PDA

View Full Version : BB fails time dilaton



lyndonashmore
2006-Jun-16, 06:16 PM
This is a very good paper. (http://www.journals.uchicago.edu/ApJ/journal/issues/ApJL/v626n1/19414/19414.web.pdf)

Firstly it admits once again that despite what posters here keep saying:


Despite the predominance of the current redshift paradigm, little direct observational evidence exists in support of cosmological expansion.

Secondly, it tells us that in measuring “time dilation” some of the techniques used in the past assumed a 1 + z dilation in the first place – and then told us that the results showed time dilation existed!

Thirdly, it tells us that an expanding universe predicts the wrong results.

Stationary Sn Ia’s have spectral features separated by 17 and 18 days.
The BB predicted 18 and 23 days for this particular supernova.
In actuality the observed separations were 25 and 31 days.

So the BB got it wrong didn’t it!
Cheers,
Lyndon

Eta C
2006-Jun-16, 06:59 PM
Lyndon, intentionaly or not, you are being intellectually dishonest here. You picked one small quote out of that paper and have totally mis-represented their result. In fact, their results are consistent with expansion and exclude tired light and theories such as Arp's. To quote the abstract.


This result is inconsistent with no time dialtion with a significance level of 99.0%, providing evidence against "tired light" and other hypotheses in which no time dilation is expected. Moreover, the observed timescale of spectral evolution is inconsistent with that expected in the "variable mass theory." The result is within 1 sigma of the ageing expected from a universe in which redshift is expected by cosmic expansion

From the discussion section (part IV)


We have shown that the spectral evolution of SN11997ex has likely been dilated by a factor of 1 + z, as expected in an expanding universe. This result also shows that hte null hypothesis of no cosmic expansion is ruled out at the 97.9% level. Thus "tired light" and other hypotheses that predict no time dilation are essentially eliminated.

Sounds to this reader as if they're confirming cosmic expansion. You've given me ample reason to never take you seriously again (assuming I ever did.)

edit to add. A closer look at the paper shows that Lyndon has misinterpreted the numbers. The 18 and 23 days are those predicted in the SN rest frame. We observed the time dilated delays of 25 and 31 days. When you take the red shift of the SN and use that to boost the observed values to the SN rest frame you get 16.97+-2.75 and 18.01 +- 3.14 days. These are in agreement with the predicted values within the significance levels quoted. So you're right lyndon. This is a great paper. It confirms expansion and excludes the ideas of Arp and other hypothesists.

lyndonashmore
2006-Jun-16, 07:11 PM
On a point of information here eta C.

picked one small quote
Since when did size matter. That was what was said and that is what was meant.
Who is being intellectually dishonest here, intentionally or not?
It does not ‘exclude’ tired light theories as you say but gives the probability of no curve broadening as small 1% or so – the same as a horse coming in at 100 to 1 (remember fionhaven in the grand national?)
The point that you seem to have missed here is that the BB’s predictions are way out and so time dilation does not ‘prove’ expansion as many posters here seem to assume.

Metricyard
2006-Jun-16, 07:11 PM
I propose a new scientific method, googleminingology. No need to read the article, just post what needs to prove a theory, the real science in the article is all fluff.

Eta C
2006-Jun-16, 07:14 PM
On a point of information here eta C.

Since when did size matter. That was what was said and that is what was meant.
Who is being intellectually dishonest here, intentionally or not?
It does not ‘exclude’ tired light theories as you say but gives the probability of no curve broadening as small 1% or so – the same as a horse coming in at 100 to 1 (remember fionhaven in the grand national?)
The point that you seem to have missed here is that the BB’s predictions are way out and so time dilation does not ‘prove’ expansion as many posters here seem to assume.

You say that the paper does not agree with BB cosmology when if fact it does. Intentionally or not, you did not understand what you read. The paper does not say what you claim it does. That is intellectual dishonesty in my book.

lyndonashmore
2006-Jun-16, 07:23 PM
OK. lets look at the facts and only the facts and let the reader decide.
The Bb predicts a time dilation of 1.31 days between the first two features.
The observed delay is 7.91 days.
In my book that is an incorrect prediction.
In your book.....

Eta C
2006-Jun-16, 07:34 PM
Well, let the reader read the paper, but here are the comparisons.


Observed Boosted to SN rest frame Predicted in SN rest frame

24.88 18.28 16.97 +/- 2.76

30.95 22.74 18.01 +/- 3.14


The observed values are those we saw here on earth. The boosted values change those numbers based on the red shift of the supernova. The predicted values are those expected on what we know about the physics of supernovas. They are consistent with the boosted observed values at the quoted confidence level. If there was no time dialtion and no expansion we would have expected to observe values of 17 and 18 here on earth, not the 25 and 31 we actually did.

I will admit to speaking in haste and withdraw accusations of dishonesty. But it is clear to me that you have mis-read the paper and do not understand what the authors did and what they are saying. The paper supports BB cosmology and excludes the ideas of Arp and others.

Celestial Mechanic
2006-Jun-16, 07:36 PM
Here's the abstract for those who don't want to load a PDF. (I think I can quote that much but no more under fair use.)

We have obtained high-quality Keck optical spectra at three epochs of the type Ia supernova 1997ex, whose redshift z is 0.361. The elapsed calendar time between the first two spectra was 24.88 days, and that between the first and third spectra was 30.95 days. In an expanding universe where 1+z represents the factor by which space has expanded between the emission and detection of light, the amount of aging in the supernova rest frame should be a factor of 1/(1+z) smaller than the observed-frame aging; thus, we expect SN 1997ex to have aged 18.28 and 22.74 days between the first epoch and the second and third epochs, respectively. The quantitative method for determining the spectral-feature age of an SN Ia reveals that the corresponding elapsed times in the supernova rest frame were 16.97+/-2.75 and 18.01+/-3.14 days, respectively. This result is inconsistent with no time dilation with a significance level of 99.0%, providing evidence against "tired light" and other hypotheses in which no time dilation is expected. Moreover, the observed timescale of spectral evolution is inconsistent with that expected in the "variable mass theory." The result is within ~1 sigma of the aging expected from a universe in which redshift is produced by cosmic expansion.
I quote; you decide. :D

lyndonashmore
2006-Jun-16, 07:47 PM
Thanks eta C,
The thrust of this thread is not to push steady state models but to point out that supernova light curve broadening (that is the observaton - not time dilation) does not 'prove' expansion - as is oft quoted.
Quasar light curves are not stretched.
We see here that the Bb predictions are over 500% out.

Are we still sure about these effects being 'time dilation?'

Eta C
2006-Jun-16, 07:55 PM
Well, if you believe the paper, yes. You started this thread by linking to a paper and claiming that pieces of it disagreed with expansion when in fact they are consistent with it. Hence my somewhat righteous indignation.

The discrepencies you quote are well within the statistical error quoted. The 17+-2.75 one is right on and the 18+-3 is just outside the 1 sigma mark, something that is expected to happen about 30% of the time. Remember, this is for just one supernova. Should the sample size be expanded to more than that, the stats should get better. So in this case, yes, the SN light curves are consistent with those expected in an expanding universe. Translated from the "scientese," yes, the SN light curves prove that time dilation is occuring in the manner predicted by an expanding universe.

lyndonashmore
2006-Jun-16, 08:02 PM
The 17+-2.75 one is right on
right on what?
This is the rest frame value.
The BB predicted a delay of 1.31 days when the observed delay was 7.91 days.
What s "right on" about that?
Cheers,
lyndon

antoniseb
2006-Jun-16, 08:12 PM
Hi Lyndon, You are right that this is a good paper, thanks for bringing it to our attention. One of the things that I think is most interesting about this paper is that it shows the imprecision we are dealing with in making these measurements with the state of the art (eight years ago) tools available to us. This paper points out (as noted in posts above) that the scale of the over estimate of the observed time dilation is about one sigma, and should happen a third of the time. Based on this one measurement, that's about all you can ask for.

Now, this observation was made eight years ago, and was just published now. I'd really like to know how many similar measurements have been made since, and whether we are now capable of describing a statistically more significant sample size.

lyndonashmore
2006-Jun-16, 08:18 PM
Hi Lyndon, You are right that this is a good paper, thanks for bringing it to our attention. One of the things that I think is most interesting about this paper is that it shows the imprecision we are dealing with in making these measurements with the state of the art (eight years ago) tools available to us. This paper points out (as noted in posts above) that the scale of the over estimate of the observed time dilation is about one sigma, and should happen a third of the time. Based on this one measurement, that's about all you can ask for.

Now, this observation was made eight years ago, and was just published now. I'd really like to know how many similar measurements have been made since, and whether we are now capable of describing a statistically more significant sample size.
So, we must ask:
"in view of the imprecision, why is the broadening of supernova light curves cited as a new 'pillar' of the bb?
Surely the jury must still be out.
This result happens one third of the time by chance - does that mean the odds against the bb are 3 to 1?

lyndonashmore
2006-Jun-16, 08:20 PM
Hi Lyndon, You are right that this is a good paper, thanks for bringing it to our attention. One of the things that I think is most interesting about this paper is that it shows the imprecision we are dealing with in making these measurements with the state of the art (eight years ago) tools available to us. This paper points out (as noted in posts above) that the scale of the over estimate of the observed time dilation is about one sigma, and should happen a third of the time. Based on this one measurement, that's about all you can ask for.

Now, this observation was made eight years ago, and was just published now. I'd really like to know how many similar measurements have been made since, and whether we are now capable of describing a statistically more significant sample size.
sorry, just to add .
the error is not an over estimate, the BB UNDERESTIMATES curve broadenng by just over 500%
Cheers,
Lyndon

Zahl
2006-Jun-16, 08:23 PM
right on what?
This is the rest frame value.
The BB predicted a delay of 1.31 days when the observed delay was 7.91 days.
What s "right on" about that?
Cheers,
lyndon

No, you got that wrong. The BB predicted time dilation was 24.88-18.28=6.6 days and the observed dilation was 24.88-16.97 +/-2.75 = 7.91 +/-2.75 days. The BB prediction was off by 1.31 +/-2.75 days. That is, the prediction is within one standard deviation of the expected figure and so right on the mark. The "no expansion" prediction was 0 days dilation and is off by 7.91/2.75=2.9 standard deviations, ie. excluded at the 99% level.

Besides, this paper does not use the old "light curve broadening" method, but instead measures time dilation directly by observing how the spectral features themselves age.

lyndonashmore
2006-Jun-16, 08:34 PM
No, you got that wrong. The BB predicted time dilation was 24.88-18.28=6.6 days and the obserbed dilation was 24.88-16.97 +/-2.75 = 7.91 +/-2.75 days. The BB prediction was off by 1.31 +/-2.75 days. That is, the prediction is within one standard deviation of the expected figure and so right on the mark. The "no expansion" prediction was 0 days dilation and is off by 7.91/2.75=2.9 standard deviations, ie. excluded at the 99% level.

Besides, this paper does not use the old "light curve broadening" method, but instead measures time dilation directly by observing how the spectral features themselves age.
Pardon?
elapsed time between the first two spectra was OBSERVED as 24.88 days.
The rest frame observation is 16.97 days.
The BB predicted 18.28 days.
Ergo, the predicted time dilation by the Bb is 18.28 - 16.97 days 1.31 days?
The actual broadening was 24.88 - 16.97 = 7.91 days.
Over 500% out.
This paper uses spectral features, as cited in the paper other methods assumed expansion and a 1 + z broadening in the first place and then cited the results as proving it.
I use curve broadening in the same sense as hubble used redshifts and not velocities.
Curve broadening is what we measure, time dilation is an interpretation of the results.
cheers,
Lyndon

lyndonashmore
2006-Jun-16, 08:40 PM
Sorry Zahl,
I see what you mean. yes this method uses spectral features such as when certan elements come throught the photosphere and not curve broadenng as such - but you know what i mean.

Zahl
2006-Jun-16, 09:16 PM
Pardon?
elapsed time between the first two spectra was OBSERVED as 24.88 days.
The rest frame observation is 16.97 days.
The BB predicted 18.28 days.
Ergo, the predicted time dilation by the Bb is 18.28 - 16.97 days 1.31 days?

No. The BB prediction for supernova rest frame aging is

(observed frame aging) x (1/(1+z)) = supernova rest frame aging

Thus, 24.88 x (1/(1+0.361)) = 18.28 days for rest frame aging. The SFA measured supernova rest frame aging is 16.97 +/-2.75 days and the BB prediction is off by 18.28-16.97=1.31 +/-2.75 days and is within 0.5 standard deviations of the expected figure.


The actual broadening was 24.88 - 16.97 = 7.91 days.

Yes.


Over 500% out.

No.

Ari Jokimaki
2006-Jun-17, 04:47 AM
In fact, their results are consistent with expansion and exclude tired light and theories such as Arp's.
How does it exclude Arp's theory specifically? In your opinion, what does Narlikar-Arp variable mass theory say about supernova time dilation?

Zahl
2006-Jun-17, 05:49 AM
How does it exclude Arp's theory specifically? In your opinion, what does Narlikar-Arp variable mass theory say about supernova time dilation?

From the paper:

"This result also shows that the null hypothesis of no cosmic expansion is ruled out at the 97.9% level. Thus, “tired light” and other hypotheses that predict no time dilation are essentially eliminated. To be consistent with the observations, alternatives such as the “variable mass theory” (e.g., Narlikar & Arp 1997) would require a highly unlikely series of coincidences, culminating with SN Ia spectral evolution that mimics the result expected with simple time dilation. Narlikar & Arp (1997) discount the results of Leibundgut et al. (1996) since the variable mass theory changes the decay rate of 56Ni (which dictates the shape of an SN Ia light curve) with redshift. However, the evolution of spectral features depends on the composition of the ejecta and opacities, and therefore on temperature and density. Although the redshift of a given spectral feature will depend on the mass of subatomic particles, the evolution of that line does not depend on the mass of these particles. Since spectral evolution should not change with redshift, we therefore can rule out the variable mass theory with our current data."

Ari Jokimaki
2006-Jun-17, 07:15 AM
From the paper:

"This result also shows that the null hypothesis of no cosmic expansion is ruled out at the 97.9% level. Thus, “tired light” and other hypotheses that predict no time dilation are essentially eliminated. To be consistent with the observations, alternatives such as the “variable mass theory” (e.g., Narlikar & Arp 1997) would require a highly unlikely series of coincidences, culminating with SN Ia spectral evolution that mimics the result expected with simple time dilation. Narlikar & Arp (1997) discount the results of Leibundgut et al. (1996) since the variable mass theory changes the decay rate of 56Ni (which dictates the shape of an SN Ia light curve) with redshift. However, the evolution of spectral features depends on the composition of the ejecta and opacities, and therefore on temperature and density. Although the redshift of a given spectral feature will depend on the mass of subatomic particles, the evolution of that line does not depend on the mass of these particles. Since spectral evolution should not change with redshift, we therefore can rule out the variable mass theory with our current data."
Thanks Zahl!

I couldn't access the paper (Lyndon's link directs me to abstract page), but now I dug out the preprint, here it is (http://arxiv.org/abs/astro-ph/0504481).

papageno
2006-Jun-17, 10:07 AM
You say that the paper does not agree with BB cosmology when if fact it does. Intentionally or not, you did not understand what you read. The paper does not say what you claim it does. That is intellectual dishonesty in my book. Unfortunately this is typical for Lyndon.

In the old BABB thread where he presented his tired-light-like "theory", most of the references he came up with actually contradicted his claims.

He even thinks that out-of-context quotes are good science!

lyndonashmore
2006-Jun-17, 10:10 AM
Yep.
Thanks Zahl, I got that bit wrong!
So where are we then?
We have had all this fuss in previous threads about ‘time dilation’ being a new pillar of the BB and yet here we have a letter by noted astrophysics published in a respected journal as late as June 2005 that states:


despite the predominance of the current redshift paradigm, little direct observational evidence exists in support of cosmological expansion

Some of the “time dilation” results used in the past using curve broadening assumed a 1 + z dilation in the processing of the results – and then told us that the results proved time dilation

moreover, for any individual SN Ia, the intrinsic width is unknown, so without assuming a 1 + z dilation, the intrinsic width and dilation cannot be separated

Thirdly, and we will get this right this time.
The observed age factor is greater than that predicted by the BB. (1.6 as opposed to 1.36)

The difference between the observed age factor and the and the expected value corresponds to a 1.03σ event, which should be observed 30.3% of the time by chance.
Now, whilst this is not enough to reject the original hypothesis of an expanding universe it certainly doesn’t give one confidence in it – certainly not the confidence that is banded about on this board.

To put it in laymans terms, if the expanding universe theory was a horse, the odds are 7 to 3 against!
Enough said?
Cheers,
Lyndon
P.S. Sorry, the link worked last night!

Zahl
2006-Jun-17, 10:45 PM
Thirdly, and we will get this right this time.
The observed age factor is greater than that predicted by the BB. (1.6 as opposed to 1.36)

Now, whilst this is not enough to reject the original hypothesis of an expanding universe it certainly doesn’t give one confidence in it – certainly not the confidence that is banded about on this board.

