look at that......... physics.stackexchange.com/questions/230763/…

someone favorited it lol

@alarge @DanielSank Does this cause you guys to cry blood as well?

looks like brainfuck to me , the crazy programming language

It's a massiva, 82 line, regular expression to parse email addresses

i dont really understand this stuff

so u mean i can type that in the terminal on linux and that parses thru email addresses like search for patterns or what?

So, a regex is like algebra to characters

what that thing does is that it checks if an email is valid or not

without using a lexer or parser but using solely regex

and that's just evil and mad

ah i see

@BernardMeurer ಠ_ಠ

Blood?

@BernardMeurer No blooed coming out of my eyes as far as I can tell.

I'm mostly just mildly annoyed :D

@DanielSank got a minor headache from that, as well as stopped to pray for the soul of who wrote it

hahahaha

Generated code (which that is) rarely looks good. You're suppose to read the input to the generater, not the output.

@dmckee You know what's funny? Typescript generates pretty reasonable javascript.

@DanielSank Exceptions to everything, I guess.

True.

Microsoft makes some pretty awesome stuff.

I used to rag on them all the time, but as much as I don't like working on Windows, Microsoft's other products are generally rather awesome.

See .NET for example.

@DanielSank They have some fine bits and pieces, but they try very hard to borg you as soon as you use one (thought I admit they don't seem quite as aggressive that way as they used to).

My school has sold it's soul for hardware, so all the programming here is taught in c#.

I've been learning little bits of it to be able to communicate with the students about programming: not a bad language aside from the 30-line "Hello, world!" problem that Java also has.

@dmckee Yes, that's true. If you play their game though the experience is reportedly rather nice.

Now that there is a .NET runtime for Linux I am much more inclined to try it out for real.

@dmckee I was pleased to discover that they provide a preconfigured compiler for python 2.7.

At least they're trying.

@DanielSank That is handy.

My wife has a windows box because (a) Civ V (b) it'll talk to her insulin pump and continuous glucose monitor and (c) money.

I find Windows unbearable which has kept me from experimenting with .NET

Still annoying after all these years, but rarely tearing your hair out bad anymore.

Civ 5 wokrs on linux

@DanielSank Should I just remove "open" whenever I say "subset" or "set"

it's the only reason I'm still sane

And a certain amount of the annoyance is down to expectation: I expect things to be organized the mac or unix way and they're not.

@0celo7 Just ping me when you want me to go through a draft.

I think "open" is known by most folks though.

@dmckee This.

@DanielSank They know what open topological sets are but not Hausdorff?

Windows is not Unix. If you play the Windows game, it makes sense (so I'm told).

@0celo7 I didn't say that.

::gnashes teeth::

@0celo7 I think most physicists when they hear "open" know that this means "don't include the boundary".

You just said open is known by most people

@0celo7 Yeah, then you added "topological".

@DanielSank Ooh! Ooh! I do.

But one step deeper into topology and I am hopelessly lost.

^ SEE @0celo7!!!

This is what I keep trying to tell you.

Closed sets? Do you know what that is @dmckee

You have to get a good feeling for where your audience's boundaries are.

@DanielSank He doesn't count

@0celo7 God help you, son.

Is it a bad thing that I have a laptop dedicated entirely to having h-bar open?

Not open, I suppose. So I get to include the boundary. But I don't understand the implications of the difference so it's rote memory.

You aren't even old enough for me to be your son, be quiet.

@BernardMeurer "Bad"... no. Interesting... yes.

@dmckee You're gonna have to define boundary then :)

@0celo7 NO HE IS NOT.

That's your job.

If you want to write a post explaining something, you do not get to demand that your readers define stuff and justify their internal picture of the words you're using.

That is 100% your job to get right.

Of course, you can't account for people just misunderstanding things, unless it's a common misunderstanding in which case you should take steps to put the reader on the right track.

Found this old macbook laying around, made it the bar-machine

Well, he can write a post targeted at a mathematical physics audience, but if he does that on a non [mathematical-physics] question it's going to be at a disadvantage with many reader. Like me.

@dmckee We talked about this. He says he wants to target a general physics audience.

General physics who want to learn more about GR

But serious @0celo7 once you pick your audience you have to accommodate the style to them. That's pretty much rule one.

@dmckee Yes.

@0celo7 Right, but that could be, oh I don't know, someone like.... me. I can follow discussions of curves, charts, and tensors if the terms are defined, but I have absolutely no idea what Hausdorff means.

I think a lot of people are in that boat.

As I said:

With physicists, assume zero knowledge but infinite intelligence.

^ That is very good advice.

Ive been at pains to learn what the typical junior at my dinkly little department can actually handle for the last few years. The answer is slightly depressing, but also slightly hopeful.

I have to start pretty low and slow, but I can push hard and they will work up to expectations.

I even had a student tell me that she resented that at first but now she wanted to thank me. So I'm doing something right.

@DanielSank So...I have to state the vector space axioms too?

I have to define a linear operator?

@0celo7 For my juniors, yes (they've probably seen it before, but they won't recall until you prompt them). For me, no. Who is your audience?

@0celo7 Probably not.

I think most folks know roughly what a vector space is by sophomore year.

Also, stating axioms tends to be useless in my experience. It's better to focus on explaining the properties of things in their respective roles.

I think people who are reading a post about index notation in GR know linear algebra.

That's where I draw the line.

@dmckee Honestly, just anyone who is confused by abstract index notation

This can be anyone from the average joe who asks on PSE to @NeuroFuzzy to @ACuriousMind (<-- he'll deny it, but he didn't get it originally)

Is 0 a natural number?

Depends. Many say no, some say yes.

^

@0celo7 BTW I'd definitely read it once you post it

for how much work you're putting in though you might as well turn it into a book and sell it :p

I'm not putting that much work into it :P

I'm thinking about it while doing other things

@NeuroFuzzy Is it reasonable that I assume the reader knows some linear algebra?

I also have to assume the reader knows some Riemannian geometry, I want to show some applications.

From a historical perspective, I think it makes more sense to consider zero a whole number; zero wasn't really considered a number until some work was done in India in the middle part of the first millennium, I think by Aryabhata (although there was some investigation of it before hand). Therefore, it's not really "natural".

Oh! @HDE226868!

@HDE226868 Do you know what the following mean: (i) Vector space (abstract definition) (ii) Linear operator (iii) Hausdorff (iv) basis of a vector space

Never mind, it was Brahmagupta. And I'll dispute all claims that the Babylonians used zero; it was only a placeholder. See hsm.stackexchange.com/questions/402/discovery-of-zero for a basic history and hsm.stackexchange.com/questions/276/… for a discussion of the Babylonian usage.

@0celo7 I think so, yeah, absolutely. "What is a tensor", "what is a differentiable manifold", etc., are all important questions but like

@0celo7 Possibly, yes, no, yes.

@HDE226868 Possibly? I'm intrigued...pls give it your best shot

you should be addressing someone who is already familiar with either abstract indexless notation, ($T=A\otimes B$ or whatever) or totally index-full notation ($\nabla_b T^a:=T^a,b+\Gamma^a_b c T^c$)

IMO

I'm addressing someone who is familiar with $\nabla_\mu T^\nu=\cdots$

My goal is to teach what $\nabla_a T^b$ means

@0celo7 "Possibly" as in I could recognize it and understand it if I saw it. I could not come up with a coherent, concise version right now, especially, as I literally just popped in to make a historical remark about zero.

@HDE226868 Protip: if $K$ is a field, then a vector space $V$ over $K$ is a $K$-module :D

@GBeau good luck tomorrow!

@0celo7 If you know a resource that gives the proof, you should cite that instead. If you don't, well, I'm not sure what to do.

@ACuriousMind Well, I wanted the proof that an embedded submanifold's tangent space agrees with the intrinsically defined one...but the only place I can find that in is Arnold, where it's an exercise.

I'm sure I can dig it up somewhere.

The reason no one gives that proof is because it is boring :P

Yeah, but we did it once :D

I have the solution written down right here

Yes - but you learn much more doing it than reading it

Exactly...I did it.

But I want to include the fact in my post.

Maybe it's in Sharpe's book, I dunno.

If you can figure it out on your own, then everything's fine - if you can't, you should do it yourself. So just state in your post "one can show that"

I know that phrase is universally hated, but sometimes it has its uses

So that's a thing that's "easy enough" that the hated phrase applies?

@0celo7 Unless you say something about line integrals, the Biot-Savart law, history, or maybe one or two other things, my brain will not process it.

@HDE226868 Line integrals?

What about them

@DanielSank What a hopelessly optimistic view :P

@ACuriousMind Why do I have to install a DirectX and C++ thing for every new game I DL?

@0celo7 Green's theorem and fun with conservative vector fields were part of my Calc III review today. Line integrals.

@0celo7 Because they all want to make sure it is there and so they install it themselves. Strictly speaking, if you have the newest versions, you would not have to install it.

@HDE226868 Calc 3 prof at my school is famous for defining conservative vector fields as those which are not liberal and then ending the class. (he then defines them properly the next class)

@ACuriousMind I just downloaded the 2008 AssCreed.

@0celo7 That's actually annoying. I'm now off to play Dire Straits music while doing magnetic field review problems, which increase productivity by 50%.

@ACuriousMind Am I the only one who finds it strange that guards in AssCreed games can chase you on roofs?

Since when is the average guard a master climber?

@0celo7 Yes, you're the first one ever to notice that.

