Hacker News new | past | comments | ask | show | jobs | submit login
Light - If you view the Earth from far enough away can you observe its past? (physics.stackexchange.com)
117 points by jjchiw on July 8, 2011 | hide | past | favorite | 65 comments



On a related note:

A 3GHz computer (ignoring multi-core, hyperthreading, other issues) executes 3 instructions every nanosecond. Light travels 1 foot per nanosecond. Your head is about 2 feet from your monitor. In the time it takes the light from this post on your monitor to reach your eye, your computer performs 6 instructions. Scale as appropriate.

When I first realized that, I wandered around in a slight daze thinking "wow, light is slow..."


Did not know this, love it and am also amazed that all of us are alive right now in an era where humans are dabbling with technology that is starting to ram right up against (perceived?) physical limitations.


Light is not slow, you're just comparing the wrong things. A processor computes instructions in parallel. It's like comparing the speed of 3 separate light beams to a single light beam. You can't say the 3 light beams were faster because they've crossed 3 times the distance a single beam could in the same time. And how many instructions do you think it takes the computer to impose a change of light?


I remember thinking of this one time as a teenager. Could we see a top down view of the rise of man? The huge caveat is that you have some capability of travelling faster than the light leaving earth, so that you can then look back and absorb the rays. It makes perfectly good sense that if you could travel faster than the speed of light you could use a massive telescope to peer back at the earth and see it's past geological events -- and I mean millions of years even, not just something trivial -- or given a strong enough telescope perhaps more detail could be seen. All of the information that left the earth as light is still out there, all over the universe.


>Could we see a top down view of the rise of man? The huge caveat is that you have some capability of travelling faster than the light leaving earth, so that you can then look back and absorb the rays.

Easier: look at the light as it was gravitationally refracted around a black hole; some paths will probably reverse the relevant light and send it back towards the Earth.

Unfortunately the size of the lens required to actually view this is impractically huge, possibly larger than the Solar System itself -- but it's fun to think about, I guess.


Also, because light is quantum and not continuous, eventually you get to the point where there isn't any light. You can't just keep cutting out X% of the light and still have some there.


>You can't just keep cutting out X% of the light and still have some there.

Yes you can.


All of the information that left the earth as light is still out there, all over the universe.

A very poetic point of view. :)

As others mentioned in the thread, you don't necessarily travel yourself. If you had a mirror far enough away, you could look into the past as well.


Right, but the mirror would have to have been placed at least as far back as the half way point to where you want to see. If you want to see back two million years, you would have had to place the mirror one million years ago.


Since the mirror would be light years away, it would take light years for the reflection to come back to Earth.


There must be a calculable limit to the size of the event that you can see at a given distance. There are finite number of photons which are emitted from an event, these spread out with the inverse square law, so as your distance increases the probability of collecting sufficient photos to reconstruct an image of the event decreases and I would think at a distance of one lightyear it would be guess hard to see something much dimmer than an atom bomb.


But light travels as a wave, and my humble (and possibly mistaken) understanding (or conjecture) is that photons are nothing but particle-like manifestations of whatever thing light is "really" made of.

Waves travel in all directions, they get weaker over long distances, but it's not as though some photon particles reach us while others don't. (or is it?)

I mean, if light is just photon particles, then we'd have to be really lucky to see that many starts that are billions of light years away.

Of course, I could be totally wrong!


Light travels as both a wave and as particles. Whether you observe light to behave as mostly particle-like or mostly wave-like depends on the circumstances. When you are dealing with very small intensities, you usually observe mostly particle-like behavior (but wave-like behavior still shows up in things like self-interference as demonstrated by the two-slit experiment).

If you think that it's unlikely for a photon from a distant star to reach us, then you are simply vastly underestimating the number of photons involved. Consider by analogy the process of smelling something: actual molecules need to fly off the object, travel through the air, and impact your nasal passages in order for you to smell it. That means that anything you smell is constantly dispersing its mass into the atmosphere, but an object can remain pungent for a long time without significant weight loss. Photons are vastly more numerous than olfactory molecules, and carry incredibly small amounts of energy.


One pottential difficulty in achieving this in practice would be that the amount of information (photons) available to measure would rapidly attenuate with distance, so you'd need to mitigate this by building larger and/or more sensitive detectors.


Fascinating question and fantastic answer.

That being said, as Edgar added to Vintage's answer, if you put a mirror 13.5 light years away, and watched your reflection from earth, that would be the same as being 27 light years away.

So if I doubled the number of mirrors (2 in orbit, 2 on earth) and halved the distance to 6.75 light years I could accomplish the same thing.

If you take that example to it's conclusion, could I construct an (expensive/complicated/etc.) device here on earth that had so many mirrors it could let me look into the past at all?

The physics-answer seems "yes" so my question is "Why haven't we tried that?" and one obvious limitation, I suppose, would be just how many mirrors you would need.

Light travels 5,878,499,562,554 miles (5.78 trillion) a year[1].

Given that, if I just wanted to see an hour into the past, I think that means I would need to observe earth from: 5,878,499,562,554 / 365 days / 24 hours = 671,061,594 miles away.

Or I could stick a mirror in space 335,530,797 miles away (~ 540,000,000 km) from earth and stare at it.

Mars, at the widest distance from Earth, is 401,000,000 km away[2], which is close enough for my purposes (I'm not picky)... so I guess if I stuck a mirror on Mars and looked at the reflection of earth I could see something like 45 minutes in the past.

The Moon is almost exactly 1000x closer to the earth than Mars[3], so I wonder if I used it for my mirror array instead if I could just put a station with 500 mirrors on it to accomplish the same thing.

Or build something on earth with millions of mirrors in it to accomplish the same thing.

I would normally think something like this impossible, but I just watched a docu on the LHC and now I wonder if even at a micro-second scale, if we have tried building something like this and observed two points in space using a computer and seeing if the visual data coming in is micro-seconds apart from each other?

For example (assume I have a camera and visual-diff software sufficient for this and that my "mirrors" have sufficient magnification capabilities to make this seem like an easy setup), if I pointed one camera at a monitor drawing a unique pattern 2' away from me, then point another camera at a mirror that has bounced the image 10 miles before being displayed... I imagine, like sound, there would be a lag in that image if we bounced it enough times.

(DOH)

It suddenly dawns on me that using this method to look into the past is effectively the same thing as recording something with a video camera and playing it back later... you are literally capturing the light for review at a later date.

So as cool as this idea is, I think I just answered my own question as to why we haven't tried to build a million-mirror-array before... cause I can buy a video camera for $300 instead :)

[1] http://www.universetoday.com/45047/how-far-does-light-travel...

[2] http://www.universetoday.com/14824/distance-from-earth-to-ma...

[3] http://www.enotes.com/science-fact-finder/space/how-far-moon...


Since you'd need to put all this stuff in place before the event you want to see actually occurs, wouldn't it be easier to just break out a video camera and record it?


I think this would have other uses. Presumably, this thing would not use energy; it would just pass photons from A to B. Make one a meter square with a twelve hour delay and point it to he sky, and you have instant lighting at night. It would he like a battery storing solar energy coupled with a lamp.

Unfortunately, even if we could fold a 12 hour light path into a tiny space, losses would probably be too high. Some of the light that got in would come out after 12 hours,but most of it would be converted into heat long before.


hugh3, you are exactly right. I updated the original post as I was thinking through how this amazing physics experiment could be helpful, and it dawned on me, it wasn't.

I could just use my Canon to accomplish the same thing.

I think my brain was more caught up in the fun of walking through the thought than accomplishing something with it.


Even so, I think it would be fascinating to interact with a device that consisted of mirrors, even if was just to experience a delay of a millisecond (or whatever is the minimum amount of time to perceive a delay).


You can slow down light http://news.bbc.co.uk/2/hi/science/nature/655518.stm and see the past!


Exactly. I think it'd be incredibly cool to walk in front of a thin transparent wall with this property, run around to the other side and see yourself walk along it.

I wonder if you could information in this fashion.


Not if you consider the resolution would be much higher then anything technically feasible. You might not now what you want to focus on before hand, but if the system is setup you can go back and decide what to look at.


Mars, at the widest distance from Earth, is 401,000,000 km away

A lot of the time, it's closer. And when it's the furthest away, both earth and mars are on different sides of the sun. A star between you and the mirror will make your mirror hard to use.


rmc, good point about the specifics of Mars. The range I found is 100-400 km. I think the (useless?) point I was making still more or less holds, but my calculations need asterisks next to them with a clarification at the bottom: "Would require you to be able to see through the Sun"


> The Moon is almost exactly 1000x closer to the earth than Mars[3], so I wonder if I used it for my mirror array instead if I could just put a station with 500 mirrors on it to accomplish the same thing.

Plus, since the moon has synchronous rotation, the mirors would at least always be facing us. Assuming all the other massive difficulties got worked out, it'd be a good place for those mirrors.


One problem with this plan is that mirrors are not perfect and some non-negligible fraction of the incoming light is absorbed and/or scattered away. Even if you could build perfect mirrors, you would still likely need to place the whole setup under vacuum to prevent the light from being scattered by air.


You are right that regular mirrors absorb or scatter light energy. However, to get 100% reflection it should possible with a setup where the light is getting refracted or bounced off critical angle. http://en.wikipedia.org/wiki/Total_internal_reflection This is commmonly used in Fibre optics.

Its more of an engineering challenge to come up with a setup to view your past than a theoretical one!


The scattering problem is one that confused me, too. At a light-year or several dozen away, there's no way you could find and piece together all the photons that bounced off the moment you're looking for. Am I misunderstanding something?


I don't get it - even if you use a mirror, the distance is still the same. Light needs to go to the mirror, reflect and come back to your eyes, there's no difference if you're n lightyears away or just n/2 and use a mirror.

me ------| <- mirror

me ---|--- me

Do I get it wrong?


farico, you are exactly right, and that is the point. You just add mirrors to make the distance needed shorter and shorter.

I was just taking the example to a logical absurdity to reason that we could potentially make a small array of millions of mirrors, here on earth, that could see back in time in "real-time" if you pointed it at something.

But as I soon realized, this is basically a video camera =(


I think this is still a worthy experiment to try out. My guess is that it would cost in the 100s of millions. Would be nice if there was a cheap way to see 1-2 seconds in the past ... cheap enough high school for science classes.


But if you could travel faster than the speed of light, you could send a mirror to the edge of the universe to figure out what happened in the past.

Figure out what happened to the tops of everyone's heads when they were outside, that is...


Or, for a more down-to-earth implementation, you might use a light-slowing material http://en.wikipedia.org/wiki/Slow_light


What? Nobody has mentioned a worm hole yet? Screw mirrors to cut the distance needed in half (although requiring the same amount of time).

How about this:

Step 1) open up a worm hole at a point along the path of light emanating from Florida on June 16th, 2008. 2) install a video camera with a giant lens around the same point pointing back at us 3) send the wireless ;) signal back through the worm hole. 4) Determine if the Anthony jury is a bunch of idiots or not.


This is something I have always wondered about.

Can the reverse be true too? i.e. if you are on the moon and can see what happens 1000km away, before someone else on earth 2000km away from that event, is that the future? Is that possible?

Can that then be extended to 27 lightyears into the future too?


Yes, but you could never signal faster than the information arriving on earth in a straight line at the speed of light.

To understand why picture the triangle between the start point (A), end point (B) and the moon (M). The distance AMB is larger than AB.

Google or Wikipedia 'light cone', pretty interesting stuff.


The man in the moon is not really observing the future he just becomes aware of the event before the man on earth.

If it wasn't light that was being transmitted the man on the moon could warn the man on earth about what's happening before it's reaching him in the same way that people near the epicentre of an earthquake can use twitter to warn people further away before it hits: http://recovery.doi.gov/press/us-geological-survey-twitter-e...


I think it is relative. Your argument considers time as a non relative entity and the past and present as static points in a timeline. So the same "event" is one man's past and one man's future. It depends on your space coordinates from the coordinates of the even if this is your past or present or how back in past.

Disclaimer: This is deduced logic and not scientific information.


It's not the future, it's just an earlier indication of the event - seen at 1000km versus 2000km. The event still happened at the same time.


"At the same time" with respect to its own frame of reference.


Yes.


forget 27 years, give me 1 minute and I'll be a rich man( I'll do sport betting :D )


To circle around a bit back to a hacker news favorite topic - this is exactly why high-frequency traders put their equipment in data centers as close to the action as possible. They're seeing the and trading on the future if your frame of reference is some data center in California.


Until just now, I always thought this was a silly practice. My line of reasoning was "What difference does a 30ms ping make?" I guess I hadn't thought about it too much though.

Now, after reading a comment elsewhere in this thread, which talks about how a cpu can execute six instructions in the time it takes the light from my monitor to reach my eyes, the data center thing seems a lot more reasonable.


Somewhat of a tangent, but the exploitation of this very phenomenon is one of the things that endears the book "Battlefield Earth" to me. Despite whatever other issues the author had, he wrote a darned good science fiction book.


I've never read Battlefield Earth, but I found The Forever War to be a really interesting view on the effects that relativity would have on light-year-scale war (and the expansion of the human race).

http://en.wikipedia.org/wiki/The_Forever_War


We definitively can, but there is an issue: The light can be absorbed by some substance or refracted. How do you account for that?

You can account for that by recording every atom/photon on the space and then using some kind of miraculous processing power, to calculate how everything goes and estimate/verify the past actions.

For the processing ability, certainly some day we'll get there. For the possibility of getting the coordination/nature of any atom/photon in the whole space, this will need a discovery of something faster (way too much or may be instantaneous) than the speed of light that gives us the ability to recognize particles.

This is not amazing. This is sick. We'll be able to watch Pharaohs with an infinite precision. See how the earth was billions of years ago and how life evolved.


You want to record / analyze every atom in the universe with a machine that exists within the universe?


Just for fun... it might not be impossible:

You can reduce R^2 to {x | 0<=x<=1}[1] and thus, by induction, we can reduce R^n to {x | 0<=x<=1} for any integer n.

If I remember correctly, there are about 10^80 particles. The position of each particle is a point in R^3 so the position of all particles is R^(3*10^80). So, the position of every particle could be stored by the position of a particle on a 1 meter (foot, inch, whatever) long stick.

Of course, you run into problems if space is discrete or, in any event, with the Heisenberg uncertainty principle but you can still store a lot of information with each particle.

Horribly impractical of course, but like I said, just for fun...

[1]http://www.earlham.edu/~peters/writing/infapp.htm


Your argument from cardinality of sets is not relevant -- current state of knowledge of nature of the universe prohibits building a device which carries on with computation using real numbers. See Bekenstein bound.

Actually, if we could perform computations using real numbers (think of it as we're back using analog computers and the universe is continuous again and not discrete/quantum), we would for instance be able to solve NP-complete (also #P-complete) problems in polynomial time.

Anyway, the rest of your argument is what philosophers were arguing about two, three and even four hundreds years ago. See Wikipedia pages for "Determinism" or "Mechanism".


"The position of each particle is a point in R^3 so the position of all particles is R^(310^80)."

There's a theory that says the the maximum information content of a region of space is proportional to its surface area*. (Honest!) This result is related to black hole physics: information has to be represented by mass and/or energy, and thus curves the space-time continuum according to general relativity. Try to put too much information in a region of space and it gets cloaked by an event horizon.