On the contrary - this paper shows that the expanding universe case is favored over the non-expanding universe case 99:1 (unless somebody can explain why spectral features appear to age faster in high-z supernovae to mimic time dilation in a non-expanding universe). It also shows that the light curve broadening studies were substantially correct despite all the criticism from ATM people. The specific prediction (that I described earlier) is well fit with just outside the 68% confidence interval.

dgruss23
2006-Jun-17, 11:03 PM
From the paper:

"This result also shows that the null hypothesis of no cosmic expansion is ruled out at the 97.9% level. Thus, “tired light” and other hypotheses that predict no time dilation are essentially eliminated. To be consistent with the observations, alternatives such as the “variable mass theory” (e.g., Narlikar & Arp 1997) would require a highly unlikely series of coincidences, culminating with SN Ia spectral evolution that mimics the result expected with simple time dilation. Narlikar & Arp (1997) discount the results of Leibundgut et al. (1996) since the variable mass theory changes the decay rate of 56Ni (which dictates the shape of an SN Ia light curve) with redshift. However, the evolution of spectral features depends on the composition of the ejecta and opacities, and therefore on temperature and density. Although the redshift of a given spectral feature will depend on the mass of subatomic particles, the evolution of that line does not depend on the mass of these particles. Since spectral evolution should not change with redshift, we therefore can rule out the variable mass theory with our current data."

Two other relevant papers to this discussion are Riess et al (1997) (http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=1997AJ....114..722R&db_key=AST&d ata_type=HTML&format=&high=3e5c03c80a23838) and Narlikar&Arp (1997) (http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=1997ApJ...482L.119N&db_key=AST&d ata_type=HTML&format=&high=3e5c03c80a17838).

I have read all three papers. I see no explanation in any of them to justify this statement by Foley et al:


Although the redshift of a given spectral feature will depend on the mass of subatomic particles, the evolution of that line does not depend on the mass of these particles. Since spectral evolution should not change with redshift, we therefore can rule out the variable mass hypothesis with our current data.

Could someone please clarify for me why the part in bold rules out the VMH? Narlikar&Arp claim that the VMH predicts the exact same time dilation as the standard model. Shouldn't the evolution of spectral features also follow the same time dilation in the VMH? Foley et al offer no explanation as to why it should not.

I also find the following statement by Foley et al unsatisfactorily vague:


To be consistent with the observations, alternatives such as the "variable mass theory" (e.g., Narlikar&Arp 1997) would require a highly unlikely serious of coincidences, culminating with SN Ia spectral evolution that mimics the result expected with simple time dilation.

What coincidences? Narlikar&Arp say this:


This mass difference applies to all subatomic particles and therefore all dynamical and atomic timescales will be scaled up by this factor. ....

The interpretation of the time dilation is different in the two cases. In the standard cosmology this is due to the expansion of space,whereas in the static model it is due to the slower timescales associated with the smaller masses of younger particles.

They expect the same results as "simple" time dilation. Why should their model not also expect the same result for spectral evolution?

To be clear - I'm not saying that their model does expect the same spectral evolution. I'm looking for someone to explain to my why Foley et al are correct that it does not. I'll require such explanation to be convinced they are correct because the mainstream does have a track record or thinking Arp's models expects things that it does not (e.g. the expectation that the bridges connecting discordant objects should show a gradual change in redshift from the lower z to the higher z object).

antoniseb
2006-Jun-17, 11:19 PM
Could someone please clarify for me why the part in bold rules out the VMH? Narlikar&Arp claim that the VMH predicts the exact same time dilation as the standard model. Shouldn't the evolution of spectral features also follow the same time dilation in the VMH? Foley et al offer no explanation as to why it should not.

The paper basically said that VMH wasn't ruled out completely, but that only a very narrow set of parameters with VMH would work supplying the correct spectral information and decay rate of the Ni56. If you aren't a supporter of Arp, you could interpret this as strong evidence that VMH is bogus, because almost all places in the VMH parameter space are ruled out. If you are a supporter of Arp and VMH, then you should be able to use the results of this paper to refine the constraints on your model of the universe and significantly make the science more detailed.

dgruss23
2006-Jun-18, 12:53 AM
The paper basically said that VMH wasn't ruled out completely, but that only a very narrow set of parameters with VMH would work supplying the correct spectral information and decay rate of the Ni56.

The paper doesn't specifically say anything about a very narrow set of parameters being consistent with VMH. What it says is: "we therefore can rule out the variable mass theory with our current data."

I'm looking for someone to provide a justification for Foley et al's claim that the VMH cannot account for the spectral evolution. I'm also looking for some clarification as to what they mean by "highly unlikely set of coincidences". Coincidences of what exactly?

As for Ni56, Narlikar&Arp (1997) say:


It is generally accepted that the light curves of Type I supernova result from the formation of 56Ni, which decays to 56Co and then 56Fe. Since these decay times run on the slower clock times of lower mass atoms, they are time dilated by exactly the (1 + z) factor derived either from the more general Narlikar-Arp or the special case Friedmann solution for the general relativistic field equations.

They are saying their model predicts the exact same time dilation as the standard model. Now perhaps they are wrong - but I'm looking for some specifics to back up the Foley et al claim that they are wrong - because none of these papers provide any real details as to how the data supports their conclusions. Both the Narlikar&Arp (1997) and the Foley et al (1995) papers seem to be engaged in a bit of hand-waving if you ask me.


If you aren't a supporter of Arp, you could interpret this as strong evidence that VMH is bogus, because almost all places in the VMH parameter space are ruled out.

Again - I don't see anything in the paper about VMH parameter space. No figures, no numbers - just an unsupported/unexplained claim that the spectral evolution is inconsistent with VMH.


If you are a supporter of Arp and VMH, then you should be able to use the results of this paper to refine the constraints on your model of the universe and significantly make the science more detailed.

A person can be a supporter of Arp without supporting VMH. As for the VMH, I really will be just as happy to learn that it is not a plausible explanation for intrinsic redshifts as to learn that it is. Either answer would advance our understanding ... but so far I haven't seen any specifics that disprove the VMH.

antoniseb
2006-Jun-18, 01:52 AM
Again - I don't see anything in the paper about VMH parameter space. No figures, no numbers - just an unsupported/unexplained claim that the spectral evolution is inconsistent with VMH.

I'm agreeing with you, though perhaps my use of the language was too obtuse. They made the claim, and it *sounds* tantalizingly like they'd actually computed something somewhere, but only gave a one sentence summary. I gave you my understanding of what they meant.

The fact that this is eight year old data suggests to me that there's a couple of groups out there trying to collect a much larger sample, and are not publishing until they're ready. So far I've heard no rumours of some major upset to cosmology over the data, so if my guess is right, we'll just see some paper covering twenty or thirty distant SNs and they'll average out to close to the expected time dilation. Odds are these groups are spending a lot of time trying to understand variations in the SN type 1a's spectra.

lyndonashmore
2006-Jun-18, 05:30 AM
On the contrary - this paper shows that the expanding universe case is favored over the non-expanding universe case 99:1 (unless somebody can explain why spectral features appear to age faster in high-z supernovae to mimic time dilation in a non-expanding universe). It also shows that the light curve broadening studies were substantially correct despite all the criticism from ATM people. The specific prediction (that I described earlier) is well fit with just outside the 68% confidence interval.
Sorry Zahl, but you are looking at comparisons between two theories. The topic of this post is the BB's prediction of the aging factor only - and it being too far out to have much confidence in it.
We keep being told that supernova time dilation is a 'proof' of the BB theory when this paper shows that it has a long way to go. Looking at the BB and its prediction ONLY it is 7 to 3 against.
Not very convincing.
cheers,
Lyndon

Tim Thompson
2006-Jun-19, 04:31 PM
Pardon?
elapsed time between the first two spectra was OBSERVED as 24.88 days.
The rest frame observation is 16.97 days.
The BB predicted 18.28 days.
Ergo, the predicted time dilation by the Bb is 18.28 - 16.97 days 1.31 days?
Pardon, indeed. What in the world are you doing? Why are you subtracting 18.28-16.97? Is that number supposed to mean something? Do you have any idea at all what you are talking about?

The OBSERVED elapsed time between the 1st & 2nd spectra is 24.88 days.
The BB PREDICTED rest frame elapsed time is 24.88*(1/1.361) = 18.28 days, based on the redshift alone.
The COMPUTED spectral aging is 16.97 +/- 2.75 days

So, the 1-sigma range on the computed rest frame aging is 14.22 -> 19.72 days. As you can see, the redshift predicted value (i.e., the BB predicted value) of 18.28 is comfortably inside that 1-sigma range, which makes the difference between 16.97 +/- 2.75 and 18.28 insignificant.

