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>
<johnregehr> @TerminateThread I definitely need a better "it's going to get squashed anyway" commit msg
egg|df|egg is now known as egg|zzz|egg
<UmbralRaptop>
TFW walking 3 miles is faster than waiting for the bus.
e14159 has quit [Killed (NickServ (GHOST command used by e_14159!~e14159@dslb-188-098-213-216.188.098.pools.vodafone-ip.de))]
e_14159 has joined #kspacademia
e_14159 is now known as e14159
<bofh>
UmbralRaptop: been there done that
<UmbralRaptop>
Bonus points if it's raining.
<bofh>
UmbralRaptop: been there done that
<bofh>
:P
* UmbralRaptop
awards bofh the requisite number of bonus points.
<UmbralRaptop>
Anyway, ChExo was fun, aside from transportation issues.
<UmbralRaptop>
Also, apparently Uranus has a weirdly active radiation belts (2nd only to Jupiter), while Neptune's are notably quiescent?
<bofh>
Wait, *what*?
<bofh>
via what proposed mechanism?
<bofh>
(mostly curious about the Uranian ones, Neptune being quiescent kinda is unsurprising).
<UmbralRaptop>
Mechanism unknown, the speaker just said that's what has been measured, and a reason for ice giant orbiters.
<tawny>
quiescent here means not as radioactive, or something else?
<UmbralRaptop>
Sorta. Lack of particles, especially at higher energies.
<bofh>
I mean the reasons for ice giant orbiters are { gigantic scroll that runs a km in length and is written in 12-point Courier New }, tbqh.
<iximeow>
this is accurate
<UmbralRaptop>
Yep.
* UmbralRaptop
does hope that they'll be a high priority on the 2020 decadal survey.
<bofh>
Same.
<bofh>
Like, we only have so many launch windows for those since we need at least Jupiter grav assists.
<bofh>
Ideally J+S.
Ellied has quit [Ping timeout: 182 seconds]
<B787_300>
UmbralRaptop: if you are so concerned about photons why don't you use a really narrow FOV and a pmt?
<bofh>
seconding the PMT suggestion
<UmbralRaptop>
PMT?
<B787_300>
Photo multiplier tube
<UmbralRaptop>
Ah. Because that's solving the wrong problem. The targets are decently bright, even spread out over an R~100k
<UmbralRaptop>
I just need absurd SNR over a night to get a precision in radial velocity slower than most people walk.
<UmbralRaptop>
"just"
<B787_300>
Right but you are trying to detect dips in brightness?
<UmbralRaptop>
minute red/blue shifts.
<B787_300>
...
<UmbralRaptop>
Radial velocity method, not transit method. >_>
<B787_300>
Oh you poor soul
<B787_300>
No wonder a 3.2 is too small for you
<bofh>
Oh, yeah, it all makes sense now.
<B787_300>
Hrm I wonder if you could get creative and use 3+ PMTs with narrow band pass filters on them
<B787_300>
Although putting a splitter block and filter is putting something that might absorb the light
<xShadowx|2>
any math geniuses? not even sure what to google for this, lets pretend there is a solar system, and a planet that revolves around its star at some speed, and theres a ship at some point in the system, and can move at some speed, what kind of math is used to calculate the intercept point between the ship and the planet? i know how intercept works with 2 straight lines, but planet moves in circle
<xShadowx|2>
oh and ignore gravity effects :|
<B787_300>
UmbralRaptop: also how were you going to detect the shift with a monochromatic sensor?
<iximeow>
parametric equations \o/
<xShadowx|2>
just straight line intercepting circle path
<xShadowx|2>
thats what i should google? :D
<B787_300>
xShadowx|2: uuum well as you said ignore gravity the ship ain't moving unless it thrusts
<B787_300>
And can the ship move fast enough that you can assume the planet doers move?
<UmbralRaptop>
B787_300: dozens of orders from an echelle grating.
