The h Bar - 2016-08-15 (page 5 of 5)

« first day (2111 days earlier) ← previous day next day → last day (2820 days later) »

11:01 PM

Looking in Rudin he defines sin and cos using complex exponentials

Man, I've just realized that String topology is the most difficult topic in mathematics that ever existed and will exist. :/

@PhysicsGuy What is "string topology" supposed to be?

@ACuriousMind en.wikipedia.org/wiki/String_topology

@PhysicsGuy AI develops quite fast, I think it would be a fruitful idea to try to use AI to produce simple or possible simplest models to describe experimental evidence. In the mathematics, the proof of the 4-color theorem was the first significant computer-aided result, I think a similar step could happen also in the physics. Computers are used mainly for "stupid but fast" tasks, i.e. to process many data, I think a step similar to the 4-color theorem will be a significant one

omg spectral sequences and string theory, I knew it

11:10 PM

@peterh What are you talking about ? What does this have to do with this ?

" The point about homology is that it measures defects, or obstructions ... Things in homology are about measuring defects. Snake lemmas, etc, tell you how those defects are related ... The snake lemma tells you that if you were looking at n-dim holes, then you next need to take into account the n-1 dim holes - or the boundaries of the n-dim holes. "
https://www.physicsforums.com/threads/exact-sequences.164213/
spectral sequences anyone?

you mean topological spectra ?

"In this paper, we investigate the behaviour of the Serre spectral sequence with respect to the algebraic structures of string topology in generalized homology theories"

@PhysicsGuy Maybe behind the string theory there is a more simple one. Note the Maxwell-equations started as 20 non-linear using mechanical analogies. Maybe the string theory has also a simplified "core". As I know, the problem of the large number of the false vacua is so hard, because finding our in them leads to the backpacking problem what is NP-hard.

@PhysicsGuy Spectral sequences are a standard tool to compute certain (co)homologies. The origin of the name is apocryphal, no one really knows where it comes from

11:13 PM

@peterh Making computers do things that are like what humans are good at, e.g. pattern recognition, is very, very difficult.

People are working rather hard on it.

My lab is called the "Quantum AI Lab" for a reason :-)

@peterh Why do you think, string theory has a simple core ? It is the second hardest mathematical thing ;)

Man, look at the equations. Not simple at all.

"Spectral sequences are to long exact sequences as iterated extensions are to extensions."

@DanielSank Right! Here's a paper with some funny examples of image recognition gone wrong

@DanielSank Yes, this is exactly what I unfortunately didn't ask as I toke part on the AMA event in readonly mode: using quantum computers to solve unsolvabe QM problems...

Also seems to arise when you want to compute the homology of a group $G$ from a normal subgroup $H$ and $G/H$ using successive approximations somehow

11:15 PM

@DanielSank Will be replaced by the AI he will create

@PhysicsGuy Maywell equations were also hard at the beginning. History of science is an important thing to recognize patterns in the steps of the science. :-) I think human brain + computers for data processing + computers using soft methods > human brain + computers for data processing

@peterh I have no idea what you mean.

@peterh Well, okay, aou're right at that point. For example, the lorentz-attractor has been solved by a computer. I think it couldn't have been solved by a human brain.

But the Maxwell equations are at a well-defined form now.

String theory wont ever be well-defined cause it's way more complicated

@PhysicsGuy If I would be a string theorist, what I would do with this false vacua problem: I would use informed search to find nearly-optimal solutions and then I would hope there is at least one between them which deserves some medallion.

"A spectral sequence is a computational tool to approximate (co)homology groups in homological algebra and algebraic topology. The sequence part in 'spectral sequences' comes from the fact that you take successive approximations to (hopefully) converge to the correct group (so a sequence of approximations in that sense)."

11:20 PM

@peterh That's what string theorists do, isn't it ?

I haven't been able to tell what people are discussing here for a few weeks.

How do you mean that ?

I'm used to dropping in and seeing folks discussing some topic that I don't understand... but lately i can't even tell what the topic is.

It's really weird.

It's not just you

"More concretely, a homology spectral sequence is a sequence of graded chain complexes that provides the higher order corrections to the naïve idea of computing the homology of the total complex Tot(V) •Tot(V)_\bullet of a double complex V •,•V_{\bullet, \bullet}: by first computing those of the vertical differential, then those of the horizontal differential induced on these vertical homology groups (or the other way around).

This simple idea in general does not produce the correct homology groups of Tot(V) •Tot(V)_\bullet, but it does produce a “first-order approximation” to them, in a useful sense. The spectral sequence is the sequence of higher-order corrections that make this naive idea actually work."

11:21 PM

These discussions don't make more sense even if one watches them from the beginning :P

I mean what maniac would even think of doing this

@DanielSank That

So much

@PhysicsGuy I hope :-) As I know, the results happened until now by computers, for example the calculation of the proton mass by its principles, has used more human brain + faster computers. And not soft methods.

@ACuriousMind Especially when people say that electromagnetic waves are ripples in spacetime.

@PhysicsGuy But I think, it will once happen as also AI will play a role to find a result, and I think it will be a significant result.

11:22 PM

@peterh This could be.

@Danu :(

What do?

@PhysicsGuy A certain subfield of string theory - that of string phenomenology - is searching for certain backgrounds that have "nice" phenomenological properties, such as being close to the standard model in the low-energy effective QFT. But that's only one aspect of string theory, there is also considerable effort in exploring more non-perturbative aspects (both in their own right, and to get more hints as to what the suggested M-theory looks like)

Quick, someone say something I understand!

@DanielSank I mentioned earlier today that I'm seriously considering setting up a chat room with moderated access.

@Danu I thought you were suggesting that because of negative tone.

11:24 PM

It's a complicated issue, but I've pretty much lost patience with this chat room

@DanielSank A combination of things

Also, my totally uniformed knee-jerk reaction is to plead with you to not do that because it would fracture the community.

Hmm:
"Remark. One can speculate whether the mystery with the name ’spectral sequence’ can be resolved as follows. The name rather likely comes from its similarities with the Taylor series in Analyis. The Taylor series of a function approximates the function better and better. As an infinite series it is given by its coefficients which one thinks of being a spectrum of numbers for the function. The first few numbers are important. They even have special names: position, velocity and acceleration. The other numbers can also be calculated but in practice not many people care about them. Now, a…

@Danu Please make it readonly and not closed

@Danu Why does everybody start hating this chatroom ?

@PhysicsGuy Two main reasons:

1) People who have no f-----g idea what they're talking about show up and start spouting off nonsense. Unfortunately, these people tend to talk more than people who actually do know what they're talking about.

11:25 PM

@DanielSank I know

This is the most serious problem with it

@PhysicsGuy Because, honestly speaking, it's a lot less nice than it used to me (insert hurr durr old people talk shit about young people all the time)

2) Some people come here and just spread discord: they make ridiculous negative accusations about the moderators and other users, complain about site policies without having enough experience to judge whether or not what they're saying is a good idea, and generally just piss around without thinking about what they're saying.

Again, this is only a real problem when such people talk too much, which is often.

@DanielSank Ok, I understand. And your plan is to open a new chat, or what ?

@PhysicsGuy No. I never proposed a new chat and right now I think that would be a step in the wrong direction.

@Ocelo7 How does your Ricci-Flow flow ?

@PhysicsGuy: Why do you always have a space before your punctuation?

11:29 PM

@DanielSank the idea actually came to me while during the past weeks, which I mostly spent in the math chat room

I've wondered that with a number of posters, but you're the first one in chat I can actually ask.

@ACuriousMind Yes please answer that.

It's so, so weird.

@ACuriousMind I don't know. I never thought about that.

That room also has some people that are excellent contributors, but they get driven away/drowned out by random crap

@Danu Who was driven away?