The OBSERVED elapsed time between the 2nd & 3rd spectra is 6.07 days.
The BB PREDICTED rest frame elapsed time is 6.07*(1/1.361) = 4.46 days
The COMPUTED spectral aging is 1.04 +/- 3.03 days.

So, the 1-sigma range on the computed rest frame aging is -1.99 -> 4.07 days. Now we see that 4.46 lies outside this 1-sigma range, at 1.13-sigma. This is not a significant discrepancy, as anything inside 2-sigma is not worth fussing over (indeed, real scientists rarely speak of any result being "positive" if it is less than 3-sigma). We might also note that this case is consistent with the absence of time dilation.

Now, the OBSERVED elapsed time between the 1st & 3rd spectra is 30.95 days.
The BB PREDICTED rest frame elapsed time is 30.95*(1/1.361) = 22.74 days.
The COMPUTED spectral aging is 18.01 +/- 3.14 days.

So, in this case, the 1-sigma range on the computed rest frame aging is 14.87 -> 21.15 days. Now we see, once again, that 22.74 is outside the 1-sigma range, at 1.5-sigma. Again, a result to be noted, but not a significant discrepancy. But we might note also that the spectrally computed aging in all 3 cases is systematically low compared to the theoretical prediction.

Now, we will use the text of the preprint (http://arxiv.org/abs/astro-ph/0504481) already provided by Ari Jokimaki, since the published paper is generally inacessible on the web without a subscription. You can get at the PDF file through a link on that page. As we see on the bottom of page 6, the 3 examples we give independently above, are combined to give an age factor of 1.602 +/- 0.234. The age factor is just 1+z, and should theoretically be 1.361. The 1-sigma range on the given age factor is 1.368 -> 1.836. We do find that the BB predicted value of 1.361 is just outside the bottom of the 1-sigma range, at 1.03-sigma, not far enough from 1-sigma to be significant. We should like to see better, but this is not a bad result. We would need to see a discrepancy beyond 2 or 3-sigma before calling the result into question. We can say that the discrepancy here, between the BB predicted & spectrally computed time dilation is not significant. This should be obvious, considering observational uncertainties, and the small size of the data set (only 3 intervals is not a major test of anything, but it can be indicative).

But, what about the "null hypothesis", the case of no time dilation at all? In that case the aging factor should be 1.000, exactly. The spectrally computed value of 1.602 +/- 0.234 differs from 1.000 by 2.57-sigma, which is significant, although not a 3-sigma discrepancy. So, if you are hard-nosed about 1.03-sigma being something you don't like (and you will be understandably lonely in this position), you will have to be even more hard-nosed about 2.57-sigma, and forced to concede that either time dilation is happening, or something that looks a lot like it, so much so as to be virtually indistinguishable from it.

But we are only looking at one paper. Note that the authors of the paper we are looking at are using a spectral aging technique developed here: Time Dilation from Spectral Feature Age Measurements of Type IA Supernovae (http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=1997AJ....114..722R&db_key=AST&d ata_type=HTML&format=&high=4366fa465123421); A.G. Riess, et al., Astronomical Journal 114: 722-729, August 1997 (you can get the PDF from a link on that page). They examined the spectrum of SN 1996bj, at a higher redshift of 0.574. Their observed elapsed time of 10.05 days should correspond to a BB predicted rest frame elapsed time of 10.05*(1/1.574) = 6.38 days. Their spectrally determined aging was 3.35 +/- 3.2 days. Once again, the 1-sigma range of 0.15 -> 6.55 days comfortably includes the BB predicted 6.38 days, once again making the discrepancy insignificant (but the spectrally computed value is once again low compared to the theoretical prediction). And this result rules out the null hypothesis at the 96.4% confidence level.

You can explore the 47 citations to the Riess, et al., paper, and seek out other examples, I didn't look. Suffice to say that in the two examples we have here, BB predictions & spectral determinations are certainly consistent, and the claim that "BB fails time dilation" (spelled wrong on the thread title; a "dilaton (http://en.wikipedia.org/wiki/Dilaton)" is a far more exotic concept). We can notice that there seems a tendency for the spectral aging determination to be low, compared to the BB prediction. That implies to me a systematic bias in the spectral determination, which can probably be eliminated, or at least mitigatged, with improved methodology.

The reeal deal is that "BB survives time dilation".

Tim Thompson
2006-Jun-19, 04:39 PM
Could someone please clarify for me why the part in bold rules out the VMH?
I think it is because the curves fit a single time dilation. If the mass were variable, then the time dilation should be variable as well, it should change as the mass changes. So the fact that one aging factor fits implies that the time dilation does not change with time. This is a guess more than a firm claim, and I have not read the papers by Narlikar & Arp on time dilation & variable mass, so I don't know where they are coming from.

ExpErdMann
2006-Jun-19, 05:19 PM
From the paper:

"This result also shows that the null hypothesis of no cosmic expansion is ruled out at the 97.9% level. Thus, “tired light” and other hypotheses that predict no time dilation are essentially eliminated."

It is important to point out that the result merely places a constraint on tired light models, albeit an important one. They must incorporate time dilation. This is not so far out as it may sound. Including tired light could actually boost tired light models as far as the Tolman surface brightness test is concerned, since we would need to include an extra (1 + z) factor for tired light models. That would give us two (1+z) factors for tired light, versus four for the Big Bang. I've read some measured values which come in around 2.

antoniseb
2006-Jun-19, 05:36 PM
It is important to point out that the result merely places a constraint on tired light models, albeit an important one. They must incorporate time dilation. This is not so far out as it may sound.

Imagine that there is a civilization that decides to communicate with us. They send out a series of pulses, one pulse per hour every hour for a hundred years. The send it with a laser that uses a wavelength of 500 nm. They are at a place so far away that we see then in a z=1 galaxy. So the wavelength is now 1000 nm, and the pulses are 2 hours apart for two hundred years.

How would you get tired light to simulate this? Wouldn't that mean that the mechanism was somehow storing information for us?

peteshimmon
2006-Jun-19, 06:09 PM
The horse was named Foinavon as I remember
and I watched the race live back in 1967.
Most of the runners all piled up on each
other after a fence and this horse strolled
up from the rear and went around them. So
nice to see an also ran beat the big names
sometime:) Stroll on!

ExpErdMann
2006-Jun-19, 06:11 PM
I don't think any information storage would be necessary. It's just that there are several ways of getting a time dilation effect, through special relativity, through GR or even through the classical Doppler shift (which was what Wilson, who is metioned in the article, based his original supernova prediction on). Considering light in motion through space, it is possible to imagine a gravitational interaction with other masses (as Zwicky did), which could somehow give a GR time dilation term; or there could be some different interaction with matter, which might yield an SR time dilation effect (due to the velocities of bodies as measured, say, in the CMBR frame). I'm just saying that it would be premature to rule out all tired light models simply because there is not one yet that has been shown to generate time dilation. Even Zwicky's model can't be totally disqualified on these grounds (I checked the citations on Zwicky's paper recently and I don't think it was ever disqualified on other grounds either).

antoniseb
2006-Jun-19, 06:44 PM
I'm just saying that it would be premature to rule out all tired light models simply because there is not one yet that has been shown to generate time dilation.

There isn't a tired light mechanism that has been shown to cause tired light, let alone time dilation.

lyndonashmore
2006-Jun-19, 07:01 PM
The horse was named Foinavon as I remember
and I watched the race live back in 1967.
Most of the runners all piled up on each
other after a fence and this horse strolled
up from the rear and went around them. So
nice to see an also ran beat the big names
sometime:) Stroll on!
Spot on!
I had two and six each way on it.
I won that outsider and I will win this one!
That's a promise.
Cheers,
lyndon

antoniseb
2006-Jun-19, 07:06 PM
I won that outsider and I will win this one! That's a promise.
That's a pretty far stretched analogy. I doubt you'll be able to keep this promise, but perhaps you can reword it less poetically (less ambiguously) and tell us exactly what it is you're promising so we can try and hold you to it.

lyndonashmore
2006-Jun-19, 07:08 PM
Tim Thompson wrote.

We might also note that this case is consistent with the absence of time dilation.
so why waste all that ink if the result is not conclusive?
Folk here use supernova 'time dlation' as a proof of the BB when it proves no such proof at all!
When you say

Now we see that 4.46 lies outside this 1-sigma range, at 1.13-sigma. This is not a significant discrepancy, as anything inside 2-sigma is not worth fussing over (indeed, real scientists rarely speak of any result being "positive" if it is less than 3-sigma
why do we say that the bb is 'proven' when the tresult is just over 1 sigma out?
The jury is still out.
If the BB was a horse the odds from this paper are 7 to 3 against it. If it were a horse it would be put down.
Cheers,
lyndon.
P.S. sory abowt the speling - I amm a sientist yu sea.

ExpErdMann
2006-Jun-19, 07:18 PM
There isn't a tired light mechanism that has been shown to cause tired light, let alone time dilation.

I'm not sure you could say that about Zwicky's model.

The time dilation findings can serve as a useful clue to those who want to construct tired light models. All I'm saying is that each such model should be evaluated independently so far as time dilation is concerned. You can't just group all the tired light models together

Jim
2006-Jun-19, 08:00 PM
why do we say that the bb is 'proven' when the tresult is just over 1 sigma out?
The jury is still out.
If the BB was a horse the odds from this paper are 7 to 3 against it. If it were a horse it would be put down.


But, what about the "null hypothesis", the case of no time dilation at all? In that case the aging factor should be 1.000, exactly. The spectrally computed value of 1.602 +/- 0.234 differs from 1.000 by 2.57-sigma, which is significant, although not a 3-sigma discrepancy. So, if you are hard-nosed about 1.03-sigma being something you don't like (and you will be understandably lonely in this position), you will have to be even more hard-nosed about 2.57-sigma, and forced to concede that either time dilation is happening, or something that looks a lot like it, so much so as to be virtually indistinguishable from it.

So, lyndon, would you then agree that the null hypothesis is worse off than BB (2.57s being a poorer result than 1.03s, or even 1.5s)? Should that horse be put down as well?

BTW, did you ever get around to that dimensional analysis of your "theory" which I suggested some time back?

And, have you run the range of the results for h your "theory" gives over the range of values for n? I realize that you can get close to the (currently) accepted value of h with n=1, but you yourself state that n can be anywhere from 0.5 to 10. Seems to me you either have to say your calculation covers a wide range of h values, or you have to show why n=1 is correct.

lyndonashmore
2006-Jun-19, 08:07 PM
Hi Jim,
Was it you that wanted H in bananas per square yard?
I skipt that because I needed to wash my hair that night.
H = 2nhr/m works perfectly well dimensionally. I never really understood why this board insisted on people answering questions that where patently obvious to all and sundry.
The range of results has nothing to do with my theory, it is due to the imprecission by whch n can be measured - and published!
Cheers,
So Jim . gettng back to the original post, what is the range of results of the predicted value of H by the BB/
Cheers,
lyndon.
P.S. I always remember!

Jim
2006-Jun-19, 09:15 PM
Hi Jim,
Was it you that wanted H in bananas per square yard?
I skipt that because I needed to wash my hair that night.
H = 2nhr/m works perfectly well dimensionally. I never really understood why this board insisted on people answering questions that where patently obvious to all and sundry.
The range of results has nothing to do with my theory, it is due to the imprecission by whch n can be measured - and published!
Cheers,
So Jim . gettng back to the original post, what is the range of results of the predicted value of H by the BB/
Cheers,
lyndon.
P.S. I always remember!

You haven't answered my questions, and you are being very patronizing. I do not appreciate either one.

I am not required to show what value(s) for h could be predicted or derived from any version of the BB. You have specifically stated that your "theory" can make such a predicition. It is incumbent on you to defend that statement. Please do so.

1. Please do a dimensional analysis of your equation. You keep stating it in SI; please state it fully in SI, and then show how it translates to any other widely used system (Imperial or US would be acceptable) units.

2. Also, please explain why you chose to use n=1 instead of n=0.5 or n=10, which you stated are also possible values for n.

BTW, you don't remember too well. You "forgot" to respond to my request in your earlier thread, and you definitely forgot that I gave the example of furlongs per fortnight, which I got from my professor who was making a point about internal dimensional consistency. I realize you were probably going for the cheap laugh, but as I stated, I find that very patronizing and do not appreciate it.

My requests are serious. Please treat them as such.

Tim Thompson
2006-Jun-20, 05:06 AM
Tim Thompson wrote.

We might also note that this case is consistent with the absence of time dilation.
so why waste all that ink if the result is not conclusive?

Not conclusive?

The spectrally computed value of 1.602 +/- 0.234 differs from 1.000 by 2.57-sigma, which is significant, although not a 3-sigma discrepancy

You have a strange notion of what constitutes "not conclusive". But of course, you were only talking about 1 out of 3 cases, and choose to ignore the other two cases. That is not a very smart thing to do.


Folk here use supernova 'time dlation' as a proof of the BB
No they don't

... when it proves no such proof at all!
Nothing ever does. Theories are never proven, why don't you know that? Theories are never "proven", but they are "consistent". In this case, as a matter of fact, time dilation and BB cosmology are very much consistent with each other.


why do we say that the bb is 'proven' when the tresult is just over 1 sigma out?
We don't say it's proven. We do say that "just over 1-sigma out" is insignificant, and we mean it. Insignificant. Unimportant. Meaningless. There is no significant difference between any two quantities which differ from each other by 1-sigma, or slightly more than 1-sigma.


If the BB was a horse the odds from this paper are 7 to 3 against it.
Absolutely false. In fact, the odds from that paper are better than 95% in favor (2-sigma is 95%, and 3-sigma is 99.7%, I am not sure of the significance level for 2.57-sigma, but it is clearly better than 95%). Your "7 to 3 against" is just some fantasy odds calculation you made up, and has no bearing on reality.

lyndonashmore
2006-Jun-20, 09:48 AM
1. Please do a dimensional analysis of your equation. You keep stating it in SI; please state it fully in SI, and then show how it translates to any other widely used system (Imperial or US would be acceptable) units.
In a word, NO! This is not a question but a request - one that I am not prepared to waste time on. If you feel that there is some mistake, dimensionally, in my formula then post here where you feel that mistake is and I will answer that.


2. Also, please explain why you chose to use n=1 instead of n=0.5 or n=10, which you stated are also possible values for n.

I use n = 0.5 to explain this result

extend beyond 30,000 km/sec and provide a dramatic confirmation of the Hubble law,
v = dD/dt = H*D
The fitted line in this graph has a slope of 64 km/sec/Mpc (http://www.astro.ucla.edu/~wright/cosmo_01.htm)
If we let hr/m in each cubic metre of space be the constant A (Ashmore’s constant) then the HST result of 72km/s per Mpc is about 1.1A this just means n is about 0.55?
One of the many benefits of my theory is that once H is known one can determine n fairly accurately.
Cheers,
Lyndon

lyndonashmore
2006-Jun-20, 09:55 AM
Absolutely false. In fact, the odds from that paper are better than 95% in favor (2-sigma is 95%, and 3-sigma is 99.7%, I am not sure of the significance level for 2.57-sigma, but it is clearly better than 95%). Your "7 to 3 against" is just some fantasy odds calculation you made up, and has no bearing on reality.
Well, actually, Tim, I got it from the paper.
It cites a 30% 'by chance'. 7 ways of being inside this value of '1 and a bit sigma', 3 ways of being where our result is ergo odds are 7 to 3 against.
Cheers,
Lyndon

Eta C
2006-Jun-20, 12:33 PM
Which just goes to show again that you don't understand what you read. Tim, Zahl, and I have tried to point it out to you several times, and you still don't understand. Now I will admit that understanding confidence intervals can be counterintuitive, but it's something that scientists learn to deal with. Frankly, you just failed STAT 101 lyndon.

Jim
2006-Jun-20, 01:04 PM
1. Please do a dimensional analysis of your equation. You keep stating it in SI; please state it fully in SI, and then show how it translates to any other widely used system (Imperial or US would be acceptable) units.

In a word, NO! This is not a question but a request - one that I am not prepared to waste time on. If you feel that there is some mistake, dimensionally, in my formula then post here where you feel that mistake is and I will answer that.

Okay, let's put it in the form of a question: What are the results of a dimensional analysis of your equation? Can you translate the equation between two widely accepted systems of units?

Your refusal to do so makes me think that either you don't know how to do it, or you are afraid your equation won't hold up.



2. Also, please explain why you chose to use n=1 instead of n=0.5 or n=10, which you stated are also possible values for n.

I use n = 0.5 to explain this result

extend beyond 30,000 km/sec and provide a dramatic confirmation of the Hubble law,
v = dD/dt = H*D
The fitted line in this graph has a slope of 64 km/sec/Mpc (http://www.astro.ucla.edu/~wright/cosmo_01.htm)
If we let hr/m in each cubic metre of space be the constant A (Ashmore’s constant) then the HST result of 72km/s per Mpc is about 1.1A this just means n is about 0.55?
One of the many benefits of my theory is that once H is known one can determine n fairly accurately.

(Emphasis added.)

Oops, my bad! You do use 0.5, not 1. But, the question remains, why that value? (Or 0.55, which you say fits the curve better.)

Also, you previously said:
However, on a point of information my theory predicts H = 2nhr/m (h planck constant, m and r mass and classical radius of the electron. 'n' is the average electron density of IG space and published values of this lie between 0.1 and 10 electrons per cubic metre of space.
H = 72km/s per Mpc is when n = 0.5! Theory is consistent with observation, no coincidences, referenced material, accepted physics. (Emphasis added.)

So, if n=10, then your "theory" must be wrong.

Also, didn't you claim that you could predict h? But then you say you chose n=0.5 because it fits observation... you selected 0.5 (0.55?) because it fits the curve. Isn't that a bit of a cop out? (I know I told you about my college roommate and his Burnop Factor... run the calculation, multiply by zero, and add the correct value. All you've done is generate a corollary.)

It seems to me you're not predicting h; you're using observed values of h to predict n.

lyndonashmore
2006-Jun-20, 03:43 PM
Which just goes to show again that you don't understand what you read. Tim, Zahl, and I have tried to point it out to you several times, and you still don't understand. Now I will admit that understanding confidence intervals can be counterintuitive, but it's something that scientists learn to deal with. Frankly, you just failed STAT 101 lyndon.
I don't see any physics or stats in this post.
please stick to science and refrain from this sort of thing.
Cheers,
Lyndon

lyndonashmore
2006-Jun-20, 03:51 PM
Okay, let's put it in the form of a question: What are the results of a dimensional analysis of your equation? Can you translate the equation between two widely accepted systems of units?
Answer: Yes I can
Question: will I do it?
No for the same reason as before. It is patently obvious the the equation holds up dimensionally so I am not gong to waste my time. If you feel there is an error then state where the error is and I will point out where you are wrong.


Your refusal to do so makes me think that either you don't know how to do it, or you are afraid your equation won't hold up.
Neither, I just have more sense




So, if n=10, then your "theory" must be wrong.

Also, didn't you claim that you could predict h? But then you say you chose n=0.5 because it fits observation... you selected 0.5 (0.55?) because it fits the curve. Isn't that a bit of a cop out? (I know I told you about my college roommate and his Burnop Factor... run the calculation, multiply by zero, and add the correct value. All you've done is generate a corollary.)

It seems to me you're not predicting h; you're using observed values of h to predict n.
Inserting published values of n show that my predicted value of H is consisted with the observed value.
Predicted.
Cheers,
Lyndon,
Who is now off to watch the football.

Tim Thompson
2006-Jun-20, 06:19 PM
Well, actually, Tim, I got it from the paper.
It cites a 30% 'by chance'. 7 ways of being inside this value of '1 and a bit sigma', 3 ways of being where our result is ergo odds are 7 to 3 against.

The observed age factor is larger than the expected age factor of 1+z = 1.361. The difference between the observed age factor and the expected value corresponds to a 1.03-sigma event, which should be observed 30.3% of the time by chance
Now, according to you, this means that BB theory is wrong, and the odds against it are 7 to 3. But that is wrong, because you ignore what is meant by "observed" age factor. The "observed" age factor is not "observed", it is "derived", from a lookup table of template spectra, and the associated algorithms, all developed by Riess, et al., 1997 (http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=1997AJ....114..722R&db_key=AST&d ata_type=HTML&format=&high=4366fa465123421). The 1.03-sigma discrepency, the "odds of 7 to 3 against" could just as easily be against the Riess et al., technique for deriving age factors. In fact, I think it is obvious that this is far more likely to be the case, which is why I bothered to point out that ...

We can notice that there seems a tendency for the spectral aging determination to be low, compared to the BB prediction. That implies to me a systematic bias in the spectral determination, which can probably be eliminated, or at least mitigatged, with improved methodology.
Systematic errors are a clear signal that something non-random is wrong. Either the theoretical (BB) age factor of 1+z is too large, or the Riess, et al., age factor is too small. The obvious place to look first is to the latter, which is a limited, empirical lookup model anyway.

So you misinterpret the meaning of what you read, and wrongly assume it means odds of 7 to 3 against BB, which it certainly does not.

But you are also ignoring the bigger point, and that is the discrepency between the "observed" age factors, and the "null hypothesis". That difference is 2.57-sigma in favor of the "observed" age factors. The "null hypothesis" is the absence of time dilation. That means that the paper shows the odds are at least 95 to 5 that the observed spectra are time dilated. And since BB theory holds that the spectra must be time dilated, it logically follows that BB theory is favored by the odds of at least 95 to 5. It is only a matter of figuring out how to calculate the theoretically expected age factor. Perhaps 1+z is not correct, but that is only a detail as to which BB model we choose. Perhaps that explains the 1.03-sigma discrepency. We could argue about that, but nobody can justifiably claim to know, one way or the other.

Or maybe there is no discrepancy at all. As Foley, et al., say, you expect to see that result 30.3% of the time, so if you look once, you have about 1/3 chance of seeing it. If you look 3 times, you have about a 1/1 chance of seeing it. I show two "looks", Foley, et al., and Riess, et al. Foley, et al., is the 1.03-sigma event, and Riess, et al., is inside the 1-sigma range. So in two "looks", we see one which is a 1/3 event. That's hardly a strange occurence.


Reference: Foley, et al., 2005 (http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=2005ApJ...626L..11F&db_key=AST&d ata_type=HTML&format=)

Eta C
2006-Jun-20, 06:32 PM
I don't see any physics or stats in this post.
please stick to science and refrain from this sort of thing.
Cheers,
Lyndon

I'll ignore the provocation and point the readers to your posts and let them decide whose are content free. As to the topic at hand, I'd simply point out again that you don't know what you're talking about. As the paper from the OP points out, the results exclude the non-expanding non-time dilated hypothesis at a 95% or greater confidence level. You can try to keep spinning it, but since the OP you've been wrong. I stand by my first response back in post 2 of the thread. You've given me ample reason never to take you seriously.

lyndonashmore
2006-Jun-20, 09:29 PM
I'll ignore the provocation and point the readers to your posts and let them decide whose are content free. As to the topic at hand, I'd simply point out again that you don't know what you're talking about. As the paper from the OP points out, the results exclude the non-expanding non-time dilated hypothesis at a 95% or greater confidence level. You can try to keep spinning it, but since the OP you've been wrong. I stand by my first response back in post 2 of the thread. You've given me ample reason never to take you seriously.
I am not provoking anyone here eta c.
This is not science that you are posting but a personal attack. (ad homiwhatsit?)
If you have no science worthy of posting then don't post.
It is as simple as that.
So, when it comes to horse racing odds, as I posted, what are the odds?
7 to 1 against is what i posted.
Cheers,
Lyndon

lyndonashmore
2006-Jun-20, 09:38 PM
Or maybe there is no discrepancy at all. As Foley, et al., say, you expect to see that result 30.3% of the time, so if you look once, you have about 1/3 chance of seeing it. If you look 3 times, you have about a 1/1 chance of seeing it. I show two "looks", Foley, et al., and Riess, et al. Foley, et al., is the 1.03-sigma event, and Riess, et al., is inside the 1-sigma range. So in two "looks", we see one which is a 1/3 event. That's hardly a strange occurence.


Reference: Foley, et al., 2005 (http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=2005ApJ...626L..11F&db_key=AST&d ata_type=HTML&format=)

Exactly.
One expects to see that result 30% of the time.
So you expect not to see it 70% of the time.
That means (and I will type this slowly to help you and Eta c understand it) that there are 7 ways of not seeing it and 3 ways of seeing it out of every ten.
ergo, the odds against it are 7 to 3 against,
Now what is difficult about that?
However,

you have about 1/3 chance of seeing it. If you look 3 times, you have about a 1/1 chance of seeing it[ so if you look often enough you will have a 100% chance of seeing whatever you want?
Your expectation may be that but surely not your chance.
As Eta C says, I think you have just failed sums 1.
cheers,
lyndon

Kesh
2006-Jun-20, 11:00 PM
I don't see any physics or stats in this post.
please stick to science and refrain from this sort of thing.
Cheers,
Lyndon
Lyndon, I'm not a professional physicist or a mathematician. I'm just an amateur with some college-level education in those. And I understand what EtaC is getting at here.

If you don't understand how margin-of-error measurements are statistics, and integral to physics, I think you've got a lot of work to do before you can promote or decry any theory.

