approach except summing from the inside for the rows closer to the top?
<bofh>
like these are all integers well within the eggsactness range for a double, so it shouldn't matter too much???
<egg>
bofh: the question is the fⱼ
<egg>
bofh: this should behave well when they are close
<egg>
bofh: these are finite difference formulae
<egg>
<bofh> egg|cell|egg: honestly I'd say the same approach except summing from the inside for the rows closer to the top? << what do you mean
<egg>
bofh: that approach is only well-defined for the central row because it's symmetric
<egg>
well, antisymmetric
<egg>
!wpn bofh
* galois
gives bofh a cobalt iridium soufflé
<bofh>
egg: I mean this whole matrix is antisymmetric
<egg>
bofh: please reread the matrix
<egg>
for more information please reread,
<bofh>
...okay, it's symmetric about the diagonal we don't care about,
<bofh>
(oops >_<)
<bofh>
(well, antisymmetric)
<bofh>
hmm
<_whitenotifier-3d18>
[Principia] pleroy opened pull request #2106: Fix minor issues found when debugging the MapNodes - https://git.io/fjJyI
<_whitenotifier-3d18>
[Principia] pleroy closed pull request #2104: Restructure the rendering of the flight plan - https://git.io/fjJc8
<_whitenotifier-3d18>
[Principia] pleroy pushed 8 commits to master [+0/-0/±17] https://git.io/fjJyL
<bofh>
egg: symmetric differences I think are pretty much always correct for any row
<egg>
wat
<bofh>
egg: okay, where would you *not* want to compute symmetric differences of the fⱼ as the first step?
<egg>
bofh: what I'm saying is that when i = (n-1)/2, sum for j in 0 .. n-1 of a_ij f_j is best computed as sum for j in 0 .. (n-3)/2 of a_ij (f_j - f_(n-j-1))
<egg>
bofh: that is not true for other values of i
<egg>
so how is that even applicable
<bofh>
why is it not applicable for other values of i?
<egg>
bofh: because the equality between sum for j in 0 .. n-1 of a_ij f_j and sum for j in 0 .. (n-3)/2 of a_ij (f_j - f_(n-j-1)) does not hold
<egg>
bofh: because the equality between a_ij and - a_i(n-j-1) does not hold, noor does a_i((n-1)/2) = 0
<egg>
s/noor/nor/
<galois>
egg meant to say: bofh: because the equality between a_ij and - a_i(n-j-1) does not hold, nor does a_i((n-1)/2) = 0
<bofh>
...*stares at matrix again*
<bofh>
okay I see what you mean now >_< it prolly makes more sense for me to continue thinking about this in a half-hour when this headache is gone, I'm normally not *this* bad at noticing things.
* egg
gives bofh a flyswatter
<egg>
!wpn -add:wpn flyswatter
<galois>
Added wpn 'flyswatter'
<bofh>
...I'm surprised that wasn't already in there.
<egg>
bofh: the galois wpn list is very smol, this isn't qboid
<egg>
same for its acr list
<egg>
!acr -list
<galois>
ValueError: not enough values to unpack (expected 2, got 1) (file "/home/snoopjedi/.sopel/modules/acr.py", line 27, in acr)
<bofh>
*snicker*
* egg
finitely differences bofh
<_whitenotifier-3d18>
[Principia] pleroy synchronize pull request #2105: Implement the main window using a separate class - https://git.io/fjJux
<bofh>
Right, finite differences. Now that my head no longer feels like it's being finitely differenced.
<bofh>
These matrix rows actually are reminding me a lot of the Taylor Series for Ai(x)/Bi(x) in how the coefficients look like
<bofh>
...which means that they likely *will* be a cancellation nightmare if naïvely summed
<egg>
bofh: yes
<bofh>
Unfortunately I don't think I ever found a good solution for dealing with cancellation in that, especially near zeroes of Ai(-x). Summing from smallest coefficient to largest if that's possible in any meaningful way helped, keeping the coefficients as integers and then multiplying by inverse scaling only after Estrin'ing the series didn't do *much* (or rather it helped less than I was eggspecting it
<bofh>
to)...
<bofh>
hm
<bofh>
Like the worst case turned out to be when your result is very smol but your summed values aren't
<bofh>
(In the end my solution is very specific to the properties of Ai(-x), so it doesn't apply here at all, so grr)
<bofh>
Like if you have to process each row independently (which I think you do, again, my knowledge of finite difference methods is decidedly lacking) then I don't think you can do much trickery with how you compute the fⱼ's
<egg>
<bofh> Like the worst case turned out to be when your result is very smol but your summed values aren't << thanks, cpt obvious
<egg>
:-p
* egg
pokes the condition of subtraction with a stick
<egg>
bofh: the coefficients are exact (or rather I put them on a common denominator already, so the numerators are exact)
<egg>
bofh: multiplication by a coefficient can be done exactly by widening, of course
<egg>
bofh: but then there's the differencing
<egg>
bofh: I suppose in the interesting cases the differencing is exact, because Sterbenz
<bofh>
egg: I'm not sure, and I'll have to poke at it a bit to have a better idée.
<PHLRaptor>
Is it my imagination, or is US customs more bureaucratic than most countries? Also, they had advertising for TSA precheck before we landed o_O
<PHLRaptor>
Other random observation, an Indian grad student with two passports (one expired), both full of (photo) visas so he has sufficient documentation
<PHLRaptor>
… I could be wrong and he's a postdoc
<egg|cell|egg>
PHLRaptor: phl is saying that you should do solar eclipses
<B787_300>
PHLRaptor: how was the flight?
<PHLRaptor>
B787_300: delayed at the start by routing issues (so they needed to get an extra 1000 gallons of fuel)
<B787_300>
that was the ZRH-PHL leg or the PHL-IAD leg?
<PHLRaptor>
egg|cell|egg: Let me try to survive midterms first
<PHLRaptor>
the former. I'm at PHL in a customs line
<B787_300>
slight rant... i hate the use for former and latter because i can never rememebr which one is which
<bofh>
19:59:57 <@PHLRaptor> Is it my imagination, or is US customs more bureaucratic than most countries?
<bofh>
Yep, that's been my eggsperience as well.
<B787_300>
huh i have always thought it wasnt that much different than the customs at BVI, UK, Calais or Bolivia
<PHLRaptor>
Form's longer than UK. CH is nicely chill
<bofh>
Okay I've never been to the UK (and quite frankly I don't see any reason to change that) so I can't confirm.
<B787_300>
CH is chill? dont you need a visa to visit China?
<whitequark>
CH is Switzerland
<whitequark>
CN is China
<B787_300>
right
<B787_300>
because Confederation Helvetica
<B787_300>
i should have remembered that converstaion about country codes in unicode
<PHLRaptor>
I guess CH figures that if they greet passengers with a guard with an MP5, they don't need a lot of forms? ¯\_(ツ)_/¯
<B787_300>
PHLRaptor: and you dont think CBP is armed?
<egg|cell|egg>
In France ít would be a famas
<PHLRaptor>
B787_300: I wasn't expecting someone right off the jetway
<egg|cell|egg>
Oh my keyboard í in việt hence the weird accents
<bofh>
20:09:24 <@B787_300> CH is chill? dont you need a visa to visit China?
<bofh>
Yes, you need a visa to visit PRC. Getting a tourist visit visa is *easy* tho and customs/entry screening was *significantly* more chill than the US one.
<B787_300>
bofh: huh apparently i misspelled china so bad my phone corrected it to chill
<egg|cell|egg>
PHLRaptor: bofh: should I share your Planetenweg pics with bofh resp. PHLRaptor
<bofh>
Obvious anecdote, but still, going to Shanghai was so much smoother than any of my US entries.
