The h Bar

2:32 AM

Hi guys I was wondering if I can get some opinion on the comment section of physics.stackexchange.com/a/127997/44205. Do you think: 1. The post, as accused, is not clear and should have clarified that a limit is something that can be saturated or not saturated. 2. The commenter is trolling 3. the commenter doesn't have a good grasp of the English language

alemi

I think its clear, but apparently not everyone does. I'd add a qualification to your point #5 along the lines of "Thus for a theoretically optimal computer...." or something like that.

Kyle Kanos

I am a native English speaker, so I should hope that I have a good grasp of the English language

alemi

As written, it is pretty easy to read it as a rather bold assertion.

Especially when you consider some of the questions the site gets each day, I don't think clarifying the statement would hurt at all

The linked paper itself uses the phrase " "ultimate computer" ".

bechira

2:50 AM

Okay judging by these I was expecting a little too much from the reader, I apologize.

David Z

3:18 AM

@bechira don't think of it as though you did something wrong in your answer, just that you could do better. Most (all?) answers can be improved in some way, so you shouldn't take it as a shortcoming in your ability to write a good answer.

In any case I don't think anyone was trolling, it was just a misunderstanding backed up by some forceful arguments.

alemi

3:31 AM

People spend a lot of time discussing what does and doesn't make a good question, especially on meta. Would anyone be willing to suggest and example of a truly stellar question?

David Z

3:47 AM

@alemi well you can look at most of the questions I've asked as examples of what I think makes a pretty good question... but stellar? I can't think of one offhand. (They're certainly out there)

alemi

@DavidZ, so, for instance, I really like the airplane question. I think its interesting, hits on some interesting physics and can certainly be relevant to others, but can it be said that you "thoroughly searched for an answer" before asking?

I guess that's my real problem, I don't really know what kind of burden people have in mind when they are asking questions to demonstrate previous effort at working towards a solution.

I think that is what keeps me from asking questions. Either the kinds of questions I come across in my daily research, I don't think would be useful to anyone else, or for the more "fun" questions, I either know the answer or have a great deal I could say about them, and would either have to play the game of writing the question pretending I don't know the answer and posting the answer myself, or have the question be a page long which would ruin the fun of the question itself.

or, take for example this question on fourier transforms that just popped up. I answered it, but didn't upvote the question itself yet because I can't decide if it is a good question. It's certainly a question I know I had as a student, and wished that I could have found a simple explanation of. But the asker didn't demonstrate any previous effort.

Though, if he had, I would honestly have had a more negative reaction to it than in its current form. Currently its short and sweet and to the point and generally applicable.

David Z

4:31 AM

@alemi I actually did do some searching, but the question certainly doesn't show it. So you're right, that wouldn't be the best question as written. (To be fair, that was asked very early in the site's lifetime, before we had a sense of what constituted proper prior research.)

Although, maybe I shouldn't quite say that. I envision the prior research requirement as an implicit thing, i.e. the questioner doesn't necessarily have to say that they did prior research, but if someone types a few key words into Google and finds the answer right away, then it's time to downvote

I usually point to this question of mine as an example of what a good specific-reference question should look like

Maybe this question is the closest I've come to asking a really good question?

alemi

That certainly demonstrates previous effort. But, to be honest, I like the airplane question a lot more (than the MV one).

The perihelion one is very good

David Z

@alemi I suppose that's reasonable because everyone knows about airplanes :-) whereas very few people know about MV initial conditions.

34

Q: Do Maxwell's Equations overdetermine the electric and magnetic fields?

Maxwell's equations specify two vector and two scalar (differential) equations. That implies 8 components in the equations. But between vector fields $\vec{E}=(E_x,E_y,E_z)$ and $\vec{B}=(B_x,B_y,B_z)$, there are only 6 unknowns. So we have 8 equations for 6 unknowns. Why isn't this a problem? A...

That seems like a pretty good question

alemi

So, what about this one. The author clearly demonstrates some previous effort, but I think everyone can agree is a bad question. He seems to be lacking some basic background knowledge, and the presentation is botched. But strictly going by the guidelines, it would seem to fit the criterion.

David Z

Which guidelines are those exactly?

I mean, we have guidelines for what constitutes prior research, guidelines for what is on topic, guidelines for how to ask a homework-like question, but no guidelines for what constitutes a good question AFAIK

alemi

physics.stackexchange.com/help/how-to-ask

I'll agree the maxwell one is very nice as well.

