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.
<awang>
I heard there was something about a possible fourth flavor of neutrino?
<awang>
Anyone know anything more about this?
<UmbralRaptor>
Results from a neutrino detector show ultrafast oscillations, suggesting an additional flavor. They match with one experiment, but I think conflict with others?
<awang>
Ah
<awang>
The best kind of result
<awang>
Sterile neutrinos?
<UmbralRaptor>
Ones that don't interact via the weak force, so are produced via oscillations only.
<UmbralRaptor>
Or maybe some gravity weirdness in the early universe.
e_14159_ has joined #kspacademia
<SnoopJeDi>
ooh, it's MiniBooNE at Fermilab
<SnoopJeDi>
also, ATLAS released results confirming the Higgs interacting with the top quark at > 6 σ
<SnoopJeDi>
ouch, these results are at 4.8σ on their own, that's a rough break.
<SnoopJeDi>
although this combined significance stuff has been really catching fire since CMS/ATLAS...
e_14159 has quit [Ping timeout: 202 seconds]
<SnoopJeDi>
uh, 4.8 for electron anti-neutrino. Lower for electron neutrinos (4.5)
<SnoopJeDi>
"Although the data are fit with a standard oscillation model, other models may provide better fits to the data." rings alarmingly strongly of Neumann's elephant
<bofh>
Meanwhile I get pissy b/c someone I've WORKED WITH published a neat result with possibly the most misleading title I've seen outside of a paper I've been asked to review, grr.
<kmath>
<colby> As is the case with kids, if you keep your printer active during the day, it will sleep better at night.
<egg|zzz|egg>
UmbralRaptor: bofh: wait no the inclinations deduced by transit impact parameter are correct if your reference plane is the plane of the sky
<egg|zzz|egg>
that's why they all take this reference system
<egg|zzz|egg>
(it means the inclinations are all very near π/2 which makes it an unusual reference system but it has that advantage)
<egg|zzz|egg>
this was today's episode of "Euler angles work in mysterious ways"
egg|z|egg has quit [Ping timeout: 202 seconds]
egg|z|egg has joined #kspacademia
egg|phone|egg has joined #kspacademia
egg|cell|egg has quit [Ping timeout: 198 seconds]
Iskierka has quit [Ping timeout: 202 seconds]
egg|z|egg has quit [Ping timeout: 182 seconds]
Iskierka has joined #kspacademia
egg|cell|egg has joined #kspacademia
egg|phone|egg has quit [Read error: -0x1: UNKNOWN ERROR CODE (0001)]
egg|phone|egg has joined #kspacademia
egg|mobile|egg has joined #kspacademia
egg|phone|egg has quit [Read error: Connection reset by peer]
egg|cell|egg has quit [Read error: Connection reset by peer]
<kmath>
<astrobites> Neufeld at #AAS232: A molecular zoo! Nearly 200 interstellar/circumstellar molecules discovered in the past 80 years https://t.co/x5mXE7vB0t
<awang>
bofh: Really? I can't remember seeing one before
<awang>
Although I've seen references to their papers
APlayer has joined #kspacademia
<egg|work|egg>
umbralraptor: bofh: spherical trig & stats question: if normal vectors are drawn from the distribution given on page 11, 1st column of https://arxiv.org/pdf/1711.05739.pdf, 1. how are the inclinations and LANs distributed; 2. given fixed inclinations, how are the LANs distributed
<APlayer>
Humm, egg, do you happen to know what integrator(s) stock KSP uses for its physics engine?
<egg|work|egg>
Forward Euler like anything else that doesn't think about it
<SnoopJeDi>
kinda surprising that it's not leapfrog
<egg|work|egg>
that's sort of not how physics engines roll
<egg|work|egg>
physics engines like having a concept of "here are all the positions and velocities at this time", and having discrete time, so offset position and velocity doesn't trivially fit
<APlayer>
And there was this orbital decay issue, was it related to the bad integrator choice?
<egg|work|egg>
force is similarly constant by frame (and often not even computed by the engine, so having it nonconstant would mean exposing something much more elaborate), so you get Euler
<SnoopJeDi>
egg|work|egg, that makes sense, and gamedev in general is KISS so I guess forward Euler makes the most sense from a design perspective to keep failure surface area small
<egg|work|egg>
APlayer: it's not even a choice
<egg|work|egg>
SnoopJeDi: doing anything else requires a conceptual leap
<egg|work|egg>
there is no such thing as "apply a force to this body now"
<egg|work|egg>
that becomes "the force is this function of <whatever>"
xShadowx|2 has joined #kspacademia
<egg|work|egg>
that's massively more complicated and not just how they do things
<SnoopJeDi>
err sorry not leapfrog, velocity verlet
<SnoopJeDi>
dunno why I always conflate those two
<egg|work|egg>
because they are the same
xShadowx has quit [Ping timeout: 182 seconds]
<egg|work|egg>
the more confusing thing is why you insist on not conflating them :-p
<SnoopJeDi>
OH, right, velocity Verlet means "do leapfrog" but Verlet by itself doesn't imply that
<egg|work|egg>
SnoopJeDi: re "at makes sense, and gamedev in general is KISS so I guess forward Euler makes the most sense from a design perspective to keep failure surface area small", you seem to always have this belief that things this deep are designed rather than the product of meaningful hysterical raisins
<kmath>
<FioraAeterna> dark secret: at least half the time, the compiler does <X> because nobody wrote any code to make it do something el… https://t.co/x0tleFUrAs
<SnoopJeDi>
my belief stands corrected
<SnoopJeDi>
anyway, I think I picked up the confusing brain-connection from someone in the plasma simulation world who had a habit of using "Verlet" to refer to velocity Verlet
<egg|work|egg>
it doesn't really take a design decision to implement Euler, it's what comes naturally if you say "the user of the game engine does things at fixed steps, sets a force, and it's back to the engine"
<SnoopJeDi>
sure, that's stating the obvious
<SnoopJeDi>
I guess I see a lack of design as an implicit design decision to stick with whatever "naturally" falls out from working the problem
<SnoopJeDi>
and certainly a very common one because it wastes a lot less time
<egg|work|egg>
well it's more that once that's how engines work, that's how engines work; the alternative would be a very different beast
<egg|work|egg>
compare "the user provides a callback that will be called at implementation-dependent times to compute forces, which may only use a subset of things that are actually being integrated, rather than the whole state of the game"
<SnoopJeDi>
you're not wrong
<SnoopJeDi>
I'm the last person on the planet who will defend game design as a place of sane architecture
<SnoopJeDi>
relatedly: RIP OpenCL
<egg|work|egg>
I mean, "the default is 'not designed'" holds everywhere, not just gamedev, not just compilers :-p
<SnoopJeDi>
sure
* egg|work|egg
pokes UmbralRaptor and bofh in the spherical trig
<SnoopJeDi>
I like what Jonathan Blow had to say about this in some YouTube video he uploaded while designing Jai. He was talking about designing "inside-out", starting with dumb naive code and successively raising it into higher abstraction as dictated by the problem
<SnoopJeDi>
although "I like what Jonathan Blow had to say..." is a good way to start a fight
<SnoopJeDi>
I wonder what integrators Open Dynamics Engine has
<kmath>
<eggleroy> @ManishEarth Hah, I was going to point you to our codebase :-p ⏎ ⏎ Re. the surrounding discussion, if you need example… https://t.co/TqcgxrIY3V
<bofh>
egg|trappist|egg: I was waiting for you to notice that thread tbqh :P
<kmath>
<bofh453> A million times this; somehow a lot of math folk don't seem to understand this & go overboard every time, no matter… https://t.co/jyDAq9TwDI
<SnoopJeDi>
yup
<egg|trappist|egg>
what do you mean making the isomorphism between the naturals and the nonnegative integers is too much,
<SnoopJeDi>
bofh, I think it has something to do with whether one recognizes an audience
<egg|trappist|egg>
you wouldn't want to confuse SS0 with the equivalence class of (0, SS0), would you,
<egg|trappist|egg>
bofh: was it Chevalley who objected to that particular map being implicit
<egg|trappist|egg>
I think Serre touches on that in How to write mathematics badly
* SnoopJeDi
grows his to-watch list
<e_14159_>
So, update: In the latest version of my student's master's thesis, I found three errors. On the title page.
<e_14159_>
(I've counted "Does not follow the university formal requirements" as one)
<egg|trappist|egg>
e_14159_: not sure mine was much better tbh
<e_14159_>
Did you write your advisot's name correctly? Did you have the correct date?
<e_14159_>
s/advisot/advisor
<Qboid>
e_14159_ meant to say: Did you write your advisor's name correctly? Did you have the correct date?
<egg|trappist|egg>
okay that I think I did :-p
<e_14159_>
Yeah, you're better.
<egg|trappist|egg>
I did initially write it with my own XeTeX style
<e_14159_>
Next feedback round will be the third time I correct my own name on the title page :-/
<egg|trappist|egg>
(which made it 20 p. shorter because it's way more compact and has figs in the margins)
<e_14159>
Currently, my plan is to note it in every feedback and, if it's still there in the pre-printing version write it explicitly in an email.
<e_14159>
If it's in the printed version, I'm not sure what I'm going to do. Despair?
<egg|trappist|egg>
do you physically meet the student?
<egg|trappist|egg>
"please write my name on the blackboard"
<e_14159>
Once per week. Except for this and next week (because feedback isn't conductive to regular meetings)
<e_14159>
Nah, he can write my name (at least first name) perfectly fine in emails.
<e_14159>
Btw, the error is in both first and last name.
<egg|trappist|egg>
:D
<e_14159>
egg|trappist|egg: Cool template. I actually think the figures in margins work well for that; it reminds me of a lecture script.
<e_14159>
Supporting the text, essentially.
<egg|trappist|egg>
yep
<egg|trappist|egg>
inspired by the Feynman lectures actually
<e_14159>
Oh, also: The figures (including stuff where you'd normally need subfigures) have clearly been drawn in some drawing program. Why do I know that? Because (a) it uses a different font, and (b) I can still see the background alignment grid because it's apparently *screenshots*
* e_14159
nods
<egg|trappist|egg>
aaaaaaaaaaa
<e_14159>
Agreed.
<e_14159>
I've asked a colleague of mine to review the thesis, because I'm a bit biased against it.
<bofh>
20:23:35 <@egg|trappist|egg> bofh: was it Chevalley who objected to that particular map being implicit
<bofh>
That souns familiar
<egg|trappist|egg>
bofh: I mean, who in their right mind would say that {} = {{{}}}, or {{}} = {{{{}}}, {{{}}, {}}}
<egg|trappist|egg>
bofh: чеб-собака? чебака? чубака? ... okay, considering his beard,
<bofh>
LOL
<egg|trappist|egg>
Thomas: I'm trying to load Gregox's SLIPPIST-1 with the latest Kopernicus, and getting nullreference spam in Kopernicus.Components.KopernicusStar.LateUpdate ()
<egg|trappist|egg>
Thomas: are you familiar with that one?
<egg|trappist|egg>
I think I've seen it already
<egg|trappist|egg>
also a dialog which tells me that there was an exception in the loading process
<egg|trappist|egg>
Thomas: hm, apparently head is broken in gregox's repo, nevermind
<egg|trappist|egg>
whitequark: blarg now i have to comment on a rust rfc >_>
awang has quit [Ping timeout: 198 seconds]
awang has joined #kspacademia
UmbralRaptop has joined #kspacademia
<egg|trappist|egg>
bofh: argh my RFC comment is turning into a weird biography
* UmbralRaptop
waves from DEN
<egg|trappist|egg>
!wpn UmbralRaptop
* Qboid
gives UmbralRaptop an Agile parabola
<UmbralRaptop>
!wpn egg|trappist|egg
* Qboid
gives egg|trappist|egg a clopen hurricane
<bofh>
is that, like, a hurricane undergoing an eyewall replacement cycle, or?