<PHLRaptor>
egg|cell|egg: yeah.
<bofh>
egg|cell|egg: feel free to link my Planetenweg pics in here even, I don't mind
<bofh>
(in fact I'm pretty sure I've done that already in the past, oops >_<)
Wetmelon has joined #kspacademia
<egg|cell|egg>
PHLRaptor: I accidentally phl on that album apparently
<PHLRaptor>
Hah!
<PHLRaptor>
The Saturn pic is decent, the rest… eh
<PHLRaptor>
bofh: how did you find the Leonardo Boutique, anyway? they mainly seem to be in DE?
<bofh>
(also gods my voice is absolutely atrocious >_<)
<egg|cell|egg>
Meh it's fine, recordings do that to oneself
<bofh>
PHLRaptor: as in how was it or how did I come across it?
<egg|cell|egg>
For how it was PHLRaptor can tell, he was there
<PHLRaptor>
Mostly the latter
<bofh>
PHLRaptor: I logged onto booking.com as per egg|cell|egg's repeated insistence, sorted by price low to high for hotels in Zürich, that was I think the second on the on the list and near ETHZ
<bofh>
a few hours and egg|cell|egg's repeated insistence that I, quote, "GET A ROOM" later, I actually booked it, lol.
<PHLRaptor>
Collinear with <egg|home|egg>, and ETHZ
<PHLRaptor>
hah
* PHLRaptor
stares at times
* PHLRaptor
is going to be entirely too close on the PHL→DCA flight
<PHLRaptor>
(still in line, have gone through a mandatoryish self service kiosk)
<bofh>
also RIGHT that's when I read Hairer, Lubich & Wanner
<bofh>
at like 4AM by ETHZ after train misadventures at Gare de Lyon
<PHLRaptor>
Have I mentioned that the temporal density of trams in Zürich is at least an order of magnitude higher than buses in DC/Fairfax?
<bofh>
I mean I like how I considered the temporal density of trams there to be a bit annoying b/c it's about half as frequent as any Paris métro ligne, but also still 1.5 orders of magnitude higher than buses in Urbana-Champaign.
<bofh>
Public transit here is so infrequent that I often just wind up walking since it's *faster*.
<PHLRaptor>
Yep, done that
<PHLRaptor>
… Either my connecting flight is getting delayed, or I'm going to miss it >_<
<B787_300>
PHLRaptor: why is it taking you so long to clear customs
<PHLRaptor>
long lines
<PHLRaptor>
Anyway, got through customs, picked up checked luggage (mandatory), rebooked for a later flight
<PHLRaptor>
Now in another line because apparently arriving internationally means going through security again
<B787_300>
yep
<B787_300>
and damn they must have been really understaffed for the incoming flights if it took you this long
<egg|cell|egg>
Bofh: probably going to a lubich seminar Tuesday
* PHLRaptor
chooses to interpret that as egg visiting Ljubljana
<egg|cell|egg>
Bofh: tram frequency depends on the time
<egg|cell|egg>
5-7 min during most of the day, 15-20 min le
<egg|cell|egg>
s/.$/ate
<galois>
egg|cell|egg meant to say: 5-7 min during most of the day, 15-20 min late
<PHLRaptor>
"It's TSA, they'll never fix it." on duty person complaining about equipment
<B787_300>
PHLRaptor: SECURITY THEATER
<B787_300>
just remmeber the GAO has been able to sneak shit through the TSA will alarming regularity
<bofh>
egg|cell|egg: yeah I prolly have a biased sample given how I'd hang around with you during daytime and then mostly use trams myself late
<bofh>
B787_300: look I sneaked fuming sulfuric acid past TSA in my carryon by accident, it's not that hard.
<B787_300>
why did you have sulfuric acid of any type in your backpack?
<egg>
B787_300: bofh
<B787_300>
yes?
<egg>
B787_300: bofh had sulfuric acid because bofh
<B787_300>
is H2SO4 part of his EDC?
<SnoopJeDi>
!acr -list
<galois>
I've uploaded a list of acronyms I know about: http://ix.io/1EpK
<SnoopJeDi>
egg, ^
<SnoopJeDi>
hooray for simple APIs \o/
<egg>
is ix.io related to iximeow
<SnoopJeDi>
!8
<galois>
SnoopJeDi: no
<iximeow>
meow
<egg>
ixi
<iximeow>
ix.i
<SnoopJeDi>
I dunno but I wish it was open-source >:(
<SnoopJeDi>
not that there aren't open-source pastebin apps, but I get the impression that ix.io is probably remarkably simpler than some of those.
<SnoopJeDi>
bofh, oh neat, I'm glad CERN did the VERY OBVIOUSLY RIGHT THING in letting Strumia go
<bofh>
SnoopJeDi: yeah, good fucking riddancce
<bofh>
B787_300: EDC?
<B787_300>
Every Day Carry
<SnoopJeDi>
an acronym dangerously close to tacticool culture
<bofh>
B787_300: I had sulfuric acid b/c I was working with eggsperiments that semester that required it a *lot*, and happened to that day fetch some from chemstores for the next day and then realized that evening I had a flight out to NYC in, like, 10 hours and it takes 4 hours to get to ORD
<bofh>
so I packed in an absurd hurry, forgetting about it
<B787_300>
lol
<B787_300>
was it over 4 oz of liquid?
<bofh>
it was 145mL in a glass bottle
<bofh>
so I think ~5 fl. oz
<B787_300>
ohhh about right at the limit of allowability (other than the whole Acid part)
<bofh>
it also very clearly had a bunch of warnings on it b/c it was 100% sulfuric acid with 12% dissolved sulfur trioxide
<SnoopJeDi>
bofh, ICYMI as further evidence that Mika is absolutely your people: she got some of the Rice geo people to let her into the Earth Science dept yesterday and wandered around looking at their samples: https://twitter.com/mikamckinnon/status/1109640733554851840
<kmath>
<✔mikamckinnon> A pair of local geos let me in to the the Rice Earth Sciences building after hours and I’m totally crushing on this… https://t.co/S5UdYDAKl0
<bofh>
SnoopJeDi: oh man that reminds me of when I snuck into Earth Sciences at U. Laval on a visit once and spent the night staring at rocks (some fluorescent and some weakly radioactive, even).
<bofh>
I should get some travel rocks (well okay I have a Ce⁺³:YAG that I still need to get around setting in a ring sometime, meant to do that months ago).
<SnoopJeDi>
travel rocks is a BIG MOOD
<SnoopJeDi>
I just gave my gf my braindump of both days of the workshop as a dry run for turning my notes into slides (and maybe a tweetstorm?)
<PHLRaptor>
shiny
<bofh>
A tweetstorm would be great for anything that's at an accessible level, which I think was a lot of it by the sounds of it.
<bofh>
PHLRaptor: so keep in mind that 03h00 is like within ideal working hours for me, so that's not a great comparison :p (better would be stuck walking 5km at what my internal clock thinks is 07h30, which... still yes on many occasions, even if we restrict to int'l travel).
<PHLRaptor>
I'll accept that
<bofh>
Like keep in mind this is *me*, I'm pretty sure I have some sort of magic ability to cause especially inter-Europe passenger rail to fail in bizarre ways
<PHLRaptor>
Flight departs at 2103, last CUE bus departs metro station at 1655, last 29K/N 2010 from one station, 2139 from the other
<PHLRaptor>
bofh: please don't cause WMATA to spontaneously implode
* egg
pokes bofh in the finite differences
<_whitenotifier-3d18>
[Principia] pleroy opened pull request #2109: Changes to compile on Linux - https://git.io/fjJQH