Huh. It really didn't strike you as strange?

Wow.

@ACuriousMind I really mean that.

The most effective talks start at the beginning and explain the meanings of things carefully, but don't dwell on logical steps which can be made by an intelligent listener.

This is an important skill, unfortunately ignored in most teaching curricula.

It's hard to teach, of course.

It's interesting how often physicists give a talk in which the rate of logical development is highly irregular.

@DanielSank Agreed.

@ACuriousMind I recently went to a workshop on quantum machine learning.

@DanielSank Agreed. You somewhat undercut your own hypothesis there.

@ACuriousMind I did?

That irregular logical development is often a massive problem, IMHO.

(Before we go on, I should say that I think 90% of talks are terrible)

I really mean that word, terrible.

@DanielSank You first said one should assume zero knowledge but infinite intelligence with physicists. Then why would physicists give talks "in which the rate of logical development is highly irregular"?

They disrespect the audience's time by not preparing and offering no consideration for the audience's background.

@ACuriousMind Because they suck.

I recently went to a quantum machine learning conference. The keynote talk was delivered by a rather famous physicist. He started off by saying that he would introduce the important quantum aspects for the members of the audience who were from classical machine learning. He then later said (not making this up) that the difference is that in quantum mechanics you draw a ket symbol and there's a superposition.

@ACuriousMind Did you play AssCreed 1?

I've heard a bunch of colloquia about topological states of matter.

They invariably start talking about some basic quantum mechanics, and then just jump to "...and then in these materials there are edge states which blah blah blah".

Wtf is an edge state?

They launch into all this jargon without explaining what it means. I could probably follow it with some help. They could... I don't know... replace the first ten minutes of their talk where they utter vagaries about the wonder of topology and instead actually deliver to me some technical information relevant to their talk.

Oh god, and the famous "and here's a spectroscopic measurement of the system."

Gee, thanks, Mr. Physics Man, how about even a brief description of wtf that means in your particular system and lab setup?

/rant

You're perfectly summing up my feeling during most "advanced" talks.

@ACuriousMind This really does make me somewhat upset because giving a bad talk is wasting other people's time.

This, to me, is inexcusable. I am baffled by the complacency of the physics community toward this transgression.

There has got to be a good T-shirt that include the line "Gee, thanks, Mr. Physics Man".

Just sayin'.

@dmckee Let's make it.

Custom shirts are easy to make. I'd even send 'em to you guys :D

But we'd need the rest of the shirt...

@ACuriousMind Another favorite is any time a presenter throws up a slide and says "...and don't worry about reading all this..."

We need to red people who do that.

Like... if that's a job talk, you're done. No job. Leave.

@DanielSank Just why the %§§& did they make than slide then, right?

@ACuriousMind More why did they show it to me? If it's a backup, fine.

I was once in a talk where a guy talked about a bump on a graph with 30 bumps

They are consciously wasting my life.

He never really explained why that one bump was so important

AND TELLING ME TO MY FACE THAT THEY'RE WASTING MY LIFE.

@DanielSank ...did you ever tell them that?

@ACuriousMind The times I've known what was going on, they're running through their 30 minute slides for a 20 minute talk.

Which is dumb.

@dmckee yes, people don't practice and time their talks.

It's horrible.

@ACuriousMind I have never told a speaker that in so many words.

Unless they are the last person to go in a meeting that's already a hour over-time. Then it's smart.

I have mentioned to a few grad students that this happens and that they should be conscious of how their presentation relates to their audience.

@DanielSank Maybe you should. I think most are so bad at talks because they never really got negative feedback

@dmckee Sure. I moderated an APS session last year. Gotta be strict with that timer.

@ACuriousMind True, but one gives up a considerable amount of social capital doing so.

This ought to come from advisers and fellow students during a practice session.

@DanielSank One of my collaboration tried giving a time to a very nearly finished grad student for the collaboration meetings.

It didn't make them finish on time, but they also never ran an hour late, either.

@dmckee You mean he tried to have a time limit for their meetings?

For whoever starred that comment: we practice our APS talks intensely. Everyone gives a timed practice, and we absolutely shred their slides. One thing I've noticed is that if you really work hard on your slides you tend to have an easier time giving a good talk.

@DanielSank Well, there was a schedule. It usual said we'd start at 9ish and finish by about 6:30 at night. Who wants to not even head out to dinner until 9?

Except the french, of course.

With great slides you don't have to remember anything, which is nice.

@dmckee True.

@dmckee We have a neat protocol regarding meetings at Google:

1) No agenda, no meeting.

@DanielSank I starred it because I got to see the same practice in action at one of my post-docs. Even for the REU kids. And their talks were (on average) better than the REU students from other disciplines.

2) Meetings end at the scheduled time.

3) If a meeting devolves into a few people discussing something specific, that conversation ends and they have their own meeting dedicated to that topic.

...what's a "REU student"?

research something undergrad?

This is why I make talks part of the upper-division classes I teach. Even if they go to industry these students are going to be giving talks for the rest of their careers, so they have to know how.

Research Experience for Undergrads. A NSF spondsered summer program.

@dmckee Yeah. Giving good talks is such a gateway to success.

It's a chance to show a whole room of people that you're intelligent and do useful things with your life.

I'm very proud of the talks given by grad students here. You can tell they seriously know what they're talking about.

"do useful things with your life" ::sobs::

@ACuriousMind Oh come on... advancing any field is useful!

Hear that?

Crying GDP that will never be...

What's not useful is doing a bunch of work and not communicating it effectively.

@DanielSank I've noticed lately that I cannot explain to others why I want to study (almost) pure mathematics. I cannot even explain why anyone should fund that.

@ACuriousMind Well that is a very tough sell.

Can you explain it to me?

Might help.

Public funds.

@0celo7 Well, to be fair most of the not-immediately-applicable research comes from public funds.

@DanielSank I'm sure if Trump was interested in String Theory that's all the justification he needs to fund it.

Sadly "it's interesting" is not a moral justification for taxation.

@0celo7 That's a tricky question.

@DanielSank There's like 3 ones that aren't.

@0celo7 Three ones what?

@DanielSank Three questions that aren't tricky.

@0celo7 Oh?

That's surprising.

I was recently asked if a food order was "for here" or "to go". I thought I navigated that relatively well and without undue mental exertion.

That's one.

LOLOLOL

Two left :D

thinking

The other day my boss asked if I'd made some figures yet. The answer was a pretty straight forward "no".

Nope.

?

A cosmic ray could have hit you in the head and deleted the memory of you having made the figures.

I'm not sure that illustrates that the question was tricky, so much as it shows that you're willing to entirely abandon Occam's razor.

I'm willing to do many things.

@DanielSank Not really. Because it is fucking awesome! I'm fascinated by the ways in which we encode our descriptions of nature, and how investigating those structures itself leads to interesting new theories

My point stands. I believe I have found 2 non-tricky questions?

@ACuriousMind Ok, so as a tax payer, right now I see little reason to give you money.

@DanielSank No.

Surely you can explain your position in a community of scientists who have taken us from the stone age to sending people to the moon.

@0celo7 Sigh. I'll try again.

Is sending people to the moon engineering or science?

Actually no I give up in light of your cosmic ray comment.

@0celo7 Certainly both.

@DanielSank Good. You see the futility of arguing with me.

There, that wasn't a tricky question! That's two.

@DanielSank It is a very tricky question.

Shut up, meta, you're drunk.

See, it hides the real question.

If PSE were a thing during the 50s, would "how do we get to the moon" be closed as "engineering."

@0celo7 shrugs

Man the python standard lib is awesome.

"Hmm, I need to produce unique pairs of elements from this set..."

import itertools

@DanielSank In that community, I'm the one who maybe finds a very well-hidden way to get us farther than the solar system but most likely will spend his life inventing theories whose experimental tests he will not live to see.

itertools.combinations(my_set, how_many_per_combination)

@ACuriousMind Dude, you just opened a door to massive popular interest.

> get us farther than the solar system

People love space exploration.

Why else would NASA be planning to travel to a planet with nothing on it?

The old "people would pay for that freely so the taxation is justified"?

@0celo7 I mean, that's sort of the definition of justified taxes...

Or am I mistaken?

Oh wait I think I see your point.

Maybe.

May I offer an analogy?

please

Suppose you're hungry and want to go to a sandwich store.

As you're walking in the door, a guy jumps out a bush and puts a gun to your head. "Go in the store or you know what," he says.

He makes no other demands.

Is this moral?

Yeah yeah.

I get it.

Taxes are for stuff that people wouldn't independently buy. I know that.

I'm saying that's irrelevant.

That's half the point of government: do projects that are valuable to the whole, but not valuable for any individual to invest in.

@DanielSank We-elll...I might have a bit exaggerated.

@0celo7 Actually, I think it is relevant.

There is, of course, the practical question "Can a libertarian paradise survive in a world of nation states?"

The answer to which is not really known.

We tax for stuff that we agree we should have, but that if any one person didn't throw into the pot, the situation for them wouldn't diminish.

Also the best test cases live in uneasy collaboration with a state host.

@dmckee I would think not.

Being old and sold-out I just try to wield my pitiful influence in favor of more libertarian instances of the institutions we have.

@DanielSank We tax for stuff that some people think we should have.

I mean, the moment some sub-group pools resources to support an army, they've stopped being libertarian and conquer their neighbors, right?

@0celo7 Oh, indeed. There is certainly not agreement on these things.