Collumbia STS 80 footage shows circular formation of seven objects

Hi. Do the objects in the STS 80 footage assume a perfectly circular formation that lies parallel with the earth's surface?

I refer to the following footage from collumbia STS 80 mission.

www.youtube.com...

I've painstakingly produced evidence the objects have a circular formation by testing whether the seven objects conform with an ellipse at 2:56 in the footage, based on the hypothesis that the central object (shown in white on graphic) becomes illuminated when the formation is in place.



The rationale is that any cross section of a circle is a sphere, and rotating a circle gives an ellipse. Irrespective of the earth's curvature and angle in xy plane of vision and z axis, a circular formation of objects at equal altitude will fit an ellipse. The ellipse should be parallel with earth's surface. The blue line in background of image is the horizon as filmed in the footage. The major axis of the ellipse lies parallel with horizon at its center.

The probability of this occurring by chance, if debris or dust, tends to zero as the degree of correspondence with the model tends toward being perfect.

This ain't debris.

I tested this graphically by plotting the coordinates of the seven objects, as measured on-screen. I modelled the formation with an ellipse, and obtained a near perfect fit to the model, as shown in the graphic above. I hope to find time to precisely state how much variance among the measurements in the two dimensions is explained, but it is clearly a very high proportion.

I estimate the angle between the major axis of the ellipse and the horizontal axis as ~0.192 rad (11 degrees). This agrees closely with the estimated tilt of the horizon, as required by the model. I estimate the angle between the face of the camera and the ojbect as ~0.76 rad (43.7 degrees) if the model is correct.

I'd appreciate help overlaying the ellipse on the actual footage. Someone "joined the dots" on the objects here:

www.abovetopsecret.com...

However, as you can see above they are much more precisely modelled with an ellipse. An ellipse is what you would expect to see if it is a perfectly circular formation viewed at an angle (tilted away relative to plane of camera).

Cheers, 9

[edit on 21-4-2008 by 987931]

Thats one weird film. I've watch that many times since it showed up awhile back.
With all the info you gave above I was thinking..maybe the aliens are sending a message. A "simple" message e.g. "we shoot hoops also" or
a picture of their planet with one moon (the center light ) or a warning! like "this circle is you, and this is going to hit you! (last light in center).

I'm reading a book now about a alien probe (new in paperback) that is being
excavated from the bottom of the ocean (its in the Moho layer) and its giving signals. After the scientist try every thing to decode them (one signal is x=1/0 )
one dude finally comes up with the meaning of the code ...ITS TOXIC TRASH DON'T TOUCH!! (to make a very long story short)

maybe someone here can figure out what they might be saying in that ST-80 sighting?

Originally posted by RUFFREADY
With all the info you gave above I was thinking..maybe the aliens are sending a message. A "simple" message e.g. "we shoot hoops also" or
a picture of their planet with one moon (the center light ) or a warning! like "this circle is you, and this is going to hit you! (last light in center).

:-P

Personally, I'd hypothesize that they took the opportunity to get into a circular formation because it is guaranteed by geometry to be recognizable and is therefore also identifiable.

It is an identified formation of unidentied objects ... an IFUFO :-)

That is, it would stand to reason they're saying hello ... anyone can identify this as not debris, dust etc. and and 'try explaining this one away'.

I know what you mean -- I wonder whether there's more in there. But it took many times looking at it to realise this is probably no rough formation, but instead a perfect one.

9


[edit on 21-4-2008 by 987931]

Good work, star & flag from me!

I'll review your methodology when I have more time. Just a few remarks in the meantime.

First, you do realize that a circle seen in perspective is not an ellipse, right? The difference is small with a telephoto lens, bigger with a wide angle. OK, I'll stop nitpicking.

Second, choosing the best matching frame makes a coincidence much less unlikely.

Third, the error margin is defined by the (very low) resolution of the video.

They are forming a deep space antenna.....

Earth-bound astronomers have been utilizing immense arrays of individual receivers, or very big radio antennas (Aricibo) for years. That's what you're seeing here...a really big antenna for extremely long wavelength communications.

I'm just guessing.....

Originally posted by nablator
Good work, star & flag from me!

I'll review your methodology when I have more time. Just a few remarks in the meantime.

Thanks very much, look forward to more input.

First, you do realize that a circle seen in perspective is not an ellipse, right? The difference is small with a telephoto lens, bigger with a wide angle. OK, I'll stop nitpicking.

Err, yes, but not sure I follow entirely -- let me be clear, I know more about acoustics than optics. I'm assuming the formation is a very long way away, and do realise not perfect ellipse, but don't understand why a difference between telephoto and wide angle. Would appreciate more info.

Edit and qn: sorry, being silly, the distance makes no difference, the ratios (or proportions) are invariant e.g. distance from centre to our side slightly larger than centre to far side, and ratio of these distances fixed or invariant ... correct?

Second, choosing the best matching frame makes a coincidence much less unlikely.

Yes, among other things, I am basically a statistician. If it is a near perfect fit, there is near zero probability it is chance. Tiny probabilities remain tiny even with many trials (we can think of each frame as a 'trial' in the statistical sense). Also, it is in formation for a period not an instant. You are quite right in principle, though, and this becomes a bigger consideration the more the observations deviate from perfect fit.

Third, the error margin is defined by the (very low) resolution of the video.

In part, as far as it can be quantified, I would say (can explain why I think this). Simple way to quantify measurement error is repeated measurements, but it would be even more compelling if this agreed with an estimate based on resolution. I have a hell of a lot of experience fitting data to models, though, and it is a damned good fit. I still want to quantify it thoroughly, though, particularly if I can get expert advice on the optics.

Thanks again and look forward to further comment. 9

[edit on 21-4-2008 by 987931]

[edit on 21-4-2008 by 987931]

Originally posted by 987931
Err, yes, but not sure I follow entirely -- let me be clear, I know more about acoustics than optics. I'm assuming the formation is a very long way away, and do realise not perfect ellipse, but don't understand why a difference between telephoto and wide angle. Would appreciate more info.

Distance doesn't matter, but the angle does. Most cameras have optical aberrations that make wide angle images distorted. Just nitpicking, the difference is probably too small to be noticed.

Yes, among other things, I am basically a statistician. If it is a near perfect fit, there is near zero probability it is chance. Tiny probabilities remain tiny even with many trials (we can think of each frame as a 'trial' in the statistical sense).

Then you know the problem with statistical evidence is that you need lots of samples to firmly establish a deviation from the expected probability. Number of occurrence: one (maybe?) in who knows how many hours of NASA TV video. Theoretical probability: unknown, but possibly tiny. So what? If you win the national lottery, will you complain?

Originally posted by nablator
Distance doesn't matter, but the angle does. Most cameras have optical aberrations that make wide angle images distorted. Just nitpicking, the difference is probably too small to be noticed.

OK, thanks .... and, for the record, I like nitpicking.

Then you know the problem with statistical evidence is that you need lots of samples to firmly establish a deviation from the expected probability. Number of occurrence: one (maybe?) in who knows how many hours of NASA TV video. Theoretical probability: unknown, but possibly tiny. So what? If you win the national lottery, will you complain?

You don't need lots of samples if you have a rational and defensible first hypothesis*. I'd expect randomness in both dimensions at any given point and simulations can be used to see how much variance is explained given the number of parameters required to fit an ellipse to the data (not many).

The expected correlation is zero between coordinates in the horizontal and vertical axes. Given this, the variance in the two dimensions is additive. I'd think an F-ratio could be used to produce a probability for a given frame, though I haven't fully worked through it.

Also, I prefer to minmize reliance on statistics to the extent possible, but it can be a useful part of a case.

On winning lotto, yes, but this can be turned around. The only reason someone (usually) wins is becomes there are millions of trials.

If you can get an estimate of probability, and it is tiny (which it will be), then you can the state for example: NASA would have to have X number of hours of footage containing debris fitting specific criteria for even, say, a 1% chance of seeing this formation.

