There has been a lot of discussions about the level of the questions and the quality of the questions.

It seems to me that we have a core group of users, that are following Physics SE from a long time, and that this group would prefer to have a Physics SE closer to mathoverflow than to Math SE as it is now.

We also have this new proposal for a theoretical physics SE.

My questions is: what should we do now ? Could we try to move the center of mass of this site more on the "research level" side ? How could we do it ?

Should we close more questions ?

We already have quite a high number of questions, so slowing things down (by closing question) is not a problem I think. In addition we still have time to modify the way things are going, we are still a the begining of the beta.

Another action would be to change the definition of the site to clearly reflect the expected level: is it possible ?

I don't think we should give up now and consider the site "dead".

What's your opinion about that ?


It appears that "research level" is a bit too restrictive: as pointed by @Noldorin, questions related to (under)grad course are interesting as well, my idea is that such an undergrad student would not ask "What is the most efficient way to destroy the universe?" ...

Questions that are not research level but coming from an (under)grad course can really benefit from answers coming from real physics practitioner.

My conclusion: in any way the level, at this moment, is too low and we should try to restrict the range of questions, and maybe even more importantly up vote good answers (which is easier to do if the questions are more interesting).

  • 2
    $\begingroup$ I'd like to think I'm in this "core group" of users, and personally would like to see this site cover the full undergraduate to graduate-level range. The current content is somewhat below that, but I think it's a sensible goal to aim for. $\endgroup$ – Noldorin Nov 14 '10 at 16:05
  • 1
    $\begingroup$ @Noldorin: Maybe I should reformulate to include "full undergraduate to graduate-level range". Obviously I am not upset about your q/a ... but I guess you also agree that more and more questions are out of range. $\endgroup$ – Cedric H. Nov 14 '10 at 18:09
  • 3
    $\begingroup$ @Cedric: As a response to your questions "Could we try to move the center of mass of this site more on the "research level" side ? How could we do it ?", I think that if people want to see 'research level' questions being asked, all what they need to do is start asking those kind of questions. By the way, I don't have such a question at the moment. If I had it, I'd ask it. $\endgroup$ – Robert Smith Nov 14 '10 at 19:12
  • 5
    $\begingroup$ @Cedric: Yes, I very much agree that we are seeing more and more low-quality (even non-sensical?) questions over time. In any case, this site needs clearer and visible guidelines. At the moment people just visit it and think anything vaguely relating to physics is acceptable! $\endgroup$ – Noldorin Nov 14 '10 at 19:32
  • $\begingroup$ @Noldorin: as this is the first SE site I am participating in from the begining, how would you proceed ? Don't we need moderators and how could guidelines made more visible ? I mean, even if it is written in the "about" page that doesn't count as "visible". $\endgroup$ – Cedric H. Nov 14 '10 at 22:45
  • $\begingroup$ @Cedric: That's a good question. I have not followed other SE sites so closely from their inception either, but I know a few things. Regarding moderators, it is about time for them - I asked the question here and the answer was that beta moderators should be chosen around one week into the public beta. $\endgroup$ – Noldorin Nov 14 '10 at 23:02
  • $\begingroup$ (contd.) And regarding guidelines... we need to be sure exactly what we want to say, and then each new user (and exiting users for one time) should be directed to them explicitly when signing up. There's no easy solution though, I admit. $\endgroup$ – Noldorin Nov 14 '10 at 23:04
  • 2
    $\begingroup$ @Noldorin: We should try to formulate the "(under)grad to research level range excluding universe destruction questions" more specifically. As a non native English speaker it is too hard for me :p $\endgroup$ – Cedric H. Nov 14 '10 at 23:16
  • 2
    $\begingroup$ @Noldorin, @Cedric: Can we all agree in: 1.- Re-formulate the intended level for the site. 2.- Close inappropriate questions. 3.- Make good use of the upcoming moderators. For the first point, undergrad level is the minimum allowed. Probably we need help from administration to change that. The second point requires moderators, right?. Also, as mentioned by Peter, we need a proper way to ensure that the best answers are voted as such, although for that we need a strong community which is still growing. $\endgroup$ – Robert Smith Nov 15 '10 at 3:29
  • $\begingroup$ Yeah, that sounds pretty sensible. The start of undergraduate sounds like a good level to start. (Maybe we can allow keen end-of-high-school people who want to go a bit beyond their course too.) $\endgroup$ – Noldorin Nov 15 '10 at 10:14
  • 3
    $\begingroup$ I think the threshold should be a "proof" that the user is interested in an answer like someone actively working or studying physics would provide, not just a "blabla" with keywords, no units, no logic,... $\endgroup$ – Cedric H. Nov 15 '10 at 12:25
  • 1
    $\begingroup$ Maybe we should have a compulsory question on relativistic electrodynamics you must answer before joining the site? :D $\endgroup$ – Noldorin Nov 15 '10 at 14:05
  • 1
    $\begingroup$ @Noldorin: Lol, I know my last comment sounds stupid, but I meant that this "proof" should be visible in the question, the way it is asked, etc. $\endgroup$ – Cedric H. Nov 15 '10 at 14:09
  • $\begingroup$ @Cedric. Yeah, I know what you meant. I was just kidding. ;) $\endgroup$ – Noldorin Nov 15 '10 at 14:25

I don't have enough reputation to post a comment, but I am very disappointed that not only are simple mechanics questions not getting good answers, but the good answers aren't getting many more votes than the bad answers. (For those who care, Tristan's answer to the linked question is the one which is most accurate.) If you don't do something to increase the level of discussion here, this site will be in danger of becoming useless. If good answers don't get more votes than bad answers, I think the site is in serious trouble.

  • $\begingroup$ "If you don't do something to increase the level of discussion here, this site will be in danger of becoming useless". Wow! What do you want us to do? More seriously, I've seen some complaints but nobody seems to have ideas to improve the site. I have asked Joe about it, with no response at all. $\endgroup$ – Robert Smith Nov 14 '10 at 19:08
  • 4
    $\begingroup$ @Robert: I agree with you. At first I suggest we close more questions like this (physics.stackexchange.com/q/770) ... $\endgroup$ – Cedric H. Nov 14 '10 at 19:13
  • 1
    $\begingroup$ I agree with Robert here; we really need to suggestions to improve the quality of answers (and questions here). I think we recognise the problem fairly well already. $\endgroup$ – Noldorin Nov 14 '10 at 19:14
  • 1
    $\begingroup$ @Cedric: Agreed. Even the top-rated answer seems very speculative and airy-fairy. $\endgroup$ – Noldorin Nov 14 '10 at 19:15
  • 4
    $\begingroup$ I would suggest everybody try to upvote as many good answers as they can. I don't think you need to worry quite as much about the questions. In the question I linked to, Tristan's answer was reasonably good, but it was languishing behind two much inferior ones. And none of them had very many votes at all (on the other hand, it's better now that I gave it as an example). $\endgroup$ – Peter Shor Nov 14 '10 at 19:15
  • 5
    $\begingroup$ @Peter: That is probably a good suggestion. Unfortunately, many users are voting on questions and answers of some complexity about which they are clueless. (I always try to support the correct answers, though I generally do nothing if I am not very familiar with the field.) We shall very shortly be electing temporary moderators for this beta site. With any luck this will be of great help to maintaining the quality of this site's content. Let's hope the right people get chosen. $\endgroup$ – Noldorin Nov 14 '10 at 19:27
  • 4
    $\begingroup$ If somebody comes here, asks a question, and the top-ranked answer isn't very useful, they may not come back. Similarly, if somebody comes here, gives a good answer, and nobody appreciates it, they may not come back, either. You want to attract people to this site, especially ones who can give good answers. I think stupid questions are much less of a problem right now (unless they overwhelm the good ones). $\endgroup$ – Peter Shor Nov 14 '10 at 19:28
  • $\begingroup$ @Cedric: I agree (although I appreciate the effort made by Matt Reece to rescue a 'bad' question giving a good answer). $\endgroup$ – Robert Smith Nov 14 '10 at 19:29
  • $\begingroup$ Good, but quite possibly over many users' heads. As an upper-graduate/research-level site, the answer would be very welcome. $\endgroup$ – Noldorin Nov 14 '10 at 19:33
  • 2
    $\begingroup$ @Peter: I think your suggestion is very reasonable (and I follow that as close as I can), however, taking the point made by Noldorin, I'm not comfortable upvoting answers when I'm not sure about their correctness because of my ignorance in that particular field. $\endgroup$ – Robert Smith Nov 14 '10 at 19:38
  • 1
    $\begingroup$ @Peter: I agree with you, but "unless they overwhelm the good ones", I think this is a (major) problem too. $\endgroup$ – Cedric H. Nov 14 '10 at 19:57

Although I am rather new to the site, I already took quite a liking to it. Nevertheless, the level is certainly low. Personally, I'd like to see the main target of the site being standard university physics courses level (i.e. undergradutate/graduate) which would put it somewhere between MO and math.SE (as Tobias suggests) instead of somewhere between high school and undergraduate (which is IMHO where math.SE is).

Now, this certainly doesn't mean that I want to prohibit other kinds of questions. But some level should certainly be enforced (by moderators and the core group of users) in order to get a reasonable mean and variance for the distribution of the level of questions.

Now, all of this was written with an implicit assumption that TP.SE takes off and become a second MO. If you think this is an unreasonable assumption and also want to have a place for some research then obviously standards for physics.SE would have to be set a lot higher. In particular, all undergrad-level questions and below should be closed immediately (and this is probably an unreasonable request). Also graduate-level and research-level question should have to be asked a lot (but here the question is: who will ask them?; and more importantly: who will answer them?).

Seeing as the previous doesn't appear to be very realistic, the first paragraph then seems a next best goal with respect to quality. Also, I think this goal is certainly attainable. In particular, I will start asking some graduate-level questions that have been interesting me lately.

Note regarding what Peter Shor was saying about the concept of good/bad answers not really working

I think this might be a problem of communication. If there is an apparently wrong answer, don't just vote but leave a comment that says so. It may be clear to someone knowledgeable that the answer is wrong but other people might not see this (except as a downvote which can be lost if the question has a few upvotes already anyway). With the comment present, this can be resolved very quickly and it also gives a hint to others to be cautious around the answer. In the same spirit leave a comment letting the others know that some answer is good. I know this is not really democratic but this site has to be built by a core of knowledgeable users. Their votes must count more than votes of someone who knows next to nothing about physics.

Second thing I noticed is that OPs don't really care about accepting good answers. But why don't they accept it? Was it (in their opinion) not all that good and/or didn't they understand something? Well, then they should obviously leave a comment addressing the issue. Or do they think that no answer to their question is even coming close? Well, then they should update the question so that everyone can see what's in their head.

Eventually, there must always appear an answer that is good enough to be accepted (given that the community knows enough about the topic and the question doesn't turn out to be an open problem; both of these conditions are usually satisfied). But if OP neither accepts an answer nor communicates then I take this as a complete ignorance of someone's work (especially when the answer is good). Why should anyone bother answering them? More importantly: why did the OP ask the question in the first place if they don't really care about getting an answer? This behavior appears quite worrisome to me. And no, I am not talking about not getting 15 pts of rep, if anyone asked.

  • 2
    $\begingroup$ +1 that makes a lot of sense, certainly the first part where you talk about targeting a university course level. That's what I'd like to see this site become as well... although considering that apparently I'm high in the running to be a temporary moderator, I'd want to know that the community accepts this view before we start enforcing it using mod powers. $\endgroup$ – David Z Nov 18 '10 at 2:40

My intention in creating the other proposal was not to kill off this site. It seems reading through a lot of the definition phase stuff that there was an intention all along to allow questions at the amateur and undergrad levels. I don't have a problem with this, and it is certainly a worthy cause to help answer physics questions in general. As I see it, concerns about the quality of answers aside, this site seems to be heading towards becoming a physics version of math.se.

Now there seems to be (at least) two strong views held by different groups in the community. One group wants a physics version of MathOverflow, while the other wants a site where the scope and level is much more broadly defined. These two goals don't have to be mutually exclusive, but I do think they require separate SE sites. This current site has already fallen into the second role, whether that was intentional or not and I think it is unlikely that it can be changed. After all SE sites are run in a very democratic way, and are setup in such a way that a group within the community will find it very hard to change the direction of the site. That is why I proposed the TP site, to fill the gap left for a research level site.


When I proposed this site my intention was to have a site for researchers, but also for the discussion of thought experiments and other things not related to their own field. These are questions also undergrads should be interested and be able to take part in and that was whom I meant with students. "Basic" questions where intended to go to Popular Natural Sciences. The problem is, pop-sci didn't take off at all, maybe being too broad.

Basically I agree with Robert Cartaino's answer. I didn't actually intend this site to become MO for physics, but also not "only" math.SE (personally, I like math.SE, but I sense some consider it not "pure" enough) but rather somewhere in between. If someone asks why a rainbow is coloured, either answer it professionally, or up-vote the correct answers, or skip that questions. I think first of all we need moderators to assist the community, but we also need everyone of us to actively help improving posts that have the potential to become good but just aren't well-posed right now. If questions are tagged and titled well, it should also be easy to skip questions of a level you don't like (at that time of the day that is...) and find the ones interesting / challenging.

My suggestion:
Ask and answer questions of the level you desire,
vote on those you accept,
ignore those you consider wrong level but are still valid,
and vote to close the really bad ones or help improving them.

  • $\begingroup$ Thanks for your input on this... I think I mainly agree. The problem is that if you diverse too much (multiple sites for different levels) none of them become too active. $\endgroup$ – Noldorin Nov 15 '10 at 14:27

My thought has been that increasing the level of the site is best done by asking questions of an appropriate level. That way, newcomers will see examples of what kinds of questions we expect here and hopefully the expectation value of the question level will rise, without us having to close all that many of the more basic questions (which can still potentially generate good answers).

Of course, I certainly do agree that some questions we get should be closed.

EDIT: I've been trying to put this into practice by asking questions which are representative of what I would like to see on the site, for example:

  • $\begingroup$ Great comment, David. $\endgroup$ – Robert Smith Nov 15 '10 at 4:56
  • $\begingroup$ Thanks. Of course, easier said than done ;-) I've been trying to do this myself, and I've only come up with a few good questions so far, but every little bit helps, I suppose. $\endgroup$ – David Z Nov 15 '10 at 5:03
  • $\begingroup$ Oh, I didn't know you were beginning with that. Maybe we can be more coordinated to create a bigger effect. $\endgroup$ – Robert Smith Nov 15 '10 at 5:26
  • $\begingroup$ Well, I was concentrating most of my effort on answering, especially at the beginning, but I think you might be right, we could perhaps benefit from a coordinated efforts to get more good questions on the site. I'll edit a couple of examples into my answer here as a start. $\endgroup$ – David Z Nov 15 '10 at 5:41
  • $\begingroup$ @David Is the question "seeding"? I.E. do you actually know the answer? It's definitely a good question, but the refined, smooth delivery is not quite what I'd expect if you'd been reading over available explanations struggling to make sense of them. See earlier discussion of seed questions: meta.physics.stackexchange.com/q/49 $\endgroup$ – Mark Eichenlaub Nov 15 '10 at 6:47
  • 4
    $\begingroup$ @Mark: Actually no, that is a legitimate question I have. I've been studying GR for several years so I have a very good sense of what the problem is, but I honestly have never found a proper answer. (A couple of my professors have tried to demonstrate that calculation but they didn't get the correct number, so they basically wound up saying "just trust me, it works." I was never satisfied with that.) $\endgroup$ – David Z Nov 15 '10 at 7:06
  • $\begingroup$ Ah, okay. I was just surprised that such a famous calculation is difficult to find spelled out well. I certainly hope someone answers it clearly, since it'll be a valuable resource. (I'm not competent to do it, unfortunately). $\endgroup$ – Mark Eichenlaub Nov 15 '10 at 7:09
  • $\begingroup$ @Mark: Yeah, I was surprised as well. I figure if we want this site to be the definitive resource on advanced physics, we should have some of those classic calculations on hand so we can draw in other people who are looking for the answers. $\endgroup$ – David Z Nov 15 '10 at 7:16
  • $\begingroup$ "Just trust me, it works" - sounds quite familiar, hah! To be fair though, the actual calculations are fairly long-winded, so it's pretty easy to make a mistake by hand. $\endgroup$ – Noldorin Nov 15 '10 at 21:53
  • $\begingroup$ @Mark, @David, @Noldorin: I myself think the canonical calculation is pretty easy and short (while the paper in the accepted answer is quite long-winded) and also that it's quite believable (in the sense that approximations done are totally reasonable). If you think it would be great to have a complete derivation on this site then I can certainly provide that. Will do when I find a bit more time. But do tell me whether I should bother at all, seeing that another answer is accepted already. $\endgroup$ – Marek Nov 18 '10 at 10:26
  • $\begingroup$ @Marek: As a general rule, I am willing to change the accepted answer if it seems appropriate to do so. $\endgroup$ – David Z Nov 18 '10 at 17:26
  • $\begingroup$ @David: thanks for clearing that up. So I'll try to write the solution in a little while. $\endgroup$ – Marek Nov 18 '10 at 17:53

@ David Z - This is too long for a comment, which is why it looks like a question.

I don't really understand what you are trying to doing by asking your question at How do you calculate the anomalous precession of Mercury? ?

Are you saying that you want a lot of standard textbook references for a standard advanced undergraduate/basic graduate course in relativity? There is no really current research oriented theory question about your question, is there? I mean there might have been 40 years ago when Dicke was asking whether the same precession could have been caused by unmeasured solar oblateness, but not now. Are you asking for references to experimental error bars? Probably not. It just seems made up and sort of hokey, or a question on the history of physics. So why would a researcher be interested, except didactically.

In contrast, a current research oriented question along the same line might be asking about analytics related to gravitational recoil from black hole mergers, or recent evidence supporting GR in binary neutron star systems with very large precession, or whether precession is expected in the galactic center, or something else more modern, a question that requires a researcher to answer properly.

  • $\begingroup$ Current research is not the only focus of the site, though. In fact, even out of the good questions that are currently on the site, very few of them focus on current research. Sure, part of the reason I want to see that question on the site is its historical significance, but I'm sure any research-level physicist who knows how to do it would get a kick out of going through the derivation. (It's not a textbook reference, by the way; at least, I have not seen that particular calculation done in any textbook I have looked at. If you know of one that does it, post an answer!) $\endgroup$ – David Z Nov 15 '10 at 7:13
  • $\begingroup$ I posted this without looking at it too carefully. Hope it is right. mathpages.com/rr/s6-02/6-02.htmi $\endgroup$ – sigoldberg1 Nov 15 '10 at 7:37

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .