It is about time to start thinking about the final design of the site; please turn on your creativity and start posting:

  • Colours, shapes, motifs, fonts... anything visual that you think is connected with physics (eg. "Magenta Rutherford atoms!")
  • Things you definitely don't want (eg. "Everything except yet another Rutherford atom model!")
  • Logo sketches/ideas (eg. "The electons' orbits should form `Physics' in Comic Sans.")
  • Catchy ideas for upvote/downvote buttons, tabs, badges, tags (eg. "Downvoting should make it change into Thomson model.")
  • Ideas for 404/Error/CAPTCHA pages (eg. "404? Obviously a photo of dark energy!")

And, as usual, upvoting ideas you like and downvoting things you don't like.

  • 1
    $\begingroup$ There's nothing wrong with this thread, though it's really the official StackExchange designer, who usually (going by the past) generates the designs himself and gets community feedback for tweaking and whatnot. $\endgroup$ – Noldorin Jan 23 '11 at 0:09

Not another atom. Absolutely. Maybe some logo with a $\Psi$.

Dark Energy for "not found", of course, will do. But I think the most suitable here would be the higgs boson (at least up to now).

Edit: I've checked how other launched sites look like and I really don't like the look of some of them. Too many bright colors or too many tiny details are really distracting. I actually like the present "scetchy" look -- maybe one can adopt it from "notebook scetchy" to "white/blackboard scetchy".

Edit: This is what I made a minute ago, using my blackboard:

Black board

I tried to use ladder operators but they didn't look nice...
While I think that bra-s or ket-s (better ket-s) should be used for tags, don't they?

  • 12
    $\begingroup$ +1 for the idea of a Higgs boson on the 404 page :-) If it weren't for copyright issues, I'd suggest using this guy. $\endgroup$ – David Z Jan 14 '11 at 0:04
  • 3
    $\begingroup$ also, just thinking: for our actual logo, I'd rather see something visual and colorful as opposed to a mathematical expression. $\endgroup$ – David Z Jan 17 '11 at 8:51
  • $\begingroup$ we are looking at this as a possible chalkboard color -- what do you think? staffwww.fullcoll.edu/rsimpson/Boy%20at%20Blackboard.jpg $\endgroup$ – Jeff Atwood Feb 17 '11 at 5:04
  • 1
    $\begingroup$ Good chalkboards are made out of slate and should be dark grey/black. $\endgroup$ – pho Feb 17 '11 at 14:36
  • $\begingroup$ That is the problem -- black would be really bad (just take a screenshot and invert the colors). Maybe, whiteboard will do? twistedphysics.typepad.com/cocktail_party_physics/images/2008/… $\endgroup$ – Kostya Feb 17 '11 at 14:41
  • 2
    $\begingroup$ Yeah, but real physicists don't use whiteboards. ;) $\endgroup$ – pho Feb 17 '11 at 16:10
  • 1
    $\begingroup$ True. :) Don't even know what to do... Hmm -- what about "hardcore" blackboard? I mean -- with chalk-filled "background", so It's color is ~grey. $\endgroup$ – Kostya Feb 17 '11 at 17:19
  • $\begingroup$ I love the blackboard look. However I think the arrows in your mockup may be slightly confusing. I'd love to make them physics related. However if they're too gimmicky they may lose the functional part. $\endgroup$ – Jin Feb 18 '11 at 16:24
  • 2
    $\begingroup$ What if you flipped the bra/ket sideways, so they point up and down like arrows? (I can work up an example later if it's not clear what I mean) $\endgroup$ – David Z Feb 18 '11 at 19:27

How about a sketchy inspired chalkboard design?


If we're to have any graphical motifs there should be a Feynman diagram included. And something that references classical physics (Maxwell's equations? The classic Keplerian areal sweeps on an elliptical orbit? Ray optics?), and something that talks to experiment (a graph with real data including error bars/bands or an event display from a Fermi top quark event or ...).

Lets have a diverse set of representations. Physics is a big field, let's show it.

The tex.SE header provides one template for how this might be done, but I confess to not liking the overall layout very much. I do like the ghostly figures in the header, however.

  • 1
    $\begingroup$ For graphs, the CMB power spectrum is pretty iconic, although I can understand if people would rather not use something that's been popularized by XKCD. I happen to like the look of the Mexican-hat potential, but it might not be so appealing without a context. I guess we could overlay it with a typical Lagrangian density that exhibits spontaneous symmetry breaking. Alternatively, a particle collision event would be quite nice, I think. $\endgroup$ – David Z Jan 16 '11 at 3:29
  • 1
    $\begingroup$ On the condensed matter side, not that I know all that much about condensed matter, but we could perhaps use something showing the molecular structure of graphene, since that seems to be all the rage these days ;-) $\endgroup$ – David Z Jan 16 '11 at 3:30
  • 1
    $\begingroup$ I'd like to have an Ising model at critical temperature; those fractals are pretty nice. But I suppose I must be among just a few to want something like that :) For the more usual stuff: phase diagrams are pretty neat sometimes; I was always mesmerized by the gas-liquid curve abrupt end and wondered what might be happening there :) $\endgroup$ – Marek Jan 16 '11 at 16:02
  • $\begingroup$ "Throwing a bone" to the experimenters by slapping a few error bars on a graph could be interpreted badly. $\endgroup$ – ptomato Feb 10 '11 at 15:56
  • $\begingroup$ @ptomato: Thus the emphasis on real data. The point of choosing one that includes error indicators (instead of say the WMAP blackbody spectrum where the error bars are smaller than the plotting symbols) is pedagogical. As an experimenter, getting vanishing error bars like that is always a goal, but for making a point about physics having non-trivial uncertainty in the results is preferred. $\endgroup$ – dmckee --- ex-moderator kitten Feb 10 '11 at 18:09

Since you mention upvote and downvote buttons, it would be neat if we could somehow tie them into the raising and lowering operators. But I can't think of a way to do it that wouldn't be excessively nerdy.

I actually kind of like the design for math.SE, with a few exceptions. The graph paper in the background I could do without, of course, and I don't think parentheses are enough to visually distinguish the tags, and I'd like the upvote and downvote buttons to be a little more colorful, and it wouldn't be that bad to have some faint colors to highlight some of the regions in the right sidebar. But I do like the general feel of simplicity, e.g. using a pure white background and simple, thin lines to separate sections of the pages, along with a bold logo and site title at the top.

Also, I think the logo for webmasters.SE is pretty nice. Of course we're all tired of classical atom models so I'm not saying I would have wanted that logo for this site, but I am a fan of the use of the three primary colors (which corresponding to quark colors, coincidentally enough), and the subtly highlighted globe which is unobtrusive against the blue background of that site.

Actually webmasters.SE has a pretty plain look as well. It's probably my favorite of all the SE site designs. My only real criticism is that it's a little too blue for my taste; I prefer to have just a tad more variety in the color scheme.

  • $\begingroup$ Indeed, webmasters.SE is very nice. But the top bar is little bit too fancy to my liking. Love the logo (probably because it's more about physics than webmastering) but not the blue-white gradient and font shadows. $\endgroup$ – Marek Jan 15 '11 at 10:08
  • 7
    $\begingroup$ +1 for ladder operators $\endgroup$ – dmckee --- ex-moderator kitten Jan 15 '11 at 20:13
  • 2
    $\begingroup$ Yes please! ladder operators for +1 and -1. Totally. $\endgroup$ – Stefano Borini Feb 3 '11 at 0:28
  • 1
    $\begingroup$ Why were the ladder operators discarded? $\endgroup$ – Abhimanyu Pallavi Sudhir Jul 3 '13 at 14:10

I actually love this look. The thing is that I've been around polished websites for so long that they glaze over my head. Looking at this page is just a sight for sore eyes. Let me tell you why. As someone who dabbles in design I think that sleek, sophisticated design is overrated. Simplicity isn't.

Right now what you have is just an unpretentious page on which the content flows out. It's like a blank sheet of paper waiting to be explored. Waiting to be written upon. There's nothing there but pure nothingness. ;)

So, why dress it up?

If you've seen the design for Hacker|News it looks so primitive at the start, but the fact is that it performs a function it acts as a barrier against the 4Chan/Digg crowd as there is nothing there on the site, but the content itself. No videos. No dressup. No "social features". Just the content. What can be more beautiful than that?

To me that's the highest possible form of design. Paul Graham has tweaked it so that it actually encourages the participation of his target audience and acts as a repellent against the gremlins. What could be more beautiful than that?

So please keep our site like this. In fact, if you can remove the nifty design tricks that you've built into it. Just make a blank sheet of paper and we, the community, will do the rest.

  • 6
    $\begingroup$ @Anna: good point that simplicity comes before design. However, that's not the end of the story. First, good design doesn't need to clash with simplicity but can use it in clever ways (think Google). Second, all of the SE beta sites look like this. We want people to know they are at physics site, and not just at any other site. $\endgroup$ – Marek Jan 13 '11 at 22:47
  • $\begingroup$ To given an example, I love MO design. It's as simple as this beta, but with much better design in my opinion. Actually, it took me a long time to get used to beta SE sites (and I am still not totally converted). $\endgroup$ – Marek Jan 13 '11 at 22:49
  • $\begingroup$ MO might be even simpler than this site is. Anyway @Marek, I suspect the beta design is supposed to seem like there's something a little off about it... maybe you're just experiencing the intended reaction. $\endgroup$ – David Z Jan 14 '11 at 0:03
  • 1
    $\begingroup$ @David: yeah. I think it's the coordinates on the background. Clearly one needs to get rid of those and accept general covariance if the site is to look any good :-) $\endgroup$ – Marek Jan 14 '11 at 0:06
  • $\begingroup$ @Marek: I understand, but what if you did it in a subtle way? Just iconology here and there on the top of this design to make it an unconscious process. Can I think for a while and then get back to all of you? I need to reflect about this for a while. $\endgroup$ – Anna Jan 14 '11 at 18:25
  • $\begingroup$ @Anna: sure, that's precisely the way I'd like it. Decent colors (not sure whether there are colors associated with physics; but I'd definitely welcome a change), some non-intrusive logo at the top (e.g. some particle physics or famous equations). But in general I'd also like to keep the design as simple as possible. Looking forward to what you'll come up with. $\endgroup$ – Marek Jan 14 '11 at 19:11
  • 1
    $\begingroup$ Hacker News and MO look horrible to my eyes. $\endgroup$ – wrongusername Jan 16 '11 at 22:26

At the moment, the most popular question tags are quantum mechanics, quantum field theory and general relativity, so Schrödinger equation or Feynman diagrams are perfect elements for the site design. On the contrary, decorating the site with elementary pictures of interference, white light dispersion or pendulum motion would give Physics.SE an 'easy' look and would probably attract lower-level questions.


I've been thinking a Feynman diagram of Higgs decay or neutrinoless double beta decay would be great for the 404 page, if we get to customize it:

  • Higgs to 4l decay via a top quark loop

    Higgs to 4l decay

  • Neutrinoless double beta decay

    neutrinoless double beta decay

(the former is my own creation, the latter is public domain from Wikipedia, so either of these could be used directly)

Dark matter or a magnetic monopole would also be decent possibilities but I can't think of a visual representation of either that carries as much "punch".


Something I think we should avoid is overly representing mathematics or equations in the design, so no Schroedinger equations or nablas. We should go for a physical phenomenon. My personal preference would be something actually visually observable, for example with the naked eye or under a microscope, not just a plot of something. My personal preference leans toward optics, so for example a diffraction pattern. Perhaps a standing wave pattern from acoustics made visible in some fashion?

  • 3
    $\begingroup$ Although I don't agree with disallowing math entirely, showing a physical phenomenon is not a bad idea. $\endgroup$ – David Z Feb 10 '11 at 18:38

You must log in to answer this question.