And it's not hours of footage without qualification, it's hours of footage with with visible objects that fit any criteria clearly relevant to the hypothesis. The more defensible criteria you have, the less footage meeting those criteria can be expected. For example, IF the formation exists for several seconds, then any footage that cannot possibly have such a formation (e.g. debris flies past rapidly after dump) cannot be relevant. NASA cannot have it both ways: they can't claim everything under the sun (pun intended) is debris then claim any debris are capable of displaying every behaviour! :-)

Having said this, as I said above I do agree that it's best to avoid statistics to the extent possible and it would be better to find lines of corroborating evidence. For example, if the altitude of the caraft is known, it is likely possible to estimate altitude of the objects GIVEN the hypothesis. There are at least ten seconds to see whether five objecs fit with the model of being in formation AND in a synchronized orbit (not necessarily geosync but with each other at least).

Also, I'd hypothesize the central object is central but at a higher altitude, which means it would have to lie in a line projecting from earth's center through center of formation.

I imagine with a few things established, others would be able to put forward more 'predictions'.

Cheers, 9


*I detest the concept of a null hypothesis for various reasons, but it is not nearly so bad to state two complementary alternate hypotheses and look at complementary conditional probabilities given each.

Originally posted by 987931
NASA cannot have it both ways: they can't claim everything under the sun (pun intended) is debris then claim any debris are capable of displaying every behaviour! :-)

Yes, if the 7 UFO stayed perfectly lined up for at least a few seconds, or if the same formation were repeated several times, it would be compelling evidence.

I tried fitting an ellipse at 2:55 with PhotoShop, it worked well enough, not perfectly, but near enough (a few pixels). At 2:56 it was impossible already. This apparently random drift is not an evidence of intelligent behavior.

How unlikely is the elliptical formation? Any 5 points can be fitted to a conic (ellipse, parabola, hyperbola). The likelihood of the ellipse is 50% maybe, or nearly 100% if you can choose points (you eliminate 2 in the STS-80 video, one in the center, one on the right). You get the 6th point essentially for free, just by waiting for the right frame. The only interesting point is the 7th, and it never exactly fits the ellipse as far as I can see with my little PS experiments. You are underestimating the likelihood of the quasi-elliptical geometry.

Having said this, as I said above I do agree that it's best to avoid statistics to the extent possible and it would be better to find lines of corroborating evidence.

Yes, but so far you don't have any.

For example, if the altitude of the caraft is known, it is likely possible to estimate altitude of the objects GIVEN the hypothesis. There are at least ten seconds to see whether five objecs fit with the model of being in formation AND in a synchronized orbit (not necessarily geosync but with each other at least).

Aren't you a bit overoptimistic? I don't see any synchronization during 10 seconds.

Also, I'd hypothesize the central object is central but at a higher altitude, which means it would have to lie in a line projecting from earth's center through center of formation.

That's wishful thinking, not evidence.

I am not convinced yet.

Thanks for the thought-provoking thread.

Originally posted by nablator
I tried fitting an ellipse at 2:55 with PhotoShop, it worked well enough, not perfectly, but near enough (a few pixels). At 2:56 it was impossible already. This apparently random drift is not an evidence of intelligent behavior.

The graphic I posted was based on measurements at 2:56. For what reason do you say it was "impossible already"?

How unlikely is the elliptical formation? Any 5 points can be fitted to a conic (ellipse, parabola, hyperbola). The likelihood of the ellipse is 50% maybe, or nearly 100% if you can choose points (you eliminate 2 in the STS-80 video, one in the center, one on the right).

Fair point about conic, when considering hypothesis 1 (debris), but I really have no idea how you make the leap to nearly 100% or what is the second point eliminated (the central one OK, more below). There is a light spot near "blinking" object to bottom right, is that the one? (its on-screen coordinates remain fixed throughout footage).

In any case, though good point, I'm imposing a very strong constraint -- that the major axis of the ellipse is parallel with the horizon's tangent at intersection of minor axis and horizon. This massively reduces the class of possible ellipses based precisely on the base rationale of the hypothesis. Hence the importance of the number of parameters: major axis length and eccentricity. And, if I can find out the altitude of craft, it may not be necessary to estimate even eccentricity, massively reducing the class yet again. I've discussed this all quite a bit with an engineer who has constructed complex structures, rather pedantic (family member).

You get the 6th point essentially for free, just by waiting for the right frame. The only interesting point is the 7th, and it never exactly fits the ellipse as far as I can see with my little PS experiments. You are underestimating the likelihood of the quasi-elliptical geometry.

I agree I've underestimated somewhat, and appreciate your points, but I think you're also underestimating the degree of constraint I'm imposing.

Aren't you a bit overoptimistic? I don't see any synchronization during 10 seconds.

I don't follow. I said there are ten seconds to SEE (put to a fairly rigorous test) whether they are synchronized. I haven't done anything but prelim work. The preliminary, and fairly crude, measurements indicate that the distance of three objects relative to central object remain fixed and the other two change marginally. I think it is worth looking at, but I stress again I only claimed it is possible to test, I didn't claim to have shown it! There is no evidence of camera movement or zoom and the objects remain continuously visible.

Also, I'd hypothesize the central object is central but at a higher altitude, which means it would have to lie in a line projecting from earth's center through center of formation.

nablatorThat's wishful thinking, not evidence.

That's wishful anti-thinking, not counter-evidence (just making a point). Again, I really don't follow how stating hypothesis can be either 'optimistic' or 'wishful thinking'. I think you're just assuming to know someone's specific thoughts, which is para .... no, won't go there, this is actually one of the most civil and constructive debates I've ever had online, I'll try to keep it on that track :-)

Look, I'd have a lot to do and I doubt I'd ever get to it. I'm putting out some calls for help. Your constructive criticism and input is help, and with a bit of luck (now optimism) I'll get some more.

Final thing, you said something about not having the stats. As I think I said, I have a hell'v'a lot of experience fitting data to models. I know I don't need them right now to keep going. I've seen many thousands of graphical tests ultimately backed (or not) by stats.

Cheers nabalator,

9

[edit on 22-4-2008 by 987931]

Further to last post, I see you were saying I don't have corroborating evidence, you weren't referring to stats (though it is one form).

Clearly, there is no chance of getting corroborating evidence if you have no further hypotheses consistent with first or predictions based on it. So, no, I don't have it but I have thought about whether it's possible, in principle, to get some!

Believe me, if the angle of the ellipse was not very close to the angle of horizon's tangent, I would have forgotten all about this by now. Oh, and I forgot to mention a third parameter above, the center of ellipse.

Alas, wish I had time to have done all of the work and obtain it all, but I do appreciate your input, as I said.

[edit on 22-4-2008 by 987931]

Originally posted by 987931
The graphic I posted was based on measurements at 2:56. For what reason do you say it was "impossible already"?

Maybe you captured a frame at 2:56.1, and I captured a frame at 2:56.9. I don't know. In less than a second there is enough drift to make it impossible. I just wanted to see how difficult it is to find the best frame. Now I know it is difficult, the elliptical shape is more a coincidence than a formation IMHO.

Fair point about conic, when considering hypothesis 1 (debris), but I really have no idea how you make the leap to nearly 100% or what is the second point eliminated (the central one OK, more below).

Well, I was thinking any 5 points that look like they are on the same ellipse are on the same ellipse. Maybe they only need to form a convex polygon. Not sure.

There is a light spot near "blinking" object to bottom right, is that the one? (its on-screen coordinates remain fixed throughout footage).

No, it's near and to the left of the rightmost point of the ellipse.

In any case, though good point, I'm imposing a very strong constraint -- that the major axis of the ellipse is parallel with the horizon's tangent at intersection of minor axis and horizon. This massively reduces the class of possible ellipses based precisely on the base rationale of the hypothesis.

That's impressive. You win.

I could say, for the sake of argument, that it is also a coincidence. You can't fight coincidence. I win.

Maybe the eccentricity is 0.6. What are the odds of having a 0.6 eccentricity? See? It does reduce the class of possible ellipses, but there is only one ellipse to test the property on.

I agree I've underestimated somewhat, and appreciate your points, but I think you're also underestimating the degree of constraint I'm imposing.

Yes. I'm trying my best to help you build a balanced argument.

Look, I'd have a lot to do and I doubt I'd ever get to it. I'm putting out some calls for help. Your constructive criticism and input is help, and with a bit of luck (now optimism) I'll get some more.

Your points are excellent. You are not afraid of constructive criticism. I like your optimism. Good luck!

In any case, though good point, I'm imposing a very strong constraint -- that the major axis of the ellipse is parallel with the horizon's tangent at intersection of minor axis and horizon. This massively reduces the class of possible ellipses based precisely on the base rationale of the hypothesis.

That's impressive. You win.

I could say, for the sake of argument, that it is also a coincidence. You can't fight coincidence. I win.

I assume you know how and why probabilities may multiply to give vanishingly small probabilities of multiple "coincidences"? I can give a few examples from science if you like. I'm not arbitrarily specifying the constraints, they're based on a reasoning and observation, just like any well-formed scientific hypothesis. They're based on a single parsimonious explanation stated as hypothesis.

On there being enough drift to make it impossible, I have spent hours doing measurements. For now, all I'll say is I think that's plain wrong, but I will get data to you if you want it.

Maybe the eccentricity is 0.6. What are the odds of having a 0.6 eccentricity? See? It does reduce the class of possible ellipses, but there is only one ellipse to test the property on.

My point was that if I know the altitude of the craft, I can estimate the angle to the horizon, then calculate distances that allow an esimate of the angle to the center of the hypothesized circle. From this, the eccentricity is given by basic trigonometry. It is then a testable an integral part of the hypothesis.

Everything independent that clearly follows from a simple hypothesis makes the test of that hypothesis more powerful (precisely because probabilities multiply if independent).

Yes. I'm trying my best to help you build a balanced argument.

Yeah, I know, and I appreciate that.

You are not afraid of constructive criticism. I like your optimism. Good luck!

Thanks, it's appreciated.

[edit on 22-4-2008 by 987931]

Originally posted by 987931
I assume you know how and why probabilities may multiply to give vanishingly small probabilities of multiple "coincidences"?

Yes, I also know it's easy to "prove" things using math.

Example 1: the Drake equation.

Example 2, just for fun: let's prove the Indian profile in Canada found with Google Earth is not a natural geological feature:
Chin: probability 10^-3
Mouth: probability 10^-4
Nose: probability 10^-4
Eye: probability 10^-3
Forehead: probability 10^-2
Indian headdress: probability 10^-5
Ear: probability 10^-2
Overall consistency: probability 10^-8
--------------------------------------------------
Probability of a natural occurrence: 10^-29!

Corroborative evidence is everything.

On there being enough drift to make it impossible, I have spent hours doing measurements. For now, all I'll say is I think that's plain wrong, but I will get data to you if you want it.

I don't doubt your results. How many frames did you capture and test? Only one? Probably I didn't capture the same one. Or the ellipses in PS are not elliptical.

My point was that if I know the altitude of the craft, I can estimate the angle to the horizon, then calculate distances that allow an esimate of the angle to the center of the hypothesized circle. From this, the eccentricity is given by basic trigonometry. It is then a testable an integral part of the hypothesis.

Wow.
Good luck.

edit: to add link

[edit on 2008-4-22 by nablator]

Yes, I also know it's easy to "prove" things using math.

What you know is that it's easy to abuse statistics. What you don't know is the fundamental reason for abuse of most use of statistics. You're also apparently missing or choosing to ignore genuine principles becauese of (trite) examples of abuse.

I'm not plucking numbers out of the air. Simple, clear hypothesis. Circular formation of objects of equal altitude.

Specific, refutible, testable.

I don't doubt your results. How many frames did you capture and test? Only one? Probably I didn't capture the same one. Or the ellipses in PS are not elliptical.

Possibly. I tested several frames, which is why I don't think your results can by right. Not as carefully, I didn't make repeated measurements -- the one I posted took the longest. But I very much doubt the others can be that far off the first.

Good luck.

Thanks if you're serious.

[edit on 23-4-2008 by 987931]