Wikipedia:Featured picture candidates/Light Dispersion

Light dispersion

 * Reason:Wow factor
 * Articles this image appears in:Optics
 * Creator:Commons User:Kieff/User:Kieff
 * Nominator: TomStar81 (Talk)


 * Support &mdash; TomStar81 (Talk) 07:45, 22 January 2007 (UTC)
 * Support Booksworm Talk to me! 08:11, 22 January 2007 (UTC) Very Good Diagram
 * weak oppose support  - beautifully implemented animation, and quite encyclopedic except for one aspect - there is no monochromatic magenta in the visible spectrum. Remove the last line of dots and it'll have my enthusiastic support. Debivort 08:30, 22 January 2007 (UTC)
 * Oppose This animation gives the false impression that light is made of "particles" - I'd prefer dashed lines instead of dots. This really doesn't explain dispersion very well. The explanation is: "At the boundary between the media, the wave's phase velocity is altered, it changes direction (not explained why), and its wavelength increases or decreases but its frequency remains constant." And, as Debivort says, scratch the magenta, substitute with indigo if you need 7 colors. --Janke | Talk 09:12, 22 January 2007 (UTC)
 * Weak support I agree the explanation needs to be improved, by relating the concept of light dispersion with the one of light refraction. I'm not sure the dots should be replaced by segments but they should be smaller and more closely spaced. Alvesgaspar 10:57, 22 January 2007 (UTC)
 * Oppose. Nice idea, but I have problems with some details. First of all the wave and particle nature are mixed in the caption and the image. And the wavelength is used in the explanation, which is confusing, as the only length unit, the distance between dots, is the same for all colors. Then the number of incoming and outgoing particles is not equal. This is a serious flaw, as the number of light quanta should be conserved in a simple dispersion. In short, the picture is oversimplifying and not having much information content, and a good portion of it is wrong or confusing too.--Dschwen(A) 10:58, 22 January 2007 (UTC)
 * Now I get it, sorry, the spacing of the dots alludes to the change in speed of light in the medium. Still confusing though. --Dschwen(A) 15:24, 22 January 2007 (UTC)
 * and the wavelength, potentially, though this falls apart when the streams exit the prism. Debivort 17:53, 22 January 2007 (UTC)
 * Oppose. This makes it look like "one photon in, 7 photons out". It might be better to just use a line/beam going into the prism. &mdash; BRIAN 0918 &bull; 2007-01-22 18:18Z
 * That wouldn't quite work as an animated model, it'd be just as good as the several (static) dispersion diagrams we have out there. I wanted to show how different frequencies have different speeds in a medium, and this conceptual "light particle" model suits this purpose. Slightly different models would probably do better, and I'll get to that eventually. &mdash; Kieff 23:32, 22 January 2007 (UTC)
 * Oh wait, now I get it. You think there shouldn't be "white" photons, but a continuous line. Hmmm... Good idea. Makes more sense that way. &mdash; Kieff 23:44, 22 January 2007 (UTC)
 * Oppose. Light as a particle/wave issue is potentially confusing. Agree with above. --Dante Alighieri | Talk 21:28, 22 January 2007 (UTC)
 * Comment. My point while making this image was to show how short wavelengths (represented by color alone in this model) slowdown in a medium, and how that's related to refraction and dispersion. I agree that dots can be misleading, but that's what I was able to do at the time (note: this is why I originally posted this at Talk:Dispersion (optics) before anywhere else, I wanted some opinions on how to improve it.) I could make little moving lines, or even little sine waves instead (this could work pretty well, I think), but I'll need to code a new thing, possibly from scratch. I can't do it right now, but I will do it once I have the time. Any suggestions are welcome. &mdash; Kieff 23:26, 22 January 2007 (UTC)
 * I don't have any problem with dots as the unit of light (wave/particle duality seems to imply you could pick either). White dots are deceptive though, as there is no such thing as a white photon. Maybe you could have 7 partially overlapping ROYGBIV colored dots going into the prism in each cycle of the animation, and where they overlapped, the combined color would obey additive light rules (i.e. red dot + blue dot = magenta dot). Where 3 dots containing R, G and B overlapped you would have white, and this would accurately show how the white light was composed. Then the mass of seven could enter the prism and the rest of the animation could stay the same. This would solve the "1 quantum in, 7 quanta out" and "no white quanta" problems. Then all that would remain is eliminating the magenta stream, and I would say it would be just about spot on at that point. Debivort 00:45, 23 January 2007 (UTC)


 * Support I think the animation is professionally done and certainly fulfills the criteria of “being eye-catching to the point where users will want to read its accompanying article.” I think criticizing it for not accurately keeping count of the dots going in and through the prism is off base. The dots going in are white and are therefore brighter because they contain all the colors. The prism properly shows the time-chopped white light (representing multitudes of photons of varying wavelengths) being split into colored dots that are darker (containing fewer photons per dot). All in all, for each dot of white light going in, there is one darker dot representing each binned value of color moving into the prism's queue (at varying speeds), and finally, one dot of each color leaving the prism. It accurately depicts the varying speed effect for each of the colors (it has to in order to keep the dots-in vs. dots-out in check). The entire animation is only eight frames and even its full-size version is only 121 kB. Accordingly, the smaller thumbnail in the optics article loads lightning quick. It captures all the important concepts of refraction and refractive index in one simple animation that no print version of an encyclopedia can match. Wikipedia needs more stuff like this.   Greg L 04:18, 23 January 2007 (UTC)
 * One can certainly rationalize the image into accuracy, but it would be very easy for a person to look at it, and come away thinking that white photons exist, or that 7 photons exit a prism for each that goes in - the discrete nature of the dots begs for this interpretation. I agree that technically it is absolutely top of the line, and with some simple technical modifications, it could have encyclopedicity that is top of the line as well. Debivort 06:52, 23 January 2007 (UTC)
 * I agree. The white photons are misleading. Doing the additive effect would be tricky, but it can be done. But, I'm thinking of something else... What if the white dots are larger than the colored dots? Would that work? &mdash; Kieff 09:27, 23 January 2007 (UTC)
 * Done this. See if it's any better. &mdash; Kieff 11:48, 23 January 2007 (UTC)


 * Comment. Maybe it would be better if white dots are replaced by all color dots going along the same line with some delay. Another problem is that it is not clear from the picture why dots of different colors bend at different angles. Olegivvit 13:13, 23 January 2007 (UTC)
 * Comment - That is precisely the effect of having different refraction angles for differente colours, caused by the fact that different wavelengths propagate with different phase velocities. The only way to show the phenomenon in the animation is to have a "wavefront" approaching the prism instead of just a stream of light "points". But that would also complicate the animation inside the prism... Alvesgaspar 15:57, 23 January 2007 (UTC)
 * Comment It looks like an animation of Newton's experiment to me: A time-chopped beam of white light goes in, and a spectrum comes out. The animation no more conveys the notion that white light is comprised of "white photons" than does simply watching a beam of light from a flashlight. Kieff, if some readers really might come away from the optics article (and this animation) with the notion that there are "white photons," perhaps a better caption could overcome some of these objections. I think your animation describes valuable concepts like how the higher refractive index of glass combined with an angle, allows the different colors to diverge. And by watching the different colors bars slew forward into an angle as they travel inside the prism, one can see how a tight pulse of light gets "chirped" across time. Now that's a neat concept I hadn't noticed before. A caption that says only “Conceptual animation of dispersion of light in a prism” doesn't help the animation as much as it could.  Greg L 20:22, 23 January 2007 (UTC)
 * Copied from the Wikimedia Commons FPC page

Here, time-chopped packets of white light, each consisting of the entire color spectrum, enter a prism. Shorter wavelengths (blue and violet) travel slower in the glass medium than do longer wavelengths like red. Consequently, shorter wavelengths are refracted (bent) more than longer ones. The prism causes the light to disperse and fan out into a rainbow-like spectrum. For each packet of white light entering the prism, a color-dispersed packet of light exits the prism. Because light travels slower in glass than in air, the packets necessarily bunch up inside the prism and only resume their normal speed (and spacing) after exiting. A very brief flash or modulation to the light entering the prism will be “chirped” (smeared out over time based on frequency) after passing through a prism; the red red modulations arrive to an observer first, and violet last. This chirping effect can be seen here as the columns of colors begin diverging inside the prism and increasingly shear forward.
 * Comment Also, see the alternate image posted here. Nothing against it or anything, but I must admit, I think it conveys little information about dispersion except the fact that it happens! There are little clues in the angle of light while it is inside the prism (though it gets wrong on the exit), but you just can't expect someone to look at it and understand what's going on, especially if the person isn't already familiarized with refraction and dispersion. For all educational purposes, it's just as good as a static image. This is what I was trying to avoid. I wanted something to show, conceptually, what goes inside the prism, and for that I needed individual parts moving, so I picked dots (mainly because it was easier to understand and follow - as well as code! - but also because I thought it would work reasonably well.) It works, but it is not entirely accurate. But again, isn't that how it usually goes? We always use simplified and inaccurate scientific models in order to teach people new concepts (Newtonian physics and gravity instead of General Relativity, frictionless systems, point masses and charges, electrons as particles in orbit, etc.)... I believe that as long as we make the inaccuracy clear, we shouldn't have a problem with them. I think this applies to this image. We teach a concept, and we filter out the inaccuracies later on with a more in-depth explanation. Anyway, I'll try making a version with little waves, but I don't think complaining about inaccuracy is a valid point against the model. It's nothing a bit more of insight can't fix, and by then the model will have done its purpose, and it would have done it well enough. &mdash; Kieff | Talk 23:07, 23 January 2007 (UTC)
 * Agree about the simplification for teaching purposes, but since you already post this image for comparison, how does your animation explian the fact the lightpath is bent? You show the fact that c in a medium is a function of wavelength, but by no means does it explain why the angles must be different. Alves is right when he says the wavefronts must be drawn for a propper explanation. --Dschwen(A) 23:16, 23 January 2007 (UTC)
 * Agreed. But it doesn't explain why because it really can't. It wasn't even the original intention, which was really just showing the relations between phase velocity in a medium and vacuum, wavelength and index of refraction. The particle model works for this, but not for the rest. I'm gonna try making a wavefront one later on whenever I have the time. I have my doubts it'll be as visual appealing, though, since lines would probably become confusing and difficult to follow inside the prism. We'll see then. &mdash; Kieff | Talk 23:46, 23 January 2007 (UTC)
 * Support Original. As Borat once said, "Very Nice! I like!". But really, either could do, but I think the original is better. Very good :-) Ilikefood 16:07, 24 January 2007 (UTC)
 * Oppose Liable to mislead about the composition of light. Pstuart84 Talk 17:26, 24 January 2007 (UTC)
 * I'm going to wait and see how the peer-review above plays out before making a decision. 82.109.231.174 17:38, 24 January 2007 (UTC) Pstuart84 Talk 17:40, 24 January 2007 (UTC)
 * Caption Suggestion I think the animation might be better served with an expanded definition. How about this proposal:
 * Greg L 03:19, 25 January 2007 (UTC)

Raven4x4x 01:50, 31 January 2007 (UTC)