<iximeow>
it's probably not the ideal way, but the way i've done that in the past is phrasing the motion of all objects as functions of t, solving X and Y independently for t, and if any t is the same for both, they intersect
<iximeow>
(extending that to also Z if you're worried about three dimensional motion, which uh.. at the time i was not)
<B787_300>
xShadowx|2: I mean iximeow is right in the construct given to us. In actual space flight you need ephemerides and isp and wonderful conic sections
<xShadowx|2>
hoping theres some formula using X,Y (ship coords, assuming star is 0,0), R (radius of planet orbit), S1 (speed of planet), S2 (speed of ship) to just say "ok heres the closest point" :3
<iximeow>
B787_300: yeah given the way it's posed i figure it's a sufficient model
<B787_300>
UmbralRaptop: what is the QE of your detector? And could you get the Dark/read current any lower on it?
<UmbralRaptop>
I suspect that given initial positions / velocities, there's an analytic solution.
<B787_300>
Because if you need SNR and the S is locked you need to reduce the N
<B787_300>
So can you get cyrogens to cool your detector?
<UmbralRaptop>
B787_300: decently high, though overall system efficiency is <5%. Read noise is smallish, dark current is assumed to be 4 electrons/hour/pixel.
<iximeow>
ah, read noise measured in single electroncs/hr
* iximeow
needs to astro better
<B787_300>
I don't want to know how much that detector cost then...
<UmbralRaptop>
hah
<B787_300>
Like the I am used to seeing ~5e/sec/pix
<iximeow>
i'm slowly trying to figure out how to align my 11" sct as i know it's still a little off
<UmbralRaptop>
Teledyne's price for their Hawaii line of sensors is "call us" o_O
<iximeow>
lol
<B787_300>
UmbralRaptop: here is a thought... do you have a coude room and optics table to use? Because then you might be able to set up so that the grating is on one side of a table the you have multiple detectors that are close together physically on the other end so you are using the grating to help separate out really close color shifts
<B787_300>
iximeow: what brand is your SCT?
<iximeow>
B787_300: celestron. i got it "used/for parts" because the previous owner uh... didn't mount the secondary properly, it fell, chipped the primary, it's a whole thing
<iximeow>
i've not yet convinced myself the primary is seriously damaged (that picture ain't bad) but..
<B787_300>
Ah... all my recent experience is with meade or takahashi
<B787_300>
But I got to use a 32 grand sensor on a 16" ... So yeah the sensor was worth more than the scope
<UmbralRaptop>
iximeow: damaged secondary on an SCT? I'm slightly impressed. O_o
<iximeow>
UmbralRaptop: another source of problems i've not yet addressed is the uh
<iximeow>
smudge, on the secondary
<UmbralRaptop>
B787_300: We have an optics table, and will eventually have a clean room, though I'm not as close to the hardware as you might think.
<iximeow>
i think the previous owner fished the secondary out of the telescope and got a thumb on the mirror or.. something
* iximeow
cringes at the thought
<UmbralRaptop>
iximeow: as long as it; s not a blood spatter, or something…
<B787_300>
I am just thinking of ways to increase S and as PMTs are a bit large you need the angular separation
* iximeow
considers trying to fiddle with alignment again tonight
<kmath>
<ChemPlayer1> A bit of sleuthing indicates that we were finally killed off for 1) chocolate cake, 2) Grignard reaction, and 3) ph… https://t.co/SUjYEsFCUY
<egg|zzz|egg>
!wpn whitequark
* Qboid
gives whitequark a giant double/goto hybrid
<kmath>
<eggleroy> Transit observations behind the TRAPPIST-1 papers [1,2]. Spitzer time red [3], VLT time blue [4], K2 campaign 12 ye… https://t.co/A0htiu4NU4
<egg|zzz|egg>
... did they typeset Spitzer in mathmode
<egg|zzz|egg>
21st february to 26th march, so a fair bit of overlap with the K2 observations
<kmath>
<lexaloffle> By 35, you should have started more projects than you can hope to complete in your lifetime.
<UmbralRaptop>
heh
<egg|zzz|egg>
bofh: physicists like reinterpret_casting things in coordinates
<egg|zzz|egg>
bofh: which violates the strict aliasing rule^W^W^W coordinate-free invariants
<bofh>
THAT'S THE BEST WAY OF SUMMARIZING IT I'VE EVER HEARD THANK YOU
<bofh>
okay the more I stare at these expressions on this wikipedia page the less sense it makes.
<bofh>
like it seems to not actually *be* a bivector? but that doesn't make sense physically then
<egg|zzz|egg>
bofh: I obviously isn't a bivector, it's symmetric
<egg|zzz|egg>
and probably positive definite
<egg|zzz|egg>
it's a bilinear form
<egg|zzz|egg>
but how do you get from a vector to a bilinear form via a sinful skew-symmetric matrix
<bofh>
it's positive semidefinite via construction, at least
<bofh>
¯\_(ツ)_/¯ my linalg prof would've docked marks for this construction in all honesty.
<bofh>
I seriously want to rewrite this section now but first I need to figure out why this construction is sensible (for some weak value of "sensible").
<SnoopJeDi>
!wpn -add:adj sinful
<Qboid>
SnoopJeDi: Adjective added!
awang has joined #kspacademia
<egg|zzz|egg>
bofh: blargh I really don't see what this [.]^2 is doing >_>
awang has quit [Ping timeout: 194 seconds]
<egg|zzz|egg>
bofh: hm so L = r ∧ p = r ∧ mωr which is linear in ω =: Iω (where I is then also a bilinear form on bivectors via the inner product), but what operation is r ↦ (ω ↦ r ∧ m ω r)
<egg|zzz|egg>
(wlog ignore the m)
<bofh>
hrm
<bofh>
I'm not seeing wtf the [.] construction makes sense to do, honestly.
<egg|zzz|egg>
oh that one certainly makes no sense
<egg|zzz|egg>
but [.]^2 might be somethnig
<egg|zzz|egg>
let ω be pure, ω = v ∧ w, then that is r ∧ (v ∧ w).r where . is "bivector acting as an alternating form"
* egg|zzz|egg
stares blankly
* bofh
blinks
<egg|zzz|egg>
bofh: okay what if we start from the insanity, then it's Iω = (*r)((*r)(*ω)) where * is the Hodge star, how do we get of all those stars now Ꙩ_ꙩ
<egg|zzz|egg>
wait no there's probably another star here, Iω = *((*r)((*r)(*ω)))
<bofh>
sec, let me try and remember how Hodge * works
<egg|zzz|egg>
it's the [silly bracket]
<egg|zzz|egg>
given an orientation the [silly bracket] is sensible
<bofh>
okay yes there's definetly a fourth one, and I *think* that lets us cancel
<bofh>
ohh. right, b/c the construction fundamentally requires a choice of coordinate
<egg|zzz|egg>
bofh: just of orientation (I mean, it's in coordinates, but the construction can be rewritten *((*r)((*r)(*ω))) which only summons the Hodge star)
<egg|zzz|egg>
bofh: blarg no, ncatlab doesn't say anything we haven't already
<egg|zzz|egg>
oh there's the Lie approach
<bofh>
hm. *can* we banish the Hodge* entirely, even?
<egg|zzz|egg>
bofh: I certainly hope so, if rigid body dynamics depend on orientation I'll be annoyed
<egg|zzz|egg>
so will the white rabbit
<bofh>
like I feel like the innermost (*r)(*ω) should be simplifiable to an equivalent expression with the * outside and a wedge inside, but... actually nevermind. hm.
<egg|zzz|egg>
so *r is a bivector, *ω is a vector, this is a bivector acting on a vector so a vector
<egg|zzz|egg>
does it have something to do with ωr
<egg|zzz|egg>
bofh: I think they're the same thing, except pseudovector is the careful physicist word?
<SnoopJeDi>
"careful" :P
<egg|zzz|egg>
well the careless one will call it all vectors
<egg|zzz|egg>
bofh: there's the so(V) = V ∧ V isomorphism that I'm using extensively
<bofh>
yeah, like they're clearly isomorphic thanks to Hodge duality
<egg|zzz|egg>
bofh: okay (*r)(*ω) = rω afaict
<egg|zzz|egg>
bofh: yeah and then it's going to be obvious that this is r ↦ (ω ↦ r ∧ m ω r)
<egg|zzz|egg>
bofh: so the insane construction is correct
<egg|zzz|egg>
bofh: but what *is* r ↦ (ω ↦ r ∧ m ω r) anyway
<bofh>
okay yeah in that case we ARE good, but... yeah.
<bofh>
uh what is m in (ω ↦ r ∧ m ω r)?
<egg|zzz|egg>
bofh: so, in coordinates, this thing is the tensor product of r with itself except on the diagonal, where instead of x^2 you have -y^2-z^2
<bofh>
this seems... weirdly more complicated than it needs to be?
<UmbralRaptop>
B787_300: er, my problem is less detection than needing huge amounts of data to get the requisite precision. eg: Altair is 1) the brightest star in my target list, and 2) basically impossible to get useful data on because it has so few absorption lines.
<B787_300>
UmbralRaptop: but if you can use less integration time to get a lot of data
<B787_300>
as you mentioned you were needing silly long integration times
<B787_300>
and if you are detecting more efficiently then you are getting more data for a shorter time
<egg|zzz|egg>
bofh: and as a form, vIw = ⟨v,r⟩⟨r,w⟩-⟨r,r⟩ ⟨v,w⟩?
<egg|zzz|egg>
bofh: wait is a bilinear form on bivectors the same as a bilinear form on vectors, or do you need an orientation
<bofh>
I... that's a good question, actually. I THINK it should be the same for both vectors and bivectors, but now I'm not at all sure.
tawny has quit [Ping timeout: 198 seconds]
<egg|zzz|egg>
bofh: we have an inner product so I'd say yes, bilinear forms are linear maps and linear maps induce linear maps on multivectors?
tawny has joined #kspacademia
<bofh>
egg|zzz|egg: oh, good point. and yeah, writing it out in terms of inner products and linear maps, yes, it better be the same.
<egg|zzz|egg>
bofh: okay so now we can ignore the bivectors and we just have the bilinear form I(r) with I(v,w) = ⟨v,r⟩⟨r,w⟩-⟨r,r⟩ ⟨v,w⟩
<egg|zzz|egg>
bofh: I guess principia can just have a SymmetricBilinearForm type :-p
<bofh>
that makes things slightly clearer I guess?
<egg|zzz|egg>
bofh: and then you have the inner product form, given a vector its tensor product with itself is a symmetric bilinear form, symmetric bilinear forms form a vector space, and that's sufficient to construct I
<egg|zzz|egg>
bofh: of course you can generalize to the symmetric product (v⊗w+w⊗v)/2
<egg|zzz|egg>
but I don't think I actually need it
<kmath>
<qntm> Hypothetically speaking would it be possible for someone to punch out the window on an Apollo command module? I nee… https://t.co/5XWeyiKoZl
* egg|nomz|egg
should build a pinyin to viet consonants transliterator
<B787_300>
UmbralRaptop: i mean hypothetically yes... their hand might not be a hand afterword as i am pretty sure you cant force it out but you might be able to cause fatigue in the bolts/support stucture by applying a repetitive force to it
<UmbralRaptop>
Ow
<bofh>
egg|nomz|egg: that seems useful
<egg|nomz|egg>
bofh: I mean mostly if you can pronounce (some dialect of) viet but not standard pinyin :-p
<egg|nomz|egg>
bofh: although viet has no such thing as pinyin q or ch, only pinyin j or zh
<egg|nomz|egg>
cc rqou for chinese phonology
<B787_300>
UmbralRaptop: yeah... i mean you have to think that: 1. it is a wedge design with the narrow part on the outside to make it so that a blowout would be hard as hell (like how Airline windows are designed) and 2. it is a designed for a 14.7 psi pressure differential
<UmbralRaptop>
B787_300: given that this is for a silly SF story, I wonder if that means someone punching in on a depressurized capsule is plausible. >_>
<B787_300>
nope
<B787_300>
as then you have the same support structure
<bofh>
egg|nomz|egg: I mean such a mapping obviously need not be bijective, I guess you pick which of injectivity or surjectivity will fail in your mapping.
<egg|nomz|egg>
bofh: pretty sure on consonants there are just more in zh-cmn
<egg|nomz|egg>
bofh: fewer tones but they're afaict different so I have no idea how to pronounce them
<egg|nomz|egg>
rqou: halp i cannot into zh
<rqou>
er, what?
<rqou>
i'm not very good either
<egg|nomz|egg>
rqou: well, vi has [ʈʂ] and [tɕ] but none of the [ʰ] stuff so that confuses me; also I don't think [ɕ] is a thing ([ʂ] is though)
<egg|nomz|egg>
bofh: the reference epoch in all TRAPPIST-1 papers seems to systematically be the first observed transit, to the extent that this now appears to be unstated
<egg|nomz|egg>
bofh: clearly we should date everything in reference to the first observation of the transit of TRAPPIST-1 c
<bofh>
egg|nomz|egg: how to convert between that and J2000?
<kmath>
<sigfig> distinctly unclear if i just have a high probability of encountering strange events or if i do something to generate them
<bofh>
the local ClearChannel top40 station just inexplicably out of nowhere played Amish Paradise and then went right back to regular programming, with *zero* mention of it at all.
awang has joined #kspacademia
awang has quit [Read error: Connection reset by peer]
awang has joined #kspacademia
<Majiir>
!wpn UmbralRaptop
* Qboid
gives UmbralRaptop a Vulcan McParseface
* UmbralRaptop
coats Vulcan McRubberface
<bofh>
!wpn Majiir
* Qboid
gives Majiir a sigma branch
UmbralRaptop has quit [Quit: Bye]
UmbralRaptop has joined #kspacademia
egg|nomz|egg is now known as egg
<egg>
bofh: uncertainty in the mid-transit time
<egg>
bofh: also, with the resonance you could probably define a whole calendrical system based on the transits!
<egg>
whitequark: are there any new котяpics?
<bofh>
oh my gods you're right and it's really tempting to make a trappist-1 calendrical system now :P
<egg>
bofh: let's do it! :-p
UmbralRaptop has quit [Ping timeout: 198 seconds]
UmbralRaptor has joined #kspacademia
<egg>
let me dig up a recent paper
<egg>
bofh: btw did you see that thing on cosmic censorship I linked yesterday
<Majiir>
egg, hello
<egg>
o/ Majiir
<egg>
!wpn Majiir
* Qboid
gives Majiir a polarized prediction which strongly resembles a minion
<iximeow>
!wpn Majiir
* Qboid
gives Majiir a rotary dust which strongly resembles a div
<Majiir>
Are you still at Google, egg?
<egg>
yeah, still getting paid to stare at number formatting at youtube
<Majiir>
I always knew you had a future in numerics
<egg>
also date formatting
<egg>
also pluralization
<Majiir>
Are you happy?
<egg>
also teaching UX writers about the grammars of 80 languages
<bofh>
egg: yea
<egg>
Majiir: I get to write emails in latin at work, so that's fun :-p
<bofh>
I'm not a black hole person so I'm not sure I properly appreciate it
<Majiir>
egg, that does sound fun
<Majiir>
Maybe I'll start opening with _salve_ from now on
<bofh>
Majiir: egg: quidquid Latine dictum sit, altum videtur. :P
<egg>
(context: I didn't know enough russian to write examples of declension of names making translation problematic, so I conjured up some latin UI strings instead)
<egg>
bofh: I think it didn't :-p
<bofh>
egg: touché :P
<egg>
bofh: hm the resonances are far from eggsact, so that would probably be lunisolar-but-much-worse
<egg>
the 8/5 is pretty close
<bofh>
can you really call them resonances in that case?