« first day (3680 days earlier)      last day (50 days later) » 

4:26 AM
DeepMind is actually insane
Have you seen their HQ, @JohnRennie?
 
4:39 AM
Both the independent variables used while generating the phase space are supposed to be canonical conjugates. Is this statement correct? (I hope yes)
 
 
1 hour later…
6:06 AM
@PrateekMourya this is a conical pendulum. Th ball will move in an oval. Offhand I don't remember if the shape is elliptical or some similr curve.
 
123
6:22 AM
Hi All..
 
 
2 hours later…
 
4 hours later…
12:30 PM
 
1:30 PM
This hadn't occurred to me until now, but in the same way "dirac delta normalised" states in QM necessitate rigged Hilbert spaces, does the same issue extend to qft? Do the momentum eigenstates $|\vec p\rangle$ lie in a "rigged fock space" of sorts?
 
@JMac Do you have experience with Aerospace Engineering or simiar?
Or any Aerospace engineer here?
 
@Charlie Yes
 
ok ty
 
Jim
@RewCie not planes, but spacecraft for me. Still technically called aerospace engineering
 
QFT is all continuous spectrum at the end of the day, even the way the scattering amplitude is set up can be traced back to continuous spectrum normalization
 
2:03 PM
@Jim Can you help designing a rover for Mars?
If you can, can you please give your mail or something to contact you?
 
Jim
@RewCie odd request. I doubt I'd be much help designing a real rover. I've been out of the game for a number of years and their design teams are usually good. If it's just an assignment or something, I can give some tips and consult a bit, but I'm not going to commit to a large role
 
2:26 PM
@Jim Okay... If you ever change your mind to help us design a rocket/launcher/rover, feel free to notify me here.
 
Jim
@RewCie who is "us"?
 
@Jim I'm gathering a startup team for that.
 
Jim
say what?
it costs some serious capital to send a rover to mars
 
No, we are just planning, doesn't mean we are starting it already. We are trying to estimate a cost for it.
I may pull out VCs and bootstrap it later we think we can do it.
or we'll see what more we need for it.
 
Jim
estimating a cost for it is pretty straightforward. Look at what others have done, shave some off in some places, add to others, multiply by 1.2
I'd guess at around $20-30M at minimum
More likely it's around $200M
 
2:30 PM
20-30 Mn$ is not very hard... I guess if it were around 60-70MN $ we can do it, otherwise we'll have to change routes and pivot things...
 
Jim
for spacex of nasa, I could see $20-30M
for a startup it's significantly more
 
Yep...
There are some indirect routes for funding too, but I don't feel that talking about it on the chat would be helpful....
@Jim anyways if you don't mind giving your E-mail for some future reference?
Also any control engineer here?
 
Jim
just start with around Earth. Jumping straight to Mars is insane
 
I never really got it, what's cargo cult science?
 
Jim
@RewCie in today dollars, designing the Saturn V rocket cost around $10B. The remainder of the Apollo program cost around $30B. That's just going to the Moon. There's a reason SpaceX isn't yet at Mars
 
2:53 PM
@Jim Ofcourse we'll start from earth.. smaller project then gradually move to bigger ones...
Curiosity took 2.5 Bn USD... So, betting it afterwards 25 Bn $ seems safer...
 
Cargo cult science is a form of pseudoscience in which an imagined hypothesis is offered after the fact for some observed phenomenon, and further occurrences of the phenomenon are deemed to be proof of the hypothesis. It can be paraphrased as, "It has been observed to work in the past, while other methods have been observed not to work." In contrast with the scientific method, there is no vigorous effort to disprove the hypothesis. The term cargo cult science was first used by physicist Richard Feynman during his 1974 commencement address at the California Institute of Technology.Cargo cults are...
 
Yeah, I've read the wiki. I just don't see how it's related to cargo cultism at all
Cargo cult programming, I get
 
Replace science with programs
 
@RewCie For context, the only missions to Mars that have ever launched were designed and operated by nation-state agencies. There has been exactly one non-nation-state-operated mission to the Moon (the Beresheet lander), which failed to land.
Moreover, Beresheet had a clear business model in that it was competing for the Google Lunar X Prize.
Without a business model (i.e. a way to extract revenue from your rover), talking about VCs is not going to convince anyone.
Unless you have a billionaire in the wings waiting to donate you $100M, of course.
 
3:14 PM
@NiharKarve cultism
the cargo is the goal of the cult
 
@NiharKarve I think Feynman's point is that "cargo cult science" doesn't drop a wrong hypothesis quickly enough, just like the actual cargo cults don't drop the hypothesis that building a runway summons airplanes when none shows up, but rather attribute it to an error in their construction of the runway or some other missing feature
 
Today, cultism is more applicable to political views.
 
but I've always found it an unhelpful analogy, tbh - cargo cult science doesn't grow from misunderstanding "real" science and just going through the motion without understanding what one is doing (like real cargo cults and cargo cult programming), it grows from personal vanity and being afraid to contradict authority even when one has the evidence on one's side
 
@EmilioPisanty Yes, I know about Beresheet Lander, it was an excellent mission, loved it, tho it missed it just by some luck...
Of course without Business Model, Business Plan, Marketing/Scientific research VCs or even other investors won't agree investment.
 
but maybe Feynman misdiagnosed the issue and thought people really just didn't understand how science is supposed to work, i.e. he might have really thought it was similar to the cargo cults in that respect
 
3:24 PM
First step would be establish the company like Lockheed Martin (I'm talking of non military ones ofc)
 
Feynman certainly didn't have much respect for authority or fear of speaking his mind, perhaps he just didn't understand that other people might not find behaving like that as easy as he did :P
 
@skullpatrol Hey pal, I've a doubt.
 
wazzup pal
 
Is there any reason for writing mov edx ecx as MOVL %EDX %ECX
Are MOV and MOVL doing same things here?
 
dunno, sorry
 
3:27 PM
I guess what happens in former case is $edx \leftarrow ecx$
Oh okay.... np....
anyone else?
In case of signed word movement, It works as $edx \stackrel s {\leftarrow} ecx$
I've tested it using as
 
@RewCie movl (or any other mov suffixed with a one-letter length encoding) is an assembly instruction in GAS (GNU Assembler) syntax. There is usually no un-suffixed mov in that syntax (but there are some formats that infer the desired size from the operand size). That is, you're looking at two different formats for your assembly if one has mov and the other has movl.
 
3:42 PM
@ACuriousMind gotcha, mov for 1 instruction movl for one 32 bit word, movq for 64 bit ones.
 
@RewCie you're learning assembler? There is no hope for you! :-)
Back in 1980 I was writing IBM OS360 assembler on an IBM mainframe. I still have the crib sheet somewhere.
 
3:59 PM
@ACuriousMind exactly what I was thinking, thanks
 
@JohnRennie Actually, they have a lot of viruses to disassemble and analyse them...
@JohnRennie I'm using CS:APP, it's good, recommended by college, first course...
 
it seems somewhat controversial to call the states created by the action of the field operators "position eigenstates" in qft
based on what I've read at least
does anyone have any input on whether or not it's generally acceptable to interpret these states this way, I've read a number of posts on pse and there seems to be a fair amount of disagreement :(
the reasoning in the picture seems reasonable enough to me, at least by analogy with regular qm
 
@Charlie it's inacceptable - there is no position operator and hence no "position eigenstate". You can called it a "localized" state or whatever but if you call it a "position eigenstate" I challenge you to show me the operator it is an eigenstate of
 
Ok that definitely makes sense
 
@RewCie aha, found it:
 
4:08 PM
@JohnRennie Wow! That's a whole reference!
 
Why I've kept a useless IBM assembler programing guide for 40 years is a good question. Probably for the reassurance that no other programming language will ever be as bad.
@RewCie that's just a cheat sheet with one line summaries of all the op codes.
 
We also have IBM/370 Asm, from this book
 
@RewCie No! Seriously? You have to learn IBM assembler?
 
@JohnRennie Everything from scratch to modern. We even have a punch card stuff...
Also we have K&R and Stephen G Kochan for ISO C89, C11 and C98 too!
Then we have CLRS book for Algorithms...
 
Well that will be ... erm ... fun :-)
 
I was about to say that looks really interesting :-)
 
@JohnRennie haha... check this book, a pretty beginner's guide to assembly, also that'd make anyone appreciate Art of Assembly
:P
 
Hi all! Does anyone have in mind a reference with the computation of the 2-2 scattering amplitude of a massive scalar field minimally coupled to graviton?
I have some doubts I wish to clarify
 
4:32 PM
does anyone use FORM to calculate the Riemann tensor symbolically?
or is it Mathematica that's popular
 
@NiharKarve I use xCoba
or xAct
 
@newUser is that Mathematica?
 
@NiharKarve to calculate the Riemann tensor from the metric?
 
@NiharKarve Yes, it's a package of Mathemtica
 
4:50 PM
I used GNU Octave once to calculate the Schwarzschild Solution. I'd never do that again!
 
@JohnRennie yes
I was thinking of "rolling my own", seems like a fun little project
 
If that's all you want to do there are loads of notebooks around for doing that. I use one written by one of the guys who used to hang around here. You're welcome to a copy if you want.
 
A physical notebook or a code notebook?
 
A code notebook i.e. a Mathematica notebook
 
Another thing I'd probably include in such a project would be symbolic gamma matrix-spinor trace tech
@JohnRennie Cool, I'd love to see it
 
5:00 PM
@NiharKarve this web site
The file is AFT-Schwarzschild.nb
 
Great, thanks
I can't see it right now because I'm on phone, but I'll check it out later
@JohnRennie To be clear, this calculates the Riemann tensor for a general metric, right?
 
Yes, you just edit it to put in your metric and it calculates all sorts of stuff. The metric, the Christoffel symbols, the Ricci and Weyl tensors, the Ricci scalar and even the Kretschmann scalar.
 
Great, thanks for the link
 
It is extraordinarily useful! :-)
 
5:34 PM
Today, my thinking about introductory physics education has centered on "the normal force". In some of my recent tutoring sessions, when I ask students to identify forces acting on the system in question, they say "normal force" (probably just out of habit) in scenarios where there is no normal force. Some examples are with hanging masses attached to strings, or with objects submerged in a fluid.
I wonder if this comes from professors not really going into detail on what constitutes a "normal force", or if it is a point students just don't tend to really think about / pick up on.
When you really get into it, normal forces can get fairly complicated, and normal forces behave in so many ways depending on the system. I think some students don't realize how different this is, and usually expect it to follow things like "the weight is always mg", "the spring force is always -kx", etc.
 
Jim
6:08 PM
@BioPhysicist a hanging mass on a string has a normal force. Tension
it's not really normal to a surface, but it is the same concept
 
@Jim It depends on what you mean by a "normal force". Typically we just say that is a tension force. Usually "normal force" is reserved for the force between two surfaces that are in contact
 
Jim
fair nuff
 
More specifically, interaction force between two surfaces has a parallel component (friction) and a perpendicular component (normal force).
So yes, a tension force could still be normal to the surface of an object... but I am not sure that constitutes what we usually mean by "normal force".
At that point it more of semantics though I suppose. There is still a force regardless of what we call it
But I am sure the students that I am tutoring are not going through this entire thought process haha
 
6:28 PM
Buoyancy as a "normal force" doesn't seem that different than the typical definition really, except that it's perpendicular to the surface everywhere in contact and not just at a point.
And calling it "normal pressure" seems kinda redundant
I mean really it's pretty much exactly the same. You're taking the force distribution normal to the contact points, and representing it as a single force
 
Yeah, that is something I had been considering as well... but I feel like when students say this in a tutoring session this is not what they mean
The buoyant force arises from the total "normal force" from the fluid I suppose
 
Yeah I get what you mean there. Calling it a normal force when you're first learning physics is probably more a sign of confusion. And yeah that's what I meant about buoyant force, but I don't know how easy that would be to teach to a student just starting physics. I probably wouldn't have thought of it like that until at least university.\
 
It seems like "normal force" is not something that is robustly defined in introductory physics classes, perhaps because it would cause more confusion than it is worth.
Sometimes I wonder if it would be better to get rid of the term entirely though
And just specify an "interaction force" or something like that
Or "force of table on block", etc. (depending on the system)? Although that is more wordy
Then students would be less inclined to just throw in a "normal force" without thinking about it first
And they wouldn't make the rookie mistake of "Normal force always balances the force component opposite of it", or even worse "normal force always balances weight and has a magnitude of mg"
 
 
2 hours later…
8:19 PM
@JohnRennie It collapsed today.
 
 
2 hours later…
10:10 PM
@FadedGiant Wow good thing they noticed it was one failure away from falling apart and people weren't trying to fix it or something. Sucks that it couldn't be salvaged easily though.
 

« first day (3680 days earlier)      last day (50 days later) »