My memory on this is vague, but didn't Richard Feynman actually look at this problem?


one of the commenters: "the question seems to lead to the idea that traveling faster than the speed of light == traveling backwards in time"

It's just a point of language, but that's silly. Seeing a video of France is very different from traveling to France, so one would imagine that seeing the past would not be described as "traveling" backwards in time.

edit: although, you may think about going faster than the speed of light as traveling backwards in time because of the "effects" of time dilation when your speed is > c


Faster than light travel isn't classical time travel, until you turn around. If you only travel away from Earth FTL, then that doesn't automatically produce nasty paradoxes. However, if instead of sticking around to see all the light pass you by, you just turn around and head back to Earth, again FTL, then you can arrive before you left, and cause all sorts of trouble.


Strictly speaking, everything we see happened in the past :)


err.. the first sentence of the link:

  From my understanding of light, you are always 
  looking into the past based on how much time 
  it takes the light to reach you from what you 
  are observing.


We cannot even resolve the lunar lander site on the moon with Hubble.

How the heck exactly are you going to take advantage of that distance or mirrors?


Or, you can videotape yourself and then watch the past whenever you want.


The question has been protected on stackoverflow, so answering here. Yes a mirror 13.5 light years away can do the trick. But there is some fundamental lack of understanding regarding relativity that has lead to such a question. In fact, the top answer is incorrect on several levels.

1. Technically the only way a person can travel 27 light years in 27 years is by travelling through out at the speed of light. Ignoring the infinite amount of energy required for such travel, there is another aspect of such travel that is not being considered. Time Dilation. So assuming you were born and take off at the speed of light. Yes you will be able to observe your birth on earth 27 years later, but you would still be a baby and not 27 years old. Time would have been effectively frozen for you as you traveled.

2. The top answer says that you are always viewing and hearing the past. You are hearing the past, but you are actually viewing the present. The light radiating out of an event you are observing are also causality horizons. In your reference frame you are observing past events depending on the distance. However thanks to relativity and Lorentz contraction, there are reference frames where the distance between you and the event can be arbitrarily close to zero, effectively making the events simultaneous i.e. when you see an event you are effectively watching it as it happens simultaneously, you are not watching the past. This is also an alternative explanation for 1. i.e. the baby cannot grow older if you put it 27 light years away in 27 years time. It will be able to watch its birth, because its only just been born.

http://en.wikipedia.org/wiki/Minkowski_diagram


If you could figure out a way to invert all the constants and principles in the universe, then we could relive our past, in reverse. tea cup fragments on the floor would assemble into a teacup then accelerate up into the air and then sit on the table. The universe would play out, right back to it's formative moments. Then you could watch yourself being born, though you would have to wait years for it to render.


I don't understand the physics but I do believe Stephen Hawking disproved this[1]; he said that after entertaining the idea that as the universe contracts time might reverse he concluded later this was wrong.

[1] A Brief History of Time (the documentary film)


I wonder if there is research into gravitational lenses or other possible sources of naturally occuring space mirrors or lenses that we could use to collect light rays from the earth's past, maybe to detect its spectral content.


There isn't any research, but only cuz it's damn impossible. The number of photons ever emitted by Earth in the distant past which are gonna ever come back to Earth is... very very small.


Yet, since there are not many similar atmospheres in the region, if there were any backscattered radiation it wouldn't be impossible to detect.


this idea reminds me of how during the cold war the US military used the moon as a reflector to measure USSR air-defense radio signatures. they were able to determine the capabilities (range, frequencies etc.) of USSR air defense without ever entering the air space.

you would assume that light from earth is reflected back from objects in deep space. the problem would be sorting out all of the weak signals and finding what you are looking for by calculating where the reflection would end up and going out and finding it. the signal would likely be too weak to interpret with today's technology, but in theory it would be possible.

(another theory is with light being bent by gravity it is possible that light from earth has been bent back around to pointing back to us (ie. around a black hole). we just don't know how to find it or where to look)




Consider applying for YC's W25 batch! Applications are open till Nov 12.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: