4
$\begingroup$

A large number of our most active members are software developers (emphasis on large, because there are way more than I'd intuitively expect). At least three of our moderators mention in their "about me" sections that they're employed by software companies (hence presumably software engineers or similar), and 2 of them are extremely active physics content creators.

To a certain extent, it makes sense, as there is (apparently) a significant overlap in skills which can encourage people from physics backgrounds to move towards software development. However, it doesn't work the other way: after software engineering and computer science degrees, I'm quite sure you won't know a huge amount of physics.

I don't have a problem at all; even I'm a somewhat active programmer. But it is definitely interesting. Why is this so? Does it imply anything about Physics SE?

There's a lot of talk about how Theoretical Physics SE's questions are supposed to be asked here, now that TPSE's closed, but if so many of our best answerers are software creators, will we be able to cater to them well? Will it be productive to try a survey similar to the StackOverflow developer survey to gauge who we're creating content for, who's using it, and who's creating it? I personally feel that it'll be very beneficial, because it's necessary to know your audience well t write for them, and it's also good to be aware of where the author of an answer is coming from.


This sounds like many questions rolled together, but at the heart of it, it's just What's the significance of the large proportion of software developers on Physics SE?

The answer may be "nothing", but it could potentially tell us something very interesting about what's going on here.


Edit: Some answers and comments have brought up the point that there're no reliable statistics/samples to prove this, which makes sense. I'll emphasize that it was as simple as me being surprised by the number of software developers, and leave the quantification at that. It's possible that I misjudged things. This observation is just my opinion and was based upon a section of active participants only. The theory cannot be extended across all users.

$\endgroup$
  • $\begingroup$ Not sure I agree with the premise of the question given the lack of concrete data. But consider -- StackOverflow was the first and is the biggest of all StackExchanges and as the network grew, many of those users likely propagated around. $\endgroup$ – tpg2114 Jun 4 '18 at 16:04
  • 1
    $\begingroup$ Maybe software developers have more time and motivation to spend on explaining physics in their free time than most who do physics for a living ;) $\endgroup$ – ACuriousMind Jun 4 '18 at 16:37
  • 4
    $\begingroup$ As for your puzzling comment on TP.SE - that site is long gone, and its content has already been migrated in full to physics.SE. You sound as if you think this is somehow an ongoing - or even future - event, but this is not the case. Maybe look at the timestamps of the posts that gave you the impression this was an active discussion... $\endgroup$ – ACuriousMind Jun 4 '18 at 16:38
  • $\begingroup$ @ACuriousMind I have significantly less time to answer questions now as a software dev than when I was a grad student. $\endgroup$ – Kyle Kanos Jun 4 '18 at 17:19
  • 1
    $\begingroup$ I'm extremely reluctant to just take as fact the anecdotal observation that a few users are software devs and then extrapolate it to the full userbase. If that doesn't scream Selection Effect at full volume, then I don't know what does. $\endgroup$ – Emilio Pisanty Jun 4 '18 at 17:56
  • 1
    $\begingroup$ It's the other way around: Most answerers at Phys.SE have a physics degree. Now if you want to make money with a physics degree, you want to work in the private sector. Software developer is one natural choice. $\endgroup$ – Qmechanic Jun 4 '18 at 19:01
  • 1
    $\begingroup$ It really just shows how bad the job market in physics is. $\endgroup$ – knzhou Jun 4 '18 at 22:41
  • 1
    $\begingroup$ @Qmechanic ... for the theoretically-inclined perhaps. For those closer to experiment there are plenty of industry jobs in nanotechnology, fabrication, optical and imaging devices etc. $\endgroup$ – ZeroTheHero Jun 4 '18 at 22:43
  • $\begingroup$ Of course, there are many jobs. $\endgroup$ – Qmechanic Jun 4 '18 at 22:55
  • $\begingroup$ @Chair Not really - you're still making a statistical claim; the extra qualifiers just make it fuzzier and harder to test. $\endgroup$ – Emilio Pisanty Jun 5 '18 at 8:31
  • $\begingroup$ @Chair It's not misjudgements - it's just how anecdotal evidence and small samples work. As for "larger than expected", that kinda depends on where you place your expectations, but if you inform them with the fact that this site grew out of a largely SO userbase, then it's rather in line with my natural expectations. $\endgroup$ – Emilio Pisanty Jun 5 '18 at 11:49
  • $\begingroup$ @Chair Well, keep in mind that what you perceive as a homogeneous population isn't actually: within the mod team, David Z was a member of SO, with hundreds of posts, at the time PSE started, whereas ACM joined more recently and was doing theoretical physics, likely with rather little computer work, before leaving academia. And that points to the other prior that needs to be set correctly: what fraction of physics graduates move on to programming jobs? $\endgroup$ – Emilio Pisanty Jun 5 '18 at 12:58
8
$\begingroup$

This is kind of a network-wide phenomenon, and of course it's rooted in the history of Stack Exchange. The first site was Stack Overflow, which, for ten years, has remained the most active site on the network. Next came Server Fault and Super User - both geared toward people who work in and are enthusiastic about fields related to computers and IT, including software development. From there, Stack Exchange branched out into some other sites about topics not as closely tied to computers, before finally adopting a new site creation process, leading to a wave of new sites about . . . everything.

Stack Exchange will always cater to programmers, developers, sys admins, and that guy in IT who knows why your computer keeps crashing. That's a huge constituency, the largest part of the userbase (the sites for those folks are the busiest). Moreover, it turns out that Stack Overflow is one of the biggest referrers of traffic to all the other sites in the network, which you can see if you have enough reputation to check out the site analytics. Therefore, a lot of users come from a programming background, and, yes, many work as software developers (I'm talking about the userbase of Physics as a whole here, not specific users).

This is starting to change, largely within the last . . . five or so years, maybe? There are over 150 Stack Exchange sites (174 at last count, but that's always increasing), and we're seeing a lot of new users come directly for non-programming-related sites. That can be good, insofar as it can attract people who really know their stuff. The network's non-computer-related userbase is changing from programmers enthusiastic about certain areas people who study and work in those areas. And I think that's a good thing. I'm one of those people.

Then again, there's a viewpoint from which much of this doesn't matter. For this site, if you like asking and answering questions about Physics, you know your stuff (or know what you don't know!), and are committed to helping the community, I don't think anyone cares if you're a physicist or an opera singer. We're here for physics. And that's what's important.

$\endgroup$
  • $\begingroup$ "about certain areas people who study" - missing a "to"? $\endgroup$ – Emilio Pisanty Jun 5 '18 at 8:32
  • 3
    $\begingroup$ Your final paragraph settles it for me: next season I'm finally auditioning for my city's opera chorus. $\endgroup$ – rob Jun 5 '18 at 11:45
  • $\begingroup$ At the end of it, do you believe in the practicality of a user survey? $\endgroup$ – user191954 Jun 5 '18 at 12:22
  • 1
    $\begingroup$ @rob Hey, it's all about resonances and fundamental frequencies and harmonics - nothing we can't calculate. And neutrinos. Oh, wait, that's a different OPERA. . . $\endgroup$ – HDE 226868 Jun 5 '18 at 12:48
  • $\begingroup$ @Chair That depends on what you'd be trying to accomplish with it. A user survey would be a step towards either confirming or refuting my answer, as well as perhaps towards learning more about physics education among the users. But I'm not sure what benefit it would have to the site. (Practical, though? Probably, yes.) $\endgroup$ – HDE 226868 Jun 5 '18 at 12:49
3
$\begingroup$

In addition to what HDE's answer mentioned, I think part of the cause is simply that, among the people who have studied physics in depth, many of them now work as programmers. (Probably not a majority, but a significant fraction.) There aren't enough jobs in physics itself for everybody with the training, so a lot of them fall back on software development.

Of course, if you're just looking at the moderators, that's almost certainly too small of a sample size to draw a meaningful conclusion. But if you did a broader survey of active members, and if that survey gave a similar result, this would probably be part of the reason why.

$\endgroup$

You must log in to answer this question.