11:30 PM

@DanielSank Mike Miller, Ted Shifrin

For the record (again) I do have an ever-growing ignore list and I've found the chat room considerably more enjoyable and usable since I started using it.

Basically the two top contributors as far as I'm concerned

Ted Shifrin used to come here? Or to math?

@PhysicsGuy How do you mean, "you never thought about that"? At some point you must have decided to press the space bar before typing a question mark, no?

(most knowledgeable people in geometry/topology stuff that I'm interested in)

@DanielSank math chat

11:31 PM

@Danu Awwww man! Shifrin is a boss!

@DanielSank Seeing how the room got ruined for them made me seriously think about this option

Here there is a similar thing going on now

@Danu Did they give a reason?

Like, actually from their own mouths?

Yeah

@ACuriousMind No, it just seems to be a habit.

I asked one of the two; (chat) names were mentioned.

11:33 PM

@Danu They left because of particular users?

That's plain stupid. Use the ignore feature.

Right?

What am I missing?

I disagree

As ACM said, when users start really dominating chat, there is not really any avoiding it

Also, when they drown out everything interesting it's just not worth sifting through the (replies to) crap

@Danu Ok I have a suggestion.

@DanielSank The SE is a high-level controlled structure, like a big company. The high level control is financially interested to make its task well, thus they are working professional. As the SE was new, they were also new, but with time their experiences (and statistical data) has improved. My low-level impression is that they are working pragmatically with increasing understand. My other low-level impression is that a dual solution would be a better idea as killed site.

@DanielSank If I ignore everyone who annoys me + everyone who converses with them, then the chat effectively becomes dead. If I only ignore those who annoy me, I still see the reactions to them. Ignore only works well when currently no one is paying attention to the users you're ignoring

I really, really think a moderated-access chat is a very bad idea.

I think we can fix the problem you mentioned simply by constructing topic-specific rooms.

11:35 PM

And honestly, I don't really care what drove those users away. What matters, to me, is that I've lost two really, really significant teachers

@ACuriousMind Exactly

On Mathemaics S.E. there are several chat rooms for each topic (Analysis, diff geometry, so on). Maybe that's a good idea.

In any case, the fact of the matter is that I'm now no longer able to effectively learn any mathematics through SE, for instance

@DanielSank Too small user base here on physics.SE chat to sustain that, imo

^

You'd need at least 3-4 regulars per room, I'd say

Then it would be a nice idea

But currently, splitting into topic-specific rooms will just create a lot of rooms that are mostly inactive

11:38 PM

like on maths S.E.

@ACuriousMind Right, so if you ignore me because I'm talking to XYZ, what do you think I'm going to do?

I'm going to get the message and stop talking to XYZ.

But the important things like "Electrodynamics", "Relativity" would always be active, I think.

@DanielSank It would be also possible to have a preliminary moderation: there would be a list of accounts capable to write anything, the messages of the others would be appear only after they passed some controlling mechanism, for example the were voted in by the list.

@DanielSank I'd swear you're new here :P

@Danu wut?

11:39 PM

@PhysicsGuy Completely disagree

@peterh If anything, raise the rep limit for chat.

I do not like this idea of "voting people in" at all.

@Danu Why ?

I do not want to come here if users have to be voted in by the oligarchy.

I would leave.

@DanielSank I don't know how many people you have on "ignore", but there are still lots of response to John Duffield, for instance.

I think.

@Danu I find ignoring a relatively low number of people has made the room very usable.

(Until more recently when the topics became so esoteric that I have no idea wtf is going on)

(however, that's not a real problem because if I wanna talk about something I start a new conversation and it's all fine)

11:41 PM

@DanielSank Yes, it would surely lead to power misuses, but also the stories of the beginning of the MO weren't dream stories

@DanielSank I concur

@peterh I don't see how MO is relevant.

For the record, if I'd set up something like this I'd rather set up a "vote out" than a "vote in" system

i.e. if you fuck up chat too much you're out

@peterh MO doesn't "vote people in", they have the same SE model as every other site, they just have a more limited scope compared to math.SE. What are you talking about?

Still a super slippery slope

But the current situation is not really acceptable to me.

11:43 PM

Jesus christ, getting awfully authoritarian, how about just relax no?

@Danu ooof I dunno dude.

@ACuriousMind The problems what @DanielSank explained, don't exist there. They solved these problems, this makes them on my opinion, at least considerable examples.

I really, really don't know.

Oh, don't worry

I already got the message; this is not going to happen

@DanielSank Power misuses don't really happen against HQ members, everybody knows their value very well.

11:44 PM

@Danu You certainly can make such a room and see what happens.

NEh

Good evening

@peterh What? Can you point to where they "solved" "these problems"? I was under the impression MO doesn't really use chat all that much, except for that "Homotopy Theory" room

If it turns into a social lounge for the users who all like each other that's great, but then what happens when a n00b shows up?

with you and ACM out there are just 3-5 "essential" members left

most of which also will be like meh

11:44 PM

"Oh look, an exclusive room where the bosses chat. I guess this site is full of itself. Thanks, k, bye."

Whoa whoa what's happening

@0celo7 The adults are discussing the future of the site.

@DanielSank Oh, I'm not worried about that actually

lol

@0celo7 Just read the chat log, it's there for a reason

11:45 PM

@ACuriousMind No, they solved it on the Q&A. What they did on the Q&A, could be done at least on the chat of the PSE.

@DanielSank Oh come on

Ideally it'd be something like this chat ~1.5 years ago

@ACuriousMind ...how do I do that

@Ocelo7 How is your Rici-flow problem going on ?

Pretty much perfectly welcoming to people who don't know anything

11:45 PM

@Danu Ah, hence the "vote out"?

Of course

Anyone who means well would be welcome

Well heck, can't we use flags and suspension for that?

lol

@Danu You don't think everyone means well?

"flags and suspensions" was not very effective

@DanielSank Neh

11:46 PM

Consider that some people may just be inept without being malicious.

@peterh What did they "solve" on the Q&A? We are currently discussing the issues this specific chat room has, I don't see what that has to do with the Q&A moderation on MO. Be more specific about what you mean

@ACuriousMind You know about spectral sequences?

@Danu there are a lot of people in the world who are simply too lazy to not be annoying.

Lots.

@DanielSank I think they wouldn't cause any serious hassle if there weren't any malicious people around

@0celo7 I...have heard of them and even used them, but only on a very superficial level.

11:47 PM

@Danu I dunno. Look at this famous article.

Just think of how much effort went into it.

@ACuriousMind So you won't be able to help when I get to them in Bott & Tu?

It's all to help the lazy people be less lazy.

My prof has already stopped helping me with that book

If people could learn from observing their surroundings, such an article wouldn't exist.

@0celo7 I now get linear dependence and independence

Linear combinations, span, and basis vectors | Essence of linear algebra, chapter 2

11:48 PM

@BernardMeurer congrats

@BernardMeurer Good, that's a really important concept.

@DanielSank I'm not sure what I'm supposed to get out of this

I actually am liking this series of videos, they aren't crap

Oh, I see your non-reply-to-me messages now

I still think this is not really relevant

@ACuriousMind More specifically: they have a relative high limit on the minimal understand of their topic. A similar limit would also solve the current problems. Yes, I am now talking against myself, but it is only the first approximation.

11:49 PM

A chat room is very different from an SE site

@Danu I'm doubtful of obtaining access to your moderated chatroom

Would I be excluded?

You should read the god damn transcript and you'd know :P

@PhysicsGuy Haven't done any lately

@peterh The problem is currently not that people with no knowledge of physics are allowed to chat.

^

11:50 PM

@Danu Reading right now!

@Danu Oh, just that someone spent a lot of time trying to explain how to ask questions well and integrate with online communities of experts in a technical field.

The point is just that a lot of people don't know how to do that, and I'm implying that it's out of incompetence, not malicious intent.

@DanielSank Right---I am pretty close to 100% certain that newbies asking random questions is not an issue in our chat room

The problem are the two kinds of users that Daniel characterized at the beginning - the clueless who think they know everything and the disgruntled with an agenda

And I'm not sure that that carries over to people not knowing how to act in a chatroom in a reasonable way

@Danu Didn't say newbies.

11:51 PM

@DanielSank Sure

@ACuriousMind No, but all of the people with them you have problems, are all non-professionals. -> filtering out the unwanted non-professionals would solve the problem

Lazy people, whatever

@ACuriousMind Correct. However I'd add a third class: the non-newbies who also don't know what they're talking about.

There are the relatively well informed but still wrong people.

@Danu Ok, it's not going to happen

But if it did happen...

Newbies who are clueless are fine. Newbies who are clueless and think they know everything are sort of ok. Newbies who won't listen are a huge problem.

11:53 PM

@peterh That is throwing out the baby with the bathwater. This chatroom should be a place where a curious person can come and ask/talk about physics. I don't want an elitist clique where only "professional" topics can be discussed, whatever that means.

I think I know everything, but when demonstrably wrong I reconsider my views.

@ACuriousMind agreed

I want n00bs to come here.

Desperately.

@DanielSank But doesn't this chatroom exist for discussing topics ? Then it doesn't matter if people don't have fully knowledge, they are here to learn from the other users of the chat.

My favorite conversations are with n00bs who listen.

^I think that's hyperbolic :P

@Danu It really isn't.

I love the thrill of "aha" coming from someone who really wants to learn.

11:54 PM

@Danu Who are the two users who drove out Mike/Ted

Paging @JohnRennie too.

@0celo7 The thing about Bott Tu is everybody says it's great and intuitive, then you open it up

Obviously, the most enjoyable conversations here are unrelated to physics ^^

@Danu -_-

@bolbteppa BT is OK

11:54 PM

@0celo7 Two users?

@DanielSank ghehe

I'm so funny I know

@Danu You said two names popped up when you asked them

@ACuriousMind Then I don't know what to do, but I am happy to read this.

@peterh I think everyone agrees, dude.

@0celo7 No I didn't

This isn't news.

11:55 PM

I said two users were driven away (Mike, Ted)

Oh, I misread.

But you said "named were mentioned"

Sure, and I'm not going to tell you which

@Danu good man

@0celo7 Implying he doesn't want to actually say those names. Subtext, it isn't that hard!

When 0celo7 starts pestering you, just don't engage.

11:56 PM

^so hard

Although very easy in this case

@bolbteppa The problem with Bott & Tu is that it's antiencyclopedic.

You have to hunt for answers

And they prove things in the discussion

It's written like a physics book at times

@Danu Much more as 2

@Danu What had they needed to remain?

And they never clearly state important things, such as the fact that bump forms with volume 1 are cohomologous

@peterh The absence of certain types of "noise"

My problem with it is that it isn't written like a physics book, but it's supposed to be :p

11:58 PM

They just didn't like the way the chat room was headed

They allude to it on $\Bbb R^n$, but somehow expect it to be obvious on general manifolds

Sooooooo...after this lively discussion, we're no wiser what to actually do about what annoys us, right?

Just like I don't like the way this chatroom is headed currently.

I'm still not sure how it works on general manifolds, actually

@ACuriousMind At least you've rejected my (most obvious) suggestion

11:59 PM

I think you have to use a careful partition of unit argument

@ACuriousMind Use the ignore feature.

^also rejected :P

False.

@DanielSank You've had me on ignore off and on

You haven't actually tried it.

@0celo7 And?

11:59 PM

@DanielSank I have, though.

Why

« first day (2111 days earlier) ← previous day next day → last day (2820 days later) »

Transcript for

Aug '1615

The h Bar

General chat for Physics SE (physics.stackexchange.com). For M...

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile