Level of questions?

Question

The level of questions (and answers) on Physics is not high enough. I'm not saying this snobbishly, but because if the answers are not high quality and accurate then the website does the community a disservice, contrary to its aims. (Perhaps, in the end, the notion of reputation will remedy the deficits that appear at start-up.)

The remedy, I think, is to attract higher-level users to the site. But this also requires that the community of users start rejecting homework-level and ill-posed questions. Unless this is done, higher-level users will not be drawn to the site. I think Math Overflow does this successfully (at least three Fields medalists figure prominently there) even though there are certainly problems of quality on that site, too. And this much was basically true from day one.

Answer 1

I don't agree and my experience bares that out: You can show your expertise, even when answering the most basic questions. You don't legislate away easy questions as if top-tier experts are just going to start filling the void. Sure, you have to ask great questions. But you also have to consistently provide awesome content with every answer, no matter how simple or mundane. That's how you will gain the respect and recognition of an expert community.

Some of the most highly-respected physicists in the world are teachers; answering Physics 201-level questions from rank-and-file students. Think about it. If you heard a seminar from someone like Einstein or Newton eloquently describing how gravity works when an apple falls to Earth, are you going to say "Nah, that's baby stuff" or do they somehow maintain that go-to source authority when it comes to areas of expertise. Be a great teacher!

In the intervening time between now and when this system is filled with world-class physicists using this site in their daily work… keep in mind that only a small percentage of the content of this site is questions. What you can offer today — right now… with the talent you have right here — are world-class, canonical answers to those questions.

Warning: Boring Anectodote— When I first started using Stack Overflow, I had 25 years of expert programming experience when I asked my first question (link). Without getting too technical, I was asking about, quite literally, one of the most basic concepts of a programming language. Surprisingly, the answer provided some deep, insightful responses I never expected. One was from a Microsoft senior developer on the team that designed the language; The other from someone who literally wrote the book on the subject. Unmatched, expert-level answers to a canonical question; That's the level of response we've come to expect from Stack Overflow.

So when someone asks "why do apples fall down to the earth," provide that world-class, canonical answer which can be improved upon indefinitely. Every time someone searches for an answer to their question, this site should have THE canonical answer on that subject — that, and every subject in physics. When an expert happens across this site, they should say, "Woah! These guys know what they are talking about. This is the site for me!"

Answer 2

I'm not sure how long I'll stick around this site. It's not analogous to MathOverflow, which has a remarkably high level of discourse. Still, this site could be useful for students, or people who have questions arising from popular books or TV shows. My worry, though, is that the average level of knowledge of the participants is quite low. People are accepting inferior answers to their questions, and in some cases better answers are getting voted down in favor of mistaken ones. In particular, just keeping up with all the questions about quantum mechanics and trying to explain the importance of decoherence for a correct understanding of what measurements are and how the classical world arises would take a significant amount of time, and if no one does it the answers will rapidly degenerate to the level of popular books (and even a lot of textbook accounts) that completely muddle the subject and make "measurement" out to be some mystical, poorly-understood, non-quantum process.

Answer 3

My feeling is this site has become analogous to math.se rather than MathOverflow, and there is little hope of transforming it into a research level website. Most of the questions here seem to me to be either at the pop sci level (with correspondingly pop sci level answers) or basic undergraduate physics. I don't think there is much hope of making this site attractive to researchers.

The solution would then perhaps be to create a physics stack exchange which is research level only and strictly enforce this rule. Actually, I suspect a separate stackexchanges may be appropriate for research level in theoretical physics and experimental physics.

UPDATE: TP research-only stack exchange proposal here.

Answer 4

Everyone seems to set Mathoverflow/math.SE as a base example but this is not the whole story; for instance on CrossValidated.com we have both popular and research level questions and both classes gets their answers. Indeed, there wasn't even a discussion about question level.
Same holds for GIS and to some extent for E&R, not to mention StackOverflow.

Answer 5

Stop upvoting questions and answers that you don't think are valuable.

The reason people post low end questions and answers is because they get upvotes.

The fact that I hit the rep cap on the first day with a few half-hearted answers and a bit of googling should be eye opening. Yes, I write in a style that appears authoritative, but I nearly flunked the two theoretical physics classes I had to take to get my engineering degree. Love calculus. Love physics. Just couldn't wrap my head around mixing the two together.

Stop upvoting answers that aren't truly great. That means you have to skip over posts that are interesting if that's all they have going for them.

I have, however, come to the conclusion that research X and practical X aren't simply two different levels of the same thing, they are actually fundamentally different topics. I hope people get the theoretical physics site off the ground, in addition to this one.

But even then, I hope people won't go around voting up every neat question or answer here - even low end questions and answers should have a minimum dress standard.

Voting is the primary feedback loop that tells people what the community should look like. Vote often, but vote carefully.

Answer 6

My position is somewhat intermediate. I am not a research physicist; I'm applying to grad school now, and have only a general physics undergraduate knowledge.

As a result, I tend to ask fairly basic questions, but ones that are not very simple or easy to answer. These are attracting lots of bad answers (and some good ones).

A research-level site would be interesting to me, but only as an observer for the time being. I would learn what sorts of things people are thinking about, but would not be able to participate in the discussion.

On the other hand, a "pop-sci" site is not very useful because there are too many uninformed responses for each good one.

What I'm really hoping to find is an online environment to replace the one I had as an undergrad, where I could go talk to people at roughly the same level (other students) or higher level (TA's and professors) than myself, but with a focus on understanding the nuances in the more fundamental material I'm learning.

Unfortunately, it sounds like neither of the ideas being tossed around - a high-level site or researchers or a low-level site for the general public - are a place that I completely fit in.

Robert Cartaino's model, a site where experts still answer basic questions very well, is what I would like, but it looks like there are practical difficulties with implementing it because the most competent people are largely interested in research-level discussion only.

Answer 7

I totally agree.

While I enjoy answering simple questions and teaching others, I joined physics.stackexchange.com hoping for 'Mathoverflow for physicists'. The proposal says clearly 'Beta Q&A site for active researchers, academics and students' - and we should cling to it.

There are plenty of good textbooks, there is wikipedia*), there is a number of other places. When there is a simple question it is easy to give an answer (it suffices to have any academician or even a good undergraduate/high-school student). Elementary questions often are good, but should go somewhere else (another stackexchange?). However, there is no place when researchers in physics can ask and answer high-end questions.

*) Some articles on wikipedia are too complicated for the beginner? Then edit them. Its much more beneficial than answering a question which has been already answered a dozen times on different fora or usenet groups.

Ah - but you need to invite your labmates, friends from conferences, active professors, etc.

Answer 8

If this site aims at emulating the success of Math Overflow and CS Theory (i.e. providing a research level Q&A site, as opposed to popular/undergrad level questions), it should be kept in mind that both sites from their very start:

a) as a rule, discouraged too basic/popular questions and stayed focused on grad level and above

b) started with a "seed" core of dedicated users (in Math Overflow case, it was a group of grad students and researchers working in algebraic geometry and related fields); experts and "big shots" (Terry Tao etc.) were attracted very quickly (in CS Theory, almost from the first day of beta version!)

As for now, I see neither a clear policy as to the target level nor the seed group of dedicated users. Simply waiting for more users to get high quality material won't work (I think it's not possible to build a critical mass solely with "casual users"). Nobody will post an expert level question if he knows it's going to be swamped by stuff like basic mechanics problems.

Answer 9

Any new thoughts on increasing the number of "research-level questions"?

Answer 10

Open source community has this problem ages ago. They have written comprehensive how to ask question guides. At first I would redirect poor questions to this link, which are in agreement with my opinions. But then I gave it another reading and found that I could revise it to make it less computer directed. Here is my initial effort, I will be slowly working on it. Some endorsement or contribution from the community would be very welcome so it show more legitimacy.

Personally I put about the effort in the question as I saw the asker did on making his question. I would probably put more if I find the question interesting and is not answered elsewhere.

Answer 11

Mathoverflow has math.stackexchange to which to send people. Maybe we could find and publicize a list of more didactic physics sites. I think I will do this with "What are some good didactic physics sites which are more appropriate for elementary misunderstandings?". Any objections to putting this question on physics.stackexchange?

That said, I also enjoy thinking about naive questions which point out my own unstated assumptions. I can see that simple ignorance will soon get tiring, however.

Answer 12

How easy would it be to alter the menu for questions at the top, so people could select how many levels of questions they want to appear?

When people posted, they would have to define the level of the question, with moderators moving questions to other levels if necessary.

Answer 13

Remember this site isn't for you - it's for the shareholders of SE.

The aim is to maximize number of visitors X quality of visitors, to sell to advertisers. Now unless Phys Rev Letters has a higher promotional budget than I think, then a site where 3 people discuss the same paper on loop quantum gravity isn't going to bring in the ad $$$$.

Similarly a site full of UFO and hollow earth cranks, or one full of the same grade school question about paths of cannon balls (the problem with physicsforums) is also going to lack readership pretty soon.

The optimum is probably something like Scientific America/ Discovery channel.

This was also the launch model for SE. The stackoverflow site deals with relatively simple questions, then programmers.se for more 'professional' topics and cstheory for the 3 people in the world that are interested in computer science theory but aren't mathematicians.

I think the aim should be a mid-level site with a future proposal for an physics-research site if this site doesn't work for you. Remember you can use tags to filter out questions that are too simple or out of your field of interest.

Answer 14

My first impression was the opposite: that the level of questions was almost to high. Granted, there are many silly questions, but there are also many daunting research-level questions. The problem is there is almost nothing in between.

There are very few university studies level questions (like what you would get as homework). These I could answer.

There are almost no questions regarding actual day-to-day physics problems, say, of a grad student or a young postdoc. These I could answer, or use.

Then consider this: An good programming question can often be answered in a few minutes, and the answer easily applied. Some of the better physics questions around here - they tend to be hardcore quantum field theory - however seem to require weeks of research to answer, and an afternoon to understand, even for someone who works in a closely related field.

What we need is more people asking real day-to-day advanced questions, that someone who came along this problem before can answer in a few minutes, rather than big research frontier questions.

Answer 15

I completely agree with Robert. I would also point out that our best questions - where best is defined as highest voted - are all interesting questions which are both accessible to laypersons and interesting for specialists to answer.

https://physics.stackexchange.com/questions?sort=votes

If we want to be the reference site for physics we need to be able to communicate, we need to be able to be authoritative by the breadth, depth and quality of our answers, and not merely by limiting the topics, audience and users to some elite few who speak lingo.

In this I do think we are lacking and we could improve massively.

Answer 16

Right, the most basic questions can be rather difficult to answer.

I'd recommend to make age a must-entry when applying for physics SE.