Recently some of our (high rep) professional users have either suspended their activities or deleted their account. For example :

  • Ben Crowell (Deleted)

  • dmckee --- ex-moderator kitten♦ (suspended his activity)

  • Luboš Motl (last posted in 2018)

  • etc.

Looking at some old post I found that in the early stage of development our Physics.SE community lost many users. For example this post talks about it:

There are many other meta posts that relate to such past issues though I believe that it would not be appropriate to post them here without any genuine reason.

I believe that if this continues our community may (again) lose much of our critical mass of professional users.

So in this post I would like to know:

  • What actions we can take, as a community, to increase the critical mass of our professional users?

  • What actions can I take, as an individual, to increase the critical mass of the professional users?

I am asking this because I am a bit selfish. I want that at the time when I (if ever) turn into a physicist there exist some channel to communicate and interact with other physicists. Much like how MathOverflow acts like an interaction channel for mathematicians (and they have Terry Tao too!)

Note that I am not implying anything, in anyway, about the quality/quantity of the current users on Physics.SE.

  • 4
    $\begingroup$ For what it's worth, I don't think we can realistically expect this site to fill a similar role to MathOverflow. In the early days of the site we considered whether we wanted to restrict our audience to experts, as MO did, or to be inclusive of questions at all levels, and as I recall, the community consensus was neutral-to-in favor of the latter option, and generally against the first (although it was certainly far from unanimous). $\endgroup$ – David Z Feb 21 '20 at 9:19
  • 4
    $\begingroup$ The recent departures/stand-downs were a result of issues (and drama) within the broader community (see meta.stackexchange.com). Going forward, how the company treats the communities will be more important than anything we can do internally for some time at least. $\endgroup$ – Jon Custer Feb 21 '20 at 14:01
  • 4
    $\begingroup$ @JohanLiebert - the problem is, that issue is the giant elephant in the room. If the direction of SE moves even further to be 'more welcoming' (e.g. don't delete homework questions, you should answer them nicely), then even more folks will depart. $\endgroup$ – Jon Custer Feb 21 '20 at 15:39
  • 2
    $\begingroup$ @G.Smith - fair enough - I think the point still stands that there is still drama playing out at a higher level in the network, and that still has chances to impact high-rep users and moderators here. Should (when, I'm an optimist) things clear up, that would be a good time to try and go looking for more folks. $\endgroup$ – Jon Custer Feb 21 '20 at 18:23
  • 2
    $\begingroup$ See this thread for a summary of how we got to where we are. To be honest, I don't think that this site could evolve into a role equivalent to MathOverflow, and I am pessimistic about the chances of a such a site ever appearing. (But I would love to be proven wrong, and I would love to contribute to credible attempts to build such sites.) $\endgroup$ – Emilio Pisanty Feb 22 '20 at 15:41
  • 2
    $\begingroup$ @JonCuster I really don't get the impression that the current SE-wide drama has impacted this community all that much. I mostly feel that we do have an issue to work through in terms of expert retention and question quality, and that this is (mostly) orthogonal to the overall SE policy issues. So let's focus on what we can fix here, and treat the SE-wide situation as an independent issue. $\endgroup$ – Emilio Pisanty Feb 22 '20 at 15:56
  • 2
    $\begingroup$ (... unless, that is, Stack Overflow Inc. has actually decided to completely stop caring about the broader SE network, its growth, and its users, which I suspect might well be the case, and in which case there's not that much point in even trying, I guess.) $\endgroup$ – Emilio Pisanty Feb 22 '20 at 15:57
  • 3
    $\begingroup$ The simple answer is that the level of the questions is too low, so the main purpose for a professional to post is popularization. Most professionals don't want to allocate much of their scarce time to popularize physics, and those that do often have more rewarding ways to do it. $\endgroup$ – knzhou Feb 23 '20 at 3:29
  • 12
    $\begingroup$ There seems to me to be a common failure mode where people with deep physics knowledge show up here, answer some questions well, but then caught up in deeply unpleasant exchanges, like arguing with crackpots with pet theories of everything, getting voted down in favor of a vague pop-sci level answer by a user who doesn't have even undergraduate-level knowledge, or getting pestered by hundreds of questions about the very basics of Newtonian mechanics that are better answered in standard books. $\endgroup$ – knzhou Feb 23 '20 at 3:32
  • 2
    $\begingroup$ This seems to be a common enough experience that when I mention to professionals that I spend time on this site, they think I must be a masochist, or at least excessively patient. When I talked to a high-rep professional ex-user, who left years ago because he thought the site was going downhill, he was amazed that the site still had any good questions at all. There is, but it takes effort and care to avoid ending up in a tarpit. $\endgroup$ – knzhou Feb 23 '20 at 3:41
  • 2
    $\begingroup$ Also, I see that the quality of answers has dropped considerably. More people are posting unclear questions thinking that this is a homework site. We should make the rules stricter. $\endgroup$ – Shishir Maharana Feb 23 '20 at 11:00
  • 2
    $\begingroup$ @ShishirMaharana I don't think stricter rules will help with homework questions being posted. There are many places on meta and the help center that discuss the homework policy. Users who post off topic homework questions either don't take the time to learn the rules, or don't care about the rules. Therefore, adding more/harsher rules really won't do anything. Instead of a rule change, what you want is more and stricter users in the close question queue for homework and low quality posts. $\endgroup$ – BioPhysicist Feb 23 '20 at 14:09
  • 2
    $\begingroup$ @ JohanLiebert @PM2Ring I'd caution against taking that type of anecdotal evidence too seriously. A certain amount of turnover is to be expected, at all levels of the expertise ladder, and it is not by itself a sign of an unhealthy site. (And certainly, the expectation of having Nobel laureates around isn't necessarily a good bar to compare the site against.) There are 'expert' users, at both postgraduate and professional level, signing on, they're just less visible. $\endgroup$ – Emilio Pisanty Feb 23 '20 at 15:46
  • 3
    $\begingroup$ @JohanLiebert I don't know whether advertising will work, or what sort of campaign we should try if we do. The internet is very different from the internet of ten years ago (OK, 9, rounded up), this site is no longer the three-month-old public-beta youngling it was when Sean Carroll wrote that post, and the overall SE-wide concerns have definitely put a dent in how comfortable I am with recommending joining to people I know, at least for now. (Sorry for the pessimism, though. It's good to see fresh faces care so much about the site!) $\endgroup$ – Emilio Pisanty Feb 23 '20 at 16:14
  • 5
    $\begingroup$ I strongly disagree with placing high-friction barriers to entry, particularly on answers -- I would argue that if the intention is discouraging low-quality question and encouraging high-quality users, the effect will backfire. Imagine telling a colleague "Oh, yeah, it's a really fun site, we really need people like you! you just need to sign up a week in advance, read twenty random posts, and then you can start looking for things to answer". $\endgroup$ – Emilio Pisanty Feb 24 '20 at 14:47

This is my view as a long-time user, not as a moderator. It's only my perspective and doesn't represent the views of the other mods.

BLUF: We need to grow organically, systematically, and create the environment in which that happens. We need to find the diamonds in the rough.

I think the challenge in these conversations we've had over the years is that we don't really have a consistent definition of what a professional physicist is exactly. Our site is targeted, in part, towards them, but what does it really mean? Is it only research-focused people in "physics?" Is it professors of physics? How many professional physicists are there, really, in the world? And would most physics-oriented people who aren't professors or research scientists in national labs/universities identify as physicists, or would they identify as an engineer or something else more applied?

We often make comparisons to StackOverflow, which has huge popularity and targets professional programmers. And they are held up as the gold standard for how things should work -- there's experts in everything there and you can get some really fantastic answers from famous people in the field. That seems to be what people are looking for here also. But I think we face a lot of different challenges that limit how well that can happen. I'll use StackOverflow and programmers in my discussion, but the same holds for IT admins and ServerFault/SuperUser, etc..


If I am a professional programmer and an absolute beast of an expert in, say, Python -- I will absolutely have used StackOverflow before and I may hang out answering questions because it's fun for awhile. But the nature of the programming field means that I may need to pick up some new tool that I don't know. Maybe I need to build something in Rust, or I have to help out on some web front end project and I need to learn that stack quickly. So I will probably ask some questions outside of my beast-mode area. So I'll be engaged everywhere on the site.

Programmers have a professional need to adapt to new technologies and new areas, and so even if they are the world's best in X, they may be a total beginner in Y. StackOverflow offers engagement on many levels.

Let's contrast that with physics. If I just won the Nobel prize for my contributions to condensed matter physics, it's unlikely that I will need to start asking questions about basic fluid dynamics to get spun up on my next project. And so if I was a user of the site, odds are pretty good that I would only be answering questions in my subject area and have little reason to be asking new questions or learning about things outside of my subject area. My only incentive to learn things outside my expertise is intellectual curiosity, rather than a professional need to keep my employment.

So our type of engagement is different than sites like StackOverflow.


When you're trying to write some code to do something and you get stuck, what's the first thing you do? For me, I head to my favorite search engine and type in some keywords, or I paste in the core error message and hit search. And 9 times out of 10, StackOverflow is in the first few results. And I can poke around and refine the search based on what I see there and I will likely find my answer -- usually on StackOverflow, but even if not, the Q&A has helped refine the keywords I need to search for so I can find what I want. This is true whether I am just starting out learning how to "print('Hello World')" or if I am paid to write software for a living.

It's impossible to use the internet to look for help with programming and not end up on StackOverflow.

Now let's look at what happens when I have a physics question. If you're a new student or a person who has never really studied physics, you do the same thing -- hop on your favorite search engine, type up some keywords or whatever your question is, and hit enter. And you probably end up coming across Physics.SE, although maybe not as prominently and frequently as in the programming example.

However -- this is 100% not how "professional physicists" search for information! Most professionals in the field, and related fields like engineering etc., will be using specialized search engines designed to return scientific information that has been peer-reviewed. If I'm using Google Scholar, there is a ~0% chance I will find something on Physics.SE. The only way I could come across it is if some journal/arXiv paper mentioned it in the citation or acknowledgement list.

So we aren't really discoverable by "professionals" looking for physics.

Tool vs product

When I go to StackOverflow and ask my question about programming, I am asking for help on how to do something very specific. I need to know how to read an XML file. Or I need help on optimizing a particular loop. I need to figure out what the 5000 line compilation error message is saying about my template class. In other words, I am asking about a tool or a building block, not asking about my final product. I'm not asking about my entire software stack, just a tiny, bite-sized chunk of it.

StackOverflow is largely self-contained and managable questions.

But physics is a little bit different. Most of the time, my bite-sized chunks and my tools I need to make my bigger product aren't physics tools! I may need to know when two operators commute (math), or how to write a for loop (programming), or how to optimize a code to run over 10,000 cores (scientific computing), or how to sample my experimental design space to reduce uncertainty (statistics), or how to write up my equations (latex). None of my tools to build my final product are physics specifically.

So I won't really be here to ask questions that help me in my professional life -- I will be elsewhere on the network.


If you look at the number of professional programmers in the world, it is huge. And they don't get together in person terribly often because it isn't a community that does conferences and seminars and the like. So StackOverflow is in many ways their community. There's so many unknown and untapped people who can help them there.

But in the physics world, it's different. Young folks may not realize this because they don't see inside that world. Earlier when I talked about engagement, I touched on specialization -- I don't need to go learn some new area frequently. And that means I know my area very well, and by extension, I know the community very well.

If I have a question, I know exactly who to ask. And I will see them in person at a conference at least once a year. I travel to conferences all over the world and I would recognize every living giant of my field. And I know the vast majority of professors/active researchers in my field. If I know it will be awhile before I see them, then I will send an email or shoot them a text. And if I don't know somebody personally, I know somebody who can make the introduction. So I don't need this community to learn things directly related to my research, I go right to the source.

What is physics?

I've alluded to this throughout, but it's an important point -- what is physics, really? If you ask a researcher in physics, you'll get a very very different answer than if you ask somebody who is in high school taking a "physics" class. To the high school student, physics is adding vectors and drawing free-body diagrams. But for a researcher, those topics probably wouldn't even be in the top 500 things they use to define physics.

Most of what the beginners are stuck on isn't physics, it's problem solving. Many times, we could replace the "physics" part of the vector addition with some other word problem and it would be the same problem to solve.

And if we go back to discoverability, who is most likely to find out this website exists? The students who think physics == adding vectors. And we see this in the many, many Newtonian mechanics questions we get at the start of every school semester.

How do we grow?

I just listed what I see as the systematic challenges we have here. How do we address them? What can we do to get world-famous scientists involved?

I think first and foremost, we probably won't get today's giants of the fields. There's so many challenges to overcome and they probably aren't going to be interested in the site as it is now, for all the reasons (plus some) above.

In my view, we need to be making sure the giants of tomorrow find us. We need to make sure that the team discovering the major breakthrough 10 years from now has grad students who asked their early questions about physics here. Those folks who hop onto Google and say "what is F in F=ma" and end up on our site. If we look at a lot of our power users now, the folks who are active professors and researchers, they started out on the site years ago as grad students and stuck around. They grew up with the site and continue to contribute to it. But this growth is slow and organic, it can't be forced.

It can't be forced, but it can be fostered. We need to be supportive without compromising quality. We can't let the 50 "plz I need this tmw" questions cause us to lose patience and snap at the 51st, because that 51st might be somebody who goes on to be a physics great as they learn and mature. We have to be polite but firm. But we also cannot allow any of those questions to stick around.

We have to make this an environment that is attractive for others to not just come to ask a question or to answer a random question. It needs to be a place where people come to engage with like-minded people between conferences and those in-person meetings. And that means it needs to be friendly and respectful. Anybody who stumbles upon a place that looks rude, or immature, or annoying, isn't likely to go "Yeah! This is where I want to spend my free time!" It needs to be a place that people want to engage with, not just a place for transactions (i.e. come to ask a question and get an answer, then leave when completed).

I think back to the best instructors I have ever had, and there is a pretty clear trend. They may have been friendly, or they may have been really gruff and unfriendly, but they took their work seriously and they figured out how to motivate me to learn. I could ask my question but they would almost never answer it. Instead, they would help me learn the tools and the building blocks I needed to answer the question. That's what we need to do here as well -- particularly with the huge number of very basic questions. We cannot cast those users off into the internet oblivion, but we also can't hand them the answer. We need to point them to the Q&A here that gives them the tools they need. And we need to help them refine their own toolset.

Ideally, some of those folks will go on to be the next greats. And they will have gone through their journey there with the people here.

  • 3
    $\begingroup$ If I could put a bounty on a meta answer I would do it here. I think this is excellent insight. $\endgroup$ – BioPhysicist Feb 25 '20 at 12:37
  • 1
    $\begingroup$ @AaronStevens Thanks -- but even if you could, I wouldn't want it... I've already got all the powers after all ;) $\endgroup$ – tpg2114 Feb 25 '20 at 12:58
  • 1
    $\begingroup$ Not on this post you don't: This is my view as a long-time user, not as a moderator. $\endgroup$ – BioPhysicist Feb 25 '20 at 13:03
  • 1
    $\begingroup$ "We need to point them to the Q&A here that gives them the tools they need. And we need to help them refine their own toolset." So would it be good if we create a community wiki post which contains some of the most helpful posts regarding a given topic so that in future we can point these out to such users? $\endgroup$ – user249968 Feb 25 '20 at 13:49
  • $\begingroup$ @JohanLiebert That's certainly part of it, and we can do that using the tag faq on the main site. We've discussed it in general here on our Meta, and people will float proposals for new canonical CW Q&A here also. $\endgroup$ – tpg2114 Feb 25 '20 at 13:53
  • $\begingroup$ It's one way to work the problem. I think it's still undecided if it's the optimal way to do it though. $\endgroup$ – tpg2114 Feb 25 '20 at 13:53
  • $\begingroup$ what is F in F=ma would be closed as a homework-style question (and rightly so IMO). $\endgroup$ – ZeroTheHero Feb 25 '20 at 14:46
  • 1
    $\begingroup$ @ZeroTheHero It absolutely would -- but that's how people end up on our site. They type those types of really basic, vague things into their search engine and get here. We need to work out how to capture that audience, without sacrificing quality and without driving them away such that the next time they have a slightly-less-basic question, they don't see our site pop up again and go "nope nope nope." $\endgroup$ – tpg2114 Feb 25 '20 at 14:48
  • $\begingroup$ I like very much your side-by-side of CS vs physics. $\endgroup$ – ZeroTheHero Feb 25 '20 at 15:22
  • $\begingroup$ Regarding your point about maturation, I like the sentiment but it doesn't seem to be true: users that begin as predominantly question askers remain question askers forever, while current question answerers have never asked many questions themselves. $\endgroup$ – knzhou Feb 26 '20 at 0:19
  • $\begingroup$ We can check this by looking at the Q/A ratios of the top rep users. Going from the top, I find ratios of 1:200, 1:1000, 1:200, zero, 1:2500, 1:20, 1:50, 1:100, 1:300, 1:100. (The next user is me, with a ratio of 1:5. AFT and I are basically the only exceptions to this rule, and even our ratios are not remotely balanced.) $\endgroup$ – knzhou Feb 26 '20 at 0:20
  • 1
    $\begingroup$ Similarly you can look at recipients of the Socratic badge, who have ratios on the opposite direction: 10:1, 80:1, 5:1, 5:1, 5:1, 10:1, 15:1. Askers stay askers and answerers stay answerers. $\endgroup$ – knzhou Feb 26 '20 at 0:23
  • $\begingroup$ @knzhou Perhaps the maturation should be broader than I specified -- we need to figure out how to capture people who end up here looking for very basic high-school physics info into long-term users. Especially those who will go on to major in and pursue post-graduate work in physics. Maybe the ones who come to ask questions will always stay askers and their questions will get more advanced. And maybe those who end up here in their intro classes end up answering questions because they are good at it, and go on to answer better questions as they mature. $\endgroup$ – tpg2114 Feb 26 '20 at 0:28
  • 1
    $\begingroup$ But I think we need to recognize that the way people end up on our site means we're only found by the beginners. If we want a strong professional core, I think the most practical is to grow it from the beginners who come through the system, whether they like to ask or to answer. I don't see how experienced professionals will end up here -- aside from seeing something on the HNQ as they go about programming/math/etc.. $\endgroup$ – tpg2114 Feb 26 '20 at 0:29
  • 2
    $\begingroup$ I think we'd also have to look at activity period -- do question askers post large numbers of questions over a decade? Or do they post them in a year or two and then stop? It could be a survivor bias, only the answerers show up in the top rep users because they are the only ones that hang out for awhile. In other words, we may have lost the question askers before we converted them into answerers. $\endgroup$ – tpg2114 Feb 26 '20 at 0:32

One thing we can do is be liberal with downvotes, in order to keep inappropriate questions off the front page, where they deter casual visitors with high standards from sticking around.

I'm always a bit surprised and a bit disappointed when I find a two-hour-old question that already has three close votes, but still has zero downvotes. With very few exceptions, questions that should be closed are questions that give a bad impression to the sort of new users who we'd like to retain. It is important to downvote those questions.

  • 3
    $\begingroup$ Yeah, I agree. I think many people are worried about scaring new users away by doing that. But what I think is that 1) If a new user isn't willing to learn the site guidelines, then that is a red flag. 2) If a new user is going to get upset over down votes, that is a red flag. 3) If a new user is not willing to learn from the down vote, then that is a red flag. $\endgroup$ – BioPhysicist Feb 25 '20 at 12:26
  • 1
    $\begingroup$ @AaronStevens : Agreed all around, though I'll add tangentially that in my view, the biggest of all red flags is the refusal to engage with a request for clarification. I'm not often undecided about whether to downvote, but when I am and when this particular red flag goes up, my decision is made. $\endgroup$ – WillO Feb 25 '20 at 14:24
  • $\begingroup$ Although something I just realized... You cannot down vote while going through the VTC queue. So that could be a possibly explanation as for why certain questions do not have more down votes than they should (when compared to votes to close the question at least). $\endgroup$ – BioPhysicist Feb 28 '20 at 13:39
  • $\begingroup$ @AaronStevens : I hadn't realized this either. Thanks. $\endgroup$ – WillO Feb 28 '20 at 13:52
  • $\begingroup$ Yes, don't forget to vote posts up or down in order to separate gold from crap. As a rule of thumb, Phys.SE in 2019 got 12000 up-votes and 2000 down-votes per month. $\endgroup$ – Qmechanic Feb 29 '20 at 13:30
  • $\begingroup$ Downvoting questions means the OP looses rep. No such “cost” to VTC; moreover it is possible to edit a closed question back into existence. Am I alone in thinking that VTC is milder than downvote? $\endgroup$ – ZeroTheHero Mar 1 '20 at 2:42
  • $\begingroup$ @zerothehero : The issue here has nothing to do with mildness and everything to do with effectiveness. Downvotes will take a question off the front page quickly. Close votes (unless I"m very misinformed) won't. $\endgroup$ – WillO Mar 1 '20 at 3:25
  • $\begingroup$ @WillO you’re right about effectiveness... not sure effectiveness (at least for newer users) is best criteria. We (the community) pretty much all agree on the symptoms, just not necessarily on the cure. I’d have to think about liberally downvoting. Of course there’s the proposal to reduce the # of VTC to 3, which would resolve (in part) the debate. $\endgroup$ – ZeroTheHero Mar 1 '20 at 4:41

I’m gonna put it out there that the direction of the mother company is to broadly increase the user base, not the number of professionals.

You can draw whatever conclusion you want from this, but for my part I will claim that there is bound to be some friction between a goal to increase/not lose professionals and the broader increase in user base.

I will also claim that the Physics site has matured nicely and that there is already a fountain of information for those who care to search for it; this last observation means that many need not register to use the site; it also means that impatient users pollute the site with duplicates (and get angry because questions are downvoted/closed/not answered), something that after many years eventually gets to you. (How many questions per year do we get on addition of relativistic velocities, or twin paradox?)

My conclusion: not much to be done at the individual level.

  • 4
    $\begingroup$ (tongue in cheek) the best one can do at the individual level is to become a professional, and stay a member. $\endgroup$ – AccidentalFourierTransform Feb 21 '20 at 23:41
  • $\begingroup$ @AccidentalFourierTransform oh well... yes I never considered that possibility... $\endgroup$ – ZeroTheHero Feb 22 '20 at 3:59
  • $\begingroup$ @AccidentalFourierTransform So much easier said than done! $\endgroup$ – knzhou Feb 23 '20 at 23:19
  • 1
    $\begingroup$ @knzhou which one? to become a professional? or to stay a member? to be honest, my motivation for either has been in steady decline for as long as I can remember :-P ʲᵏ ʲᵏ ᶦᵐ ⁿᵒᵗ ᵈᵉᵖʳᵉˢˢᵉᵈ ʲᵏ ʲᵏ $\endgroup$ – AccidentalFourierTransform Feb 23 '20 at 23:28
  • 3
    $\begingroup$ @AccidentalFourierTransform Not only are both individually hard, but given how much time I spend on this site instead of doing research, the combination is just about impossible... :P $\endgroup$ – knzhou Feb 23 '20 at 23:41

You can't be "liberal with downvotes" when you only have one per post, and AFAIK serial downvoting is some sort of SE crime in any case.

The thing that annoys me most on SE (not just this site) are a few members (I'm not going to name them) who regularly post answers that are somewhere between poor and wrong. If I had nothing better to do, I could post a better answer, but I'm not going to get any thanks (either real or in virtual internet points) for spending maybe an hour doing so. So I tend to let Gresham's law take is course. If the site wants to "reward" them by accumulating thousands of rep points for rubbish, so be it. Life's to short to be the guy here: https://xkcd.com/386/

The second most annoying thing (again not just this site) are a few members who seem to enjoy posting cryptic comments which are not necessarily "wrong", but IMO don't serve any purpose except to show that they know more than the OP.

To be fair, my general attitude to SE doesn't include the notion that somehow it should be the internet's "best" repository of useful questions and answers for all time. There are plenty of websites which started out with that sort of ambition, didn't achieve it, and withered away. I don't see why SE should be fundamentally different. Of course the top management or SE now has a vested interest in preserving it as a source of their salaries, but that is a very different objective.

So far as I'm concerned, if there is interesting content here I will browse it, and I don't mind contributing a bit of my 70 years of life experience in return - but that's as far as my relationship with SE ever going to go. As for "rep" - right now, I have more than 2,000 unread messages in my SE inbox, and there used to be more before something (presumably automatic) deleted a few thousand of them. And all my SE email gets autodeleted.


This is just one factor among many, but I believe that one of the main things that would need to change to attract and retain professional physicists is the somewhat narrow definition of "physics" that the site has developed over the years.

Let me explain: I don't know if it's just because of the company I keep, but virtually every professional physicist I've ever met has had interests in some traditionally non-physics field, whether it's machine learning, evolutionary biology, economics, neuroscience, the origin of life or something else. It's in their nature to want to push the boundaries of their field, and a major way they do that is to apply the mathematical tools of physics to other disciplines. Questions about this sort of thing will generally be closed immediately on this site, which is one source of discouragement for professionals.

Similarly, even if you are a 'pure' professional physicist, working on understanding some physical system, it's somewhat unlikely that the practical questions you face will be about the physics of the system you're studying. As a professional working on the topic, you most likely know that already, or share an office with someone who does. It's much more likely that you will have a question that is purely about the mathematical tools you need to model your system. Such questions will be closed here because they are considered to be about mathematics rather than physics.

In short, a somewhat necessary condition for attracting professional physicists is to widen the scope a bit so that the site can include the questions that are most likely to be of interest to them. (This could go hand in hand with narrowing it a bit in the opposite direction, so that it includes a bit less of the repetitive homeworky stuff, but that's another story.)

  • 5
    $\begingroup$ I think this follows with my experience -- the questions I have during my research are about math or programming, and so I end up using other parts of the network instead to get those answered. I suppose we could look at the problem differently... do we want professional physicists asking questions here, or do we want them answering questions here? If the former, then we'd really have to broaden the scope and if the latter, we need to find a way to make it engaging and interesting. I'm not advocating one or the other, but it would help to frame the problem others see. $\endgroup$ – tpg2114 Mar 3 '20 at 14:33
  • 1
    $\begingroup$ @tpg2114 that makes sense. I would imagine that if the main goal is to get professionals answering questions here, it would still help a lot if there were more questions from professionals as well, but I take the point about them being two different goals. $\endgroup$ – Nathaniel Mar 3 '20 at 16:31

Pro or not, people who are into physics are into physics for the physics, not for energy spent on and arguments about question curation or refinement.

Look for ways that Physics SE is a little bit more like waking down the hall, grabbing a piece of chalk and posing a question, and a little bit less like, well, arguments about question curation or refinement.

Smaller SE communities are often more cordial and/or collegial exactly because they are a little bit more like waking down the hall, grabbing a piece of chalk and posing a question than the larger ones.

It doesn't have to be angsty like this. Bigger is not always better.

How? I don't know exactly. Has an SE site ever split into two or three? Is that known to be bad?

Off the top of my head I'd propose:

  1. Beginner through undergraduate Physics
  2. Advanced Physics
  3. Experiments and experimental techniques

An advantage would be less moderator burn-out, but the corollary is that six more mods would have to be found. And of course there's the "that's impossible" aspect.

  • 1
    $\begingroup$ Has an SE site ever split into two or three? — yes: English Language and Usage (ELU) split into ELU and English Language Learners (ELL). Maths is also split. $\endgroup$ – gerrit Mar 4 '20 at 12:29
  • $\begingroup$ @gerrit yay! though Maths split a bit differently didn't it? Like it has evolved to a higher plane of consciousness or at least in some ways distinct from the rest of the SE ecosystem? $\endgroup$ – uhoh Mar 4 '20 at 12:54
  • 1
    $\begingroup$ Personally, I don't see how this suggestion would help at all... For one thing, people who are into physics can also be into helping make sure that a physics Q&A site is well moderated/curated. "Arguments" (or ideally discussions) about curation or refinement belong on meta, which is ideally separate enough from the main site. The main site shouldn't have arguments about curation and refinement at all. Ideally you would only see that if you go to meta to seek it out. But your proposal seems blatantly counter-productive. I don't see how splitting up the site would help us grow. $\endgroup$ – JMac Mar 4 '20 at 13:27
  • 1
    $\begingroup$ The current physics site is a merger of a general physics, theoretical physics, and astronomy site when the latter two couldn't sustain themselves. Astronomy has since re-launched, but related questions are on topic here still. Given the relative lack of experimental questions here, I suspect that spin-off would fail or end up just moving to Engineering.SE or something. I'm not sure the environment has changed enough for an advanced site to re-launch -- but maybe it could, since astronomy seems to be doing okay. $\endgroup$ – tpg2114 Mar 4 '20 at 13:29
  • 1
    $\begingroup$ @tpg2114 About the lack of experimental questions here, I wonder if how many might have been closed or otherwise discouraged because "not about physics" or just go unanswered and forgotten (my early experience was one about deposition of aluminum in a vacuum chamber, it's invisible now, may not have been that good, but the close reason was "not about physics"). The experimental physics tag says 10 this week, 33 this month though I don't know if that means anything $\endgroup$ – uhoh Mar 4 '20 at 13:36
  • $\begingroup$ @JMac When you say " help us grow" are you after raw on-topic questions per minute, or something more important? And I hate it when people use "blatant" in that way as if I've just violated some law by sharing my thoughts. I think making the site more attractive to professionals (not my word) is what this question is about in part, so no, not blatantly counter-productive considering the question asked. Oblique to productive I can accept though. $\endgroup$ – uhoh Mar 4 '20 at 13:40
  • $\begingroup$ @uhoh I didn't address either qualitatively or quantitatively, because to me, it still seems blatantly counter productive. I can't think of any way that splitting up a site would encourage it's growth. It's literally the exact opposite. That's why I said it seems blatantly counter productive (and note that "seems" only addresses my opinion; just because you don't agree, it doesn't mean that it doesn't seem blatantly counter productive to me). I think this answer needs to explain a lot more why you expect splitting the site up to lead to growth. Those seem like clear polar opposites. $\endgroup$ – JMac Mar 4 '20 at 13:50
  • 1
    $\begingroup$ @uhoh I'm not sure, I know I've pushed to make sure they can stay open -- but they do need to be appropriately scoped. There's some posts on meta here about how to make them suitably on-topic and if you find ones that have been incorrectly closed, we can see what work needs to be done to reopen them through the queue. I wasn't trying to critique your idea here, just providing some context about how our scope became what it is today. $\endgroup$ – tpg2114 Mar 4 '20 at 13:57
  • $\begingroup$ @tpg2114 okay I'll look for those and do some reading up, thanks! Being experimental and facing some pressing experimental questions these days, perhaps I'll... do an experiment as well. $\endgroup$ – uhoh Mar 4 '20 at 14:05
  • $\begingroup$ @JMac the question starts with several notable contributors leaving, so that seems to be an element of the question. I don't think they left because the question rate was too low. I also don't think just getting more questions asked will suddenly produce "some channel to communicate and interact with other physicists" as the question discusses. Making the site more amenable to physicist-to-physicist interaction is important, there may be a negative correlation between quality and quantity for quantity above some critical value. $\endgroup$ – uhoh Mar 4 '20 at 14:10
  • 1
    $\begingroup$ @gerrit Maths has two sites, but it didn't "split" in any sense; the history is documented here. $\endgroup$ – Emilio Pisanty Mar 5 '20 at 23:41
  • $\begingroup$ @EmilioPisanty I know, it was always split. I didn't mean to imply it became split during its existence. My choice of the word split may have been confusing here (but my dictionary seems to suggest the word may still be correct even if it has always been two sites). $\endgroup$ – gerrit Mar 6 '20 at 8:28

I debated whether or not to post this; I'm not a long-time user and have little standing in this community, and the question may be inactive anyway. But while I saw bit and pieces (especially from knzhou), I didn't really see the following perspective stated very clearly, so I thought I'd give it a shot. My apologies if this is redundant or doesn't add value.

My perspective starts with numbers. As a rough guess, let's say there are 5000 research physicists employed at universities in the US. Say we double that for physics faculty members who are more focused on teaching. Add a few thousand more for physicists employed at the national labs and similar places.

That's tiny. That's only two or three times bigger than my old high school. Compare that to literally millions of professional software developers, and the difference in scale becomes obvious. [And yes, Physics.SE is global, but I'll leave that extrapolation as an exercise for the reader. It doesn't change the conclusion.]

Those numbers tell me that something like Physics.SE will never be equivalent to StackOverflow (the scales are too different), and it is not realistic for it to serve as the "go to" source for physicists or grad students to ask each other questions (there will always be better alternatives made possible by the fact that the community is small). So its main purpose has to be something else.

To me, that "something else" is the popularization of physics. The number of professional physicists is small, but the number of people interested in physics is quite a bit larger. And physicists tend to love physics -- they love to think about physics, talk about physics, debate physics, and explain physics to others. And I'll further assert (without proof) that every professional physicist has a hidden but deep-seated belief that the average person should know more about physics than they currently do. So Physics.SE is a good place for people without a background in physics to ask curiosity-driven questions.

So then the question becomes: how does Physics.SE attract enough high-quality answers that people will want to ask questions? How do we make it more rewarding for experts to participate? I don't have a complete answer, but I have a couple thoughts:

  • First, be respectful of those asking the question. This is not usually a problem (as far as I can tell); I can't remember any instance of someone not being respectful. But still, if a person is interested enough in physics to ask a question, that interest should be encouraged.

  • Second, and more importantly, experts such as professional physicists and graduate students should be publicly supportive of each other. This isn't like an editorial board or peer review session where it's okay to tear apart an argument to make sure it's sound. This is public. You can usually tell who is an expert and who isn't. If you disagree with a portion of an otherwise good answer/comment, make sure to say that the answer/comment is good before going on to the disagreement. If you see an answer that you think is particularly good, add a comment saying why you think so. If you see an answer/comment being attacked by a crackpot, put in a good word for the person who provided the good answer. People are motivated by shows of support.

  • 1
    $\begingroup$ Fair warning for your second point: Comments that just say that another answer/comment is good without making another point will usually be deleted sooner or later. $\endgroup$ – ACuriousMind Mar 7 '20 at 10:48
  • $\begingroup$ @curious: that makes sense. I suppose I really meant that one should specify or elaborate on what makes it a good answer, otherwise you're right, there isn't much value. I'll edit and make that more explicit. Thanks for pointing it out. $\endgroup$ – Richter65 Mar 7 '20 at 14:30
  • 3
    $\begingroup$ Have you got any sources for your numbers? They look off by at least an order of magnitude to me. The APS has around 50,000 members - not all of them are in the US, but not all physicists in the US are members. $\endgroup$ – Emilio Pisanty Mar 7 '20 at 15:18
  • $\begingroup$ @emilio: Good question. I saw the 5000 number somewhere, but I don't remember where. However, AIP keeps track of all faculty members; their estimate is about 10,000 (tinyurl.com/rceqwwf). The national labs was a complete guess. I suspect many APS members are either students, not actively practicing physicists (I'm one example; I left the field many years ago), or in heavily applied fields as closely related to engineering as to the types of areas that tend to stimulate the public's curiosity. But still, even if you take the 50,000 number at face value, it's still a small community. $\endgroup$ – Richter65 Mar 7 '20 at 16:39
  • $\begingroup$ @EmilioPisanty is closer to the mark, at least for the US. A “best guess” for the # of physicists worldwide is about 1/2-million: physicstoday.scitation.org/do/10.1063/PT.5.010310/full $\endgroup$ – ZeroTheHero Mar 7 '20 at 16:56
  • 1
    $\begingroup$ Talking about numbers is ultimately a red herring. (Though that does not mean that you get to pull numbers out of a hat.) I'd be extremely surprised if there are more working mathematicians than working physicists, and yet MathOverflow is a perfectly healthy site. $\endgroup$ – Emilio Pisanty Mar 8 '20 at 22:07

You must log in to answer this question.