David Z

4:42 AM

@alemi ah, well I'm pretty sure that's not meant to be a complete checklist for how to ask a good question ;-)

It's just a list of things to do to avoid the most common ways of asking a bad question

but questions can be bad for other reasons

alemi

It is titled "How do I ask a good question?" ;-)

David Z

Yes, but then it says "To improve your chances of getting an answer, here are some tips:"

It doesn't say "if you do these things, your question will be good"

I don't dispute that the wording of that page could be improved a bit, but we don't have any control over that

alemi

I know, I know. I'm just trying to get a sense of the communities feelings here. You've been here a while, just looking for some wisdom

It just seems like there is a discrepancy between what people say constitutes a good question when they put in close votes, versus when they are talking on meta, versus what the FAQs say, etc

David Z

That's fair

It definitely is the nature of SE sites that there are unwritten "rules" and guidelines

alemi

I mean when you mouse over the upvote button it says "This question shows research effort; it is useful and clear"

David Z

4:46 AM

Especially when it comes to upvoting and downvoting, it really comes down to people's preferences

alemi

Which always freaks me out a little, as it puts a strong emphasis on prior work, but when you look at the most highly voted stuff and the questions people usually praise, it seems as though that isn't the most important thing

David Z

and then we write meta posts and close reasons etc. to try to codify the preferences that people have already indicated with their votes

@alemi yeah, also keep in mind that the same description is used across many different SE sites. Some places it fits better than others.

I mean, there's no way to truly describe what causes a question to get upvoted in a short blurb of text like that.

For the research effort thing, I think the mere fact that a question doesn't have an answer which is easily findable through Google is a way of showing research effort. It doesn't have to say in the question itself "I looked here and here and here..."

alemi

Cool. Just to probe you a little more... where would you fall on the fourier transform question? Good or bad? and why?

Thanks for this by the way, I've found it very helpful. In general I think I'm going to stop worrying as much about what is or isn't truly good and go with my gut a little more.

David Z

@alemi IMO it's not great, but it's okay. I wouldn't downvote it. I think the OP could have showed a little more effort to work through it themselves, and that would make me like it a bit more, but honestly a question on that topic will never be a great question because it's relatively simple math.

alemi

Also, out of curiosity. The last couple stackexchange podcasts have talked about the apparent "fall from grace" of the main stackoverflow website, saying lots of people have seen its quality go down in recent times. Has the physics SE shown similar evolution?

David Z

4:53 AM

@alemi sounds good. Especially when it comes to upvotes and downvotes.

@alemi probably... I think it's a natural consequence of the growth in popularity of a community that its quality goes down a bit. But I'm okay with the level of quality we have now. I seem to recall some meta questions about this, if you care to go look them up.

Stack Overflow gets a lot more crappy questions than we do ;-)

alemi

I've poked around quite a bit on the meta, as I wanted to get a sense of the community as a new user, but that's what led to this current discussion, trying to sort out some of the discrepancies I was seeing in how people talked about questions, I just wanted to see some examples, you know? Again, thanks for the engaging and frank discussion.

Awal Garg

5:38 AM

@FlipFlapFlop also, the material of the balls matters. If they are made of a conducting material like copper, current would induce them as they would be accelerating under earth's gravity and magnetic field. the induction of current would oppose its motion due to lenz's law. Since density of the two balls is different, there would be different amount of current generation, and thus different opposition to acceleration. So this answer is wrong in that context. But true if the balls are perfectly made of wood or something... — Awal Garg 58 secs ago

^do I make some sense?

Awal Garg

6:16 AM

@Bernhard ping... (as we discussed on main).

user54412

4:29 PM

@alemi Terry Bollinger tends to ask very deep, probing questions that show how much thought he's put into them. AlanSE too.

user54412

You may also want to check out this question, especially AlanSE's answer:

user54412

32

Q: Question self-destruction: why don't experts ask more questions?

I've been kicking around this site for a little while now and I've realized that I'm very rarely even tempted to post a question. I'm an astronomy grad student with a background in physics, so naturally as I go about my life and work questions relevant to physics occur to me; it's one of my main ...

discussion questions

Kyle Kanos

8:02 PM

Gah! Effing cluster man...

Code runs fine last week, now that I need to restart the files I get nothing but stupid error messages

HYDT_dmxu_poll_wait_for_event (./tools/demux/demux_poll.c:77): callback returned error status WTF does that mean?

Kyle

@KyleKanos it means HPC environments are a pain in the ass :(

Kyle Kanos

Tell me about it

Kyle

8:19 PM

The best (worst) issue I ever ran into was having a job crash after several hours, right when it went to write a checkpoint to disk. Problem persisted for a couple more tries, eventually I contacted support who told me "sometimes things crash, you should make sure to checkpoint your code" which was infuriating since I crashed on a checkpoint, repeatedly. Eventually it turned out there was an issue with their disk allocation code and my checkpoint write was being assigned to a full disk.

Kyle Kanos

After it happened twice, I added a backtrace to see where it was occurring. Something is up with my IO, it seems, because it is failing during write (level 2 of 6 AMR levels)

I've added even more debug parameters to see if there's some sort of pointer error.

Kyle Kanos

8:40 PM

:D

0

A: How would wormhole-based FTL violate causality?

Two preliminaries: (1) Wormholes are a type of curved spacetime. Therefore there can't be any complete analysis of this problem in special relativity. You need general relativity. (2) The basic logic is not that causality violation implies FTL, it's that FTL implies causality violation. There a...

Transcript for