The h Bar

Pigeon

00:12

Anyone know of a program for germanium gamma spectroscopy available online?

dmckee

@Pigeon What do you mean by "for"?

Do you have a spectrum and want to analyze it?

Most of the detector manufactures ship something with their product, but they are generally only so good. As a young post-doc I ended up rolling my own when I needed to do something moderately sophisticated.

Because the actual gain tends to be slightly non-linear and the background subtraction is difficult to do well and you couldn't know the scale of the actual errors without knowing everything that went on inside the code.

Nor was I the only one who did this...there were three code bases floating around the building each optimized for slightly different measurements.

Kyle Kanos

01:09

This FSU vs UO football game is pretty epic.

hwlau

02:18

Do they really spend 2 trillion on R&D

dmckee

02:47

@hwlau Which "they" did you have in mind? That seems pretty reasonable for global spending (keeping in mind that the "D" is a big cost in industry).

hwlau

gov?

so you mean the company are also counted?

dmckee

I don't image that governments spend that much. Hundreds of billions, maybe.

@hwlau All I know is that the figure says "Global investments".

hwlau

yup, it seems too much to me.

but if it also counts the technology sector by Intel, whatever

it seems reasonable

dmckee

To put the investment scale into perspective the US dept. of Energy spends on order of a billion dollars a year just on "the intensity frontier". There are two other frontiers in the DOE scheme and that doesn't cover NSF or the biomed people.

hwlau

ya, but it is trillion. A thousand of billions.

We need to figure out a thousand of such projects

DanielSank

03:23

@KyleKanos: My SO might be able to answer your bountied question. I mentioned it to her and she seemed interested. If all goes according to plan this will suck her in as a recurring user!
* maniacal laughter

Sabre Tooth

@KyleKanos I could probably join that site and ask about the sun spotter (how to make part, as opposed to the the optics one here)

Nick

04:16

I have a quick QM question that I should really know the answer to -- having a brain fart here. Any linear combination of solutions to HΨ = EΨ is also a solution, correct? So any linear combination of energy eigenstates is also a solution. So assume you put the Hilbert space in a basis of the orthonormal energy eigenstates. Now any function can be represented as a linear combination of basis functions, right? But not every function is a solution to HΨ = EΨ. So where am I messing up here?

DanielSank

05:00

@SabreTooth Yeah, probably.

@Nick Your first statement is wrong.

Suppose $H|A\rangle = E_A |A\rangle$.

Also suppose $H|B\rangle = E_B|B\rangle$, and that $E_A \neq E_B$.

There is no $E_{A+B}$ such that $H(|A\rangle + |B\rangle) = E_{A+B}(|A\rangle + |B\rangle)$

^ Actually there might be cases where that works but not in general.
...thinking...

David Z

@DanielSank your argument there makes sense to me

DanielSank

@DavidZ: Right, but there are some trivial cases like $E_A = 0$, etc.

Other than those cases you get $H(|A\rangle + |B\rangle) = E_A|A\rangle + E_B|B\rangle$

and then if you try to factor out something as an eigenvalue you get

David Z

Even if $E_A = 0$ though, assuming $E_B \neq 0$, you get $H(\lvert A\rangle + \lvert B\rangle = E_B \lvert B\rangle$

DanielSank

$E_A(|A\rangle + (E_B/E_A)|B\rangle)$ which is not a scalar times the original vector.

Nick

Okay, thanks that makes sense. Then where was I getting that linear combinations of solutions to Schrodinger's equation are also solutions? Am I just misremembering something...

DanielSank

05:06

@DavidZ good point

David Z

@Nick you might be mixing up the time-independent and time-dependent versions?

Any arbitrary linear combination of solutions to the time-independent equation - that is, any arbitrary wavefunction - can be the initial condition to the time-dependent equation

Nick

Hmm... I bet that's it. Thanks.

DanielSank

But being an initial condition isn't the same thing as being a solution.

Actually, this is a vocabulary game in a sense.

David Z

@DanielSank yeah, I'm being imprecise in order to avoid explaining all those details

DanielSank

Ah.

@DavidZ: Is there a way to customize mathjax?

My web-programming fu isn't very strong and I haven't figured out how it works yet.

David Z

05:09

Customize in what sense?

I mean, yes, it is technically possible, but it may require a lot of Javascript hacking depending on what you want to do

DanielSank

For example, in real life I type \ket{\Psi} a lot.

It would just be nice to have some physicsy macros.

David Z

Ah, you can do things like \newcommand I believe

DanielSank

But, where?

David Z

Just between dollar signs. $\newcommand\ket[1]{\lvert #1\rangle} \ket{\psi}$ becomes $\newcommand\ket[1]{\lvert #1\rangle} \ket{\psi}$

hm... testing: $\ket{A}$

DanielSank

$\newcommand\ket[1]{| #1 \rangle}$

$\ket{A}$

Oh snap!

It lasts from entry to entry!!

David Z

05:11

yeah, so the \newcommand only needs to appear once in any given page. Since I defined it, in this case, it also works for your messages and so on.

DanielSank

Sick!

I can just copy paste my macros.tex file here!!

Groovy.

David Z

well one wouldn't want to define too many macros ;-)

Just \ket and \bra should be fine, but a large TeX macro file might cause MathJax to become very slow

DanielSank

Just a few:
\ket
\braket{A}{B} -> $\langle A | B \rangle$
\bbraket{A}{B}{C} -> $\langle A | B | C \rangle$

@DavidZ Oh

David Z

And I'm still holding out home that we can get Dirac notation wrapped up into an official MathJax package that can be enabled on the main site

"large" meaning probably O(hundreds) or more

DanielSank

^ If I can lend my programming fu to that I am hereby volunteering.

David Z

05:15

Dirac notation isn't a matter of programming-fu - the implementation can be easily adapted from the braket LaTeX package. It's a matter of convincing either the MathJax or SE developers to include the module in a script which is loaded on the site, and then convincing SE to enable it.

Implementing units in the style of the siunitx package might take some more programming-fu. That package does some disturbingly complicated things with TeX.

DanielSank

Can I lend my pestering-the-right-person-fu?

David Z

If I find anything out, I will let you know who to pester

DanielSank

k

Thanks for caring about this. Good tex environment a happy scientist makes.

David Z

I guess, to be fair, the Dirac notation package for MathJax doesn't actually exist yet. Everyone's probably assumed that there's no way it would get enabled, and thus nobody bothered to actually make it.

So if you wanted to create it, it couldn't hurt.

though I'm not sure if @ManishEarth has already done something along those lines... it would be like him

DanielSank

So... you said O(hundreds) of macros. What are you thinking?

I can think of three.

Oh, I misunderstood. Never mind.

David Z

05:18

oh ok

DanielSank

Could you explain something:

Why does having a TeX package help? Does mathjax send the content to a server which renders it into unicode or something?

David Z

No, MathJax is all Javascript. It runs entirely in your browser.

DanielSank

Ah.

David Z

I was referring to the braket LaTeX package because that package has already solved the problem of automatically sized Dirac brackets, for LaTeX. And I believe the implementation is fairly simple, and could be roughly "translated" into MathJax.

DanielSank

Looking at the js behind the "start chatjax" link, I have no idea what it's doing :P

I will investigate this further and if/when I have some idea what's going on I will try to get something useful done.

David Z

05:21

hehe :P I can't claim to fully understand it myself

that is probably the sort of thing one should understand if one is going to add a package to MathJax

DanielSank

^ That.

Oh I see. It's pulling a (probably much larger) script from another URL.

David Z

Yeah, it loads MathJax from another URL, configures it, and then sets a delayed callback to render the math markup

DanielSank

Yep. I guess I should just email the guy who owns that web page.

Have you already done that?

David Z

What for?

about Dirac notation you mean?

DanielSank

About appropriate channels to add packages/macros.

I'm guessing someone who bothered to make that page knows more than me, and might actually want to help.

David Z

05:28

He's a moderator on Mathematics and probably too busy for such things

In any case, he's not the one you'd need to talk to about adding Dirac notation. You'd have to talk to the actual MathJax developers.

DanielSank

@DavidZ Ah.

Oh, ok, just realized mathjax is a real project.

Know what to do now.

Thanks.

David Z

mathjax.org

a very large real project

DanielSank

Indeed.

David Z

Some information about writing third-party MathJax extensions

docs.mathjax.org/en/latest/options/ThirdParty.html

and a tutorial on extension-writing: docs.mathjax.org/en/latest/extension-writing.html

DanielSank

$\newcommand\braket[2]{\langle #1 | #2 \rangle}$

@DavidZ: Could you type \braket{a}{b} and see what happens?

David Z

05:37

sure: $\braket{a}{b}$

DanielSank

Did it render for you?

David Z

yep

DanielSank

I'm guessing no

WHAT?!

David Z

why wouldn't it?

DanielSank

How did your browser know to do that. Oh, I guess when it rendered my chat message your local js environment processed the command.

Cute.

David Z

05:38

yeah

DanielSank

very cute

Oh man, this is an incredible opportunity for pranks.

David Z

:-P

DanielSank

It is not good that I know this.

lol

David Z

hm, now I wonder if there's a way to use MathJax to invoke other arbitrary Javascript? (there shouldn't be, if the security is done properly, but it wouldn't entirely surprise me if someone missed thinking about that)

DanielSank

I was just thinking that too.

Ok real question: It looks like what we should do is make a "Physics.SE" extension and put it up on the ChatJax extensions repository.

Yes/no?

David Z

05:44

I'm all for it, except instead of making a physics.SE extension it would be more useful to make specific extensions for the features that physicists use

DanielSank

@DavidZ Ok, that's why I'm asking this.

David Z

i.e. an extension for braket notation, another one for units, etc.

DanielSank

@DavidZ: Indeed, but then for ease of use I would make an umbrella package which just imports those.

Yes?

David Z

There are already several extensions incorporated into MathJax itself, such as mhchem which lets you render chemical symbols. They use it on Chemistry

@DanielSank that would probably be something to do for your own personal use

DanielSank

Perhaps I don't understand how this works yet. I will investigate further.

David Z

05:45

I mean, not that I would stop you, but in my experience it seems to be more generally useful to break things down by feature, when you're sharing with other people

DanielSank

@DavidZ Yes, of course.

I think I don't understand how specific pages control which extensions they use. I need to read more.

David Z

so you certainly could make an umbrella package which imports both braket and units and so on, but I just think it's kind of an unnecessary design decision. But that's just me.

If it does get implemented in any form (umbrella package or not), I wouldn't complain.

DanielSank

@DavidZ I only mentioned the umbrella thing because I don't understand how this works. I agree with you about the design in general.

David Z

ah okay

well, it's not like I have time to do any of this myself, so I don't really get a say in it anyway

DanielSank

@DavidZ Ooooh, don't know how I feel about that kind of reasoning.

I write a lot of software tools at work. Step #1 is ask other people how they would use it, etc.

DanielSank

06:40

This statement is meant to gauge the possible agreement among other users:
I think the quantum-interpretations tag should be removed. "Interpretations" are not falsifiable, and therefore are not science and not physics. Any question which might attract this tag is either not a science question or simply doesn't need that tag.

...and now I see that this is impossible and doesn't make sense given how SE works.
Never mind.

ManishEarth