Kesh
2006-Jun-20, 11:04 PM
Exactly.
One expects to see that result 30% of the time.
So you expect not to see it 70% of the time.
That means (and I will type this slowly to help you and Eta c understand it) that there are 7 ways of not seeing it and 3 ways of seeing it out of every ten.
ergo, the odds against it are 7 to 3 against,
Now what is difficult about that?
However,
so if you look often enough you will have a 100% chance of seeing whatever you want?
Your expectation may be that but surely not your chance.
As Eta C says, I think you have just failed sums 1.
cheers,
lyndon

... stop. Right there. :hand:

I'm going to try to explain this so it's easier to comprehend. You have a 30% chance of an error. 30% of 100 is roughly 1/3. That means that out of every 3 observations, you are likely going to see one that's outside the accepted margin of error.

Which means if you make three observations, your chances of seeing one that's outside the margin are near 100%, statistically. It's not guaranteed, but it's very likely. Just like if you make ten observations, there's a good chance 3 of them are outside the margin of error. Thus, your odds of seeing a result outside the margin of error, from a set of 3 observations is basically 1/1.

Does that make more sense?

Eta C
2006-Jun-21, 01:27 AM
Exactly.
One expects to see that result 30% of the time.
So you expect not to see it 70% of the time.
That means (and I will type this slowly to help you and Eta c understand it) that there are 7 ways of not seeing it and 3 ways of seeing it out of every ten.
ergo, the odds against it are 7 to 3 against,
Now what is difficult about that?
However,
so if you look often enough you will have a 100% chance of seeing whatever you want?
Your expectation may be that but surely not your chance.
As Eta C says, I think you have just failed sums 1.
cheers,
lyndon

Sorry Lyndon your naive understanding of statistics (and I'm not going to sugar-coat it. It is naive.) has led you astray. Tim and I are correct and you're the one who needs to get it straight. I'll try one more time and if you don't get it now, you're helpless.

First off, you've got the idea of the confidence interval backwards. When I define a 1-sigma confidence interval around an expected value (x +/- y) I'm saying that 70% of the time I would expect to find a value within +/- y of that value (x). The comparison from the OP paper falls just outside the 1-sigma range, and is not unusual in that respect. As an experimental physicist I'd say the measurements are consistent.

Now for the null case in the paper (no time dilation) the value is about 3 sigma out from the expected value. If you go to a table of the cumulative gaussian distribution you'd find that you expect a statistical fluctuation of that size less than .3% of the time. That's highly unlikely and I'd say (as the authors of the paper do) that their results exclude the no-expansion hypothesis.

So, a comparison of the probability of a statistical fluctuation causing the 1-sigma offset of the expanding, time-dilated hypothesis (30%) to that of the 3-sigma non-expanding hypothesis (0.3%) is fairly significant.

So, one more time. Get a statistics book! As Kesh says, it's an integral part of understanding any experimental or observational result. I wouldn't give two hoots for an experiment that didn't quantify its uncertainties. If you were my student, you would have gotten an F from me for your performance here, or at least Pauli's evaluation.

Later.

Tim Thompson
2006-Jun-21, 01:35 AM
That means (and I will type this slowly to help you and Eta c understand it) that there are 7 ways of not seeing it and 3 ways of seeing it out of every ten.
ergo, the odds against it are 7 to 3 against, ...
And what, exactly, does that mean? What does it imply, and why?

lyndonashmore
2006-Jun-21, 06:15 AM
Tim Thompson wrote:


Or maybe there is no discrepancy at all. As Foley, et al., say, you expect to see that result 30.3% of the time, so if you look once, you have about 1/3 chance of seeing it. If you look 3 times, you have about a 1/1 chance of seeing it.
Kesh Wrote:

Thus, your odds of seeing a result outside the margin of error, from a set of 3 observations is basically 1/1.
Eta C wrote:

Sorry Lyndon your naive understanding of statistics (and I'm not going to sugar-coat it. It is naive.) has led you astray. Tim and I are correct and you're the one who needs to get it straight. I'll try one more time and if you don't get it now, you're helpless.

Mathematics lesson For Tim, Kesh, Eta C

Sums 1 – probabilities.

Teacher, Lyndon Ashmore.

Question, if there is a one third probability of seeing an event what is the probability of seeing it if you look three times.
(Tim, Kesh, Eta C all put their hands up)
“Yes, you boy” says Lyndon.
1/1 says the Boy.

“Rubbish” says Lyndon. That would mean that if you looked four times the probability would be 4/3 – impossible!

“Lets look at it this way:
Probability of seeing it first time = 1/3
Probability of seeing it second time = 2/3 x 1/3
That is P(not seeing it=2/3)xP(probability of seeing it=1/3)
Probability of seeing third and last time = 2/3 x 2/3 x 1/3
To get the probability of seeing it we add up these three to give: 19/27.
NOT APPROXIMATELY ONE AT ALL.
“Of course” says Lyndon “better students will realise that the probability of ‘seeing the event’ is one minus the probability of ‘not seeing the event’
1 – 2/3 x2/3 x 2/3 = 19/27 same as before – but good scientists always repeat and check.
Lesson over for today.

‘F’s’ all round

Need any more help with your stats boys?
Cheers,
Lyndon

tusenfem
2006-Jun-21, 08:55 AM
Probability calculations like I take 1 ball out of a pot with 10 red and 10 blue balls is one thing, and you have nicely described it in your message above.

This however is not the same as confidence intervals, which work differently. If you have a averaged value bar(x) with standard deviation std(x) for some process you will find that about 30% of the measurements will lie within 1 std(x), 95% of the measurements will lie within 2 std(x) and 99% of the measurements will lie within 3 std(x). It is a bell-shaped curve and is not a multiplicable chance distribution like I wrote above, and which you seem to think describes the data.

Eta C
2006-Jun-21, 12:56 PM
Tim Thompson wrote:


Kesh Wrote:

Eta C wrote:


Mathematics lesson For Tim, Kesh, Eta C

Sums 1 – probabilities.

Teacher, Lyndon Ashmore.

Question, if there is a one third probability of seeing an event what is the probability of seeing it if you look three times.
(Tim, Kesh, Eta C all put their hands up)
“Yes, you boy” says Lyndon.
1/1 says the Boy.

“Rubbish” says Lyndon. That would mean that if you looked four times the probability would be 4/3 – impossible!

“Lets look at it this way:
Probability of seeing it first time = 1/3
Probability of seeing it second time = 2/3 x 1/3
That is P(not seeing it=2/3)xP(probability of seeing it=1/3)
Probability of seeing third and last time = 2/3 x 2/3 x 1/3
To get the probability of seeing it we add up these three to give: 19/27.
NOT APPROXIMATELY ONE AT ALL.
“Of course” says Lyndon “better students will realise that the probability of ‘seeing the event’ is one minus the probability of ‘not seeing the event’
1 – 2/3 x2/3 x 2/3 = 19/27 same as before – but good scientists always repeat and check.
Lesson over for today.

‘F’s’ all round

Need any more help with your stats boys?
Cheers,
Lyndon

Ha. Ha. Ha. It is to laugh. Sorry Lyndon. Wrong application of basic combinatorics. Frankly, from what you've posted here, I doubt you'd recognize a normal distribution, or know what it means even though Tim, Tusenfem, and I have patiently (or not so patiently in my case) tried to explain it to you. If you were to try and pass that last post off in an actual scientific setting (as opposed to an internet BB) I doubt the responses you'd get would be as "polite" as those posted here.

I've said all I have to say on this topic (more than once) and have nothing else to add. You can add more scorn if you want. But scorn like that is not troubling to me.

I'm done on this topic.

Later Lyndon. Take the last word if you want.

Kesh
2006-Jun-21, 03:43 PM
Eh, I'm with EtaC. Lyndon, you seem unwilling to actually try and figure out what we're patiently explaining to you. And given events in my personal life, I've no desire to bang my head against your wall, so to speak.

lyndonashmore
2006-Jun-21, 08:44 PM
Wrong application of basic combinatorics. Frankly, from what you've posted here, I doubt you'd recognize a normal distribution
Nice parting shot. But let me call your bluff.
The point in question was if you look three times at an event that has a probability of being seen equal to one third, then, as Tim posted, the chances are almost 1/1.
I showed that this is wrong.
There is no normal distribution here,
I was wrong earlier in this post.
I admitted to it.
Why won't Tim. Kesh and Eta C do the same.
We all make mistakes.
Show me a scientist who hasn't made a mistake and I will show you one that is either a liar or devoid of any orignal thought.
Tim has yet to post.
A simple "yeh!, you are correct on this point and this point alone" would suffice,
Can we ever trust a Big banger again?
Cheers,
a sad Lyndon

papageno
2006-Jun-21, 09:06 PM
Lyndon, have you addressed this:


But you are also ignoring the bigger point, and that is the discrepency between the "observed" age factors, and the "null hypothesis". That difference is 2.57-sigma in favor of the "observed" age factors. The "null hypothesis" is the absence of time dilation. That means that the paper shows the odds are at least 95 to 5 that the observed spectra are time dilated. And since BB theory holds that the spectra must be time dilated, it logically follows that BB theory is favored by the odds of at least 95 to 5. It is only a matter of figuring out how to calculate the theoretically expected age factor. Perhaps 1+z is not correct, but that is only a detail as to which BB model we choose. Perhaps that explains the 1.03-sigma discrepency. We could argue about that, but nobody can justifiably claim to know, one way or the other.


Let me use JayUtah's words (http://www.bautforum.com/showpost.php?p=749135&postcount=464):


The only meaningful challenge to one line of induction is another line of induction whose inductive leap is shorter. The question is thus not that X isn't proved sufficiently to remove the inductive leap altogether and thus reject Y categorically. It isn't that X's inductive leap is so long that you're just better off believing Y on general principles. The question -- the only proper question, that is -- is whether the inductive leap associated with Y is greater or lesser than X's leap.

So, which one requires the shorter inductive leap?
Time dilation (BBT) with ~1 sigma or No Time Dilation with ~2.5 sigma?

Tobin Dax
2006-Jun-22, 05:16 AM
Nice parting shot. But let me call your bluff.
The point in question was if you look three times at an event that has a probability of being seen equal to one third, then, as Tim posted, the chances are almost 1/1.

Not really. All you're doing there is trying to construct an empty victory over a "banger." Get over it. The point was that, for the standard Gaussian error curves, you will observe an error greater than 1 sigma 30% of the time. 70% of the time, then, you will see an error less than 1 sigma. So, with this error of 1.something sigma, this condition will be seen about 1/3 of the time. (Heck, better yet, go look at papageno's post. He makes that point that this leans strongly toward the BB in an easier way.) The 2.5 sigma error of tired light will only be seen less than 5% of the time.

If you still have trouble understanding this, look here (http://en.wikipedia.org/wiki/Theory_of_errors) and here (http://en.wikipedia.org/wiki/Normal_distribution).

Tim Thompson
2006-Jun-22, 02:15 PM
Nice parting shot. But let me call your bluff.
The point in question was if you look three times at an event that has a probability of being seen equal to one third, then, as Tim posted, the chances are almost 1/1. I showed that this is wrong.
No, you fouled up your little lesson but good ...

Probability of seeing it first time = 1/3
Probability of seeing it second time = 2/3 x 1/3
Oops, wrong already. Some lesson. The probability of seeing it the second time is 1/3, and so are the odds of seeing it on any other trial. Now, suppose I am going to do 2 trials, and I want to know the odds of seeing it at least once in those 2 trials. Thats 2 trials * 1/3 per trial = 2/3. And for three trials it's 3 * 1/3 = 1 exactly. And for 4 trials it's 4/3, which is greater than 1, which means it will happen (the same as =1).

Count on me being gone a few days observing on a mountaintop. See if you can learn how to do arithmetic between now & then.

snowflakeuniverse
2006-Jun-22, 03:17 PM
Post 93 of my thread on critiques of the proposed Uniform Expansion Theory gives a physical explanation for the lack of observed time dilation in the energy variation from quasars with respect to cosmological red shift.
http://www.bautforum.com/showthread.php?t=42188&page=4

Snowflake

Sp1ke
2006-Jun-22, 04:03 PM
I know a lot of the probability banter is off the original topic but isn't the example in post 66 more like:

Probability of it not occurring in first trial: 2/3 = ~0.67
Probability of it not occurring in first or second trial: 2/3*2/3 = ~0.44
Probability of it not occurring in all three trials: 2/3*2/3*2/3 = ~0.3

So there's never a 100% chance of seeing the event although the more trials you do, the less chance there is of not seeing it.

lyndonashmore
2006-Jun-22, 07:39 PM
No, you fouled up your little lesson but good ...

Oops, wrong already. Some lesson. The probability of seeing it the second time is 1/3, and so are the odds of seeing it on any other trial. Now, suppose I am going to do 2 trials, and I want to know the odds of seeing it at least once in those 2 trials. Thats 2 trials * 1/3 per trial = 2/3. And for three trials it's 3 * 1/3 = 1 exactly. And for 4 trials it's 4/3, which is greater than 1, which means it will happen (the same as =1).

Count on me being gone a few days observing on a mountaintop. See if you can learn how to do arithmetic between now & then.
So Tim the
Don't try this at home - We're what you call "professionals"
When you saw it on the second time, what happened the first time?
My post of "If the BB was a horse, these results show that the odds of it being correct are 7 to 3 against" was/is perfectly correct.
I am not into humiliating people Tim.
Quit whilst you are ahead.
Like I am going to do.
Cheers,
lyndon

Nereid
2006-Jun-23, 01:18 AM
... as enunciated in this thread.

This method/math may prove useful, in future discussions; for example, on lyndonashmore's ideas on H.

lyndonashmore
2006-Jun-23, 06:22 AM
... as enunciated in this thread.

This method/math may prove useful, in future discussions; for example, on lyndonashmore's ideas on H.
"You ain't seen nothin yet!"
Cheers,
Lyndon

Ari Jokimaki
2006-Jun-23, 07:03 AM
Oops, wrong already. Some lesson. The probability of seeing it the second time is 1/3, and so are the odds of seeing it on any other trial. Now, suppose I am going to do 2 trials, and I want to know the odds of seeing it at least once in those 2 trials. Thats 2 trials * 1/3 per trial = 2/3. And for three trials it's 3 * 1/3 = 1 exactly. And for 4 trials it's 4/3, which is greater than 1, which means it will happen (the same as =1)
Let's apply this to coin flipping:

Probability of getting tails in coin flipping is 1/2 for one trial. So if things work as you say, probability for two trials will be 2/2 = 1. This means that if we flip a coin two times, we will get one occurance of tails. Getting heads has the same probability, so every time we flip a coin two times we will get heads once and tails once. In the end this means that when you are flipping coin your results will look like this:

..., heads, tails, heads, tails, heads, tails, heads, tails, heads, tails, heads, tails, heads, tails, heads, tails, ...

You can easily test this, can you do it?

P.S. If there's need for responses, I'll be away for a week.

Thanatos
2006-Jun-23, 07:08 AM
There isn't a tired light mechanism that has been shown to cause tired light, let alone time dilation.Expansion of the universe might work! Aside from that, all 'tired light' theories have serious issues with energy conservation. And, none of them work very well when confronted with observational evidence [e.g., the black body spectrum of the CMB].

Lyndon, I think your model relies on too many 'a priori' assumptions to be palatable. For example, a coordinate system.

papageno
2006-Jun-23, 01:14 PM
I thought that my question in post #64 (http://www.bautforum.com/showpost.php?p=767417&postcount=64) was clear, direct and pertinent.
So, Lyndon, how about giving an answer before you go on holiday?

lyndonashmore
2006-Jun-23, 02:59 PM
Papageno wrote:

So, which one requires the shorter inductive leap?
Time dilation (BBT) with ~1 sigma or No Time Dilation with ~2.5 sigma?

Unfortunately your question is off topic. The OP was that the Big bang idea makes a wrong prediction and is therefore probably (7 to 3 against if it were a horse) not the correct interpretation. Two wrongs don't make a right regardless of which is nearest!
Cheers,
Lyndon

Nereid
2006-Jun-23, 03:30 PM
(with one or two, very minor, edits):
[snip]

Unfortunately your question is off topic. The OP was that the {X} idea makes a wrong prediction and is therefore probably ([a] to [b] against if it were a horse) not the correct interpretation. Two wrongs don't make a right regardless of which is nearest!
Cheers,
{BAUT member}Yep, likely to be a good piece of boilerplate ...

papageno
2006-Jun-23, 03:56 PM
Unfortunately your question is off topic. The OP was that the Big bang idea makes a wrong prediction and is therefore probably (7 to 3 against if it were a horse) not the correct interpretation. Two wrongs don't make a right regardless of which is nearest!
How can it be off-topic? It is about the paper you introduced in the OP.
If the BBT gives a prediction which agrees with the experimental result within 1 sigma, then any undergraduate student can tell you that the BBT does not fail this test.

You obviously have no experience whatsoever of measurements and experimental tests of theoretical predictions. Otherwise you would not try to argue that a correspondence between prediction and experiment within the errors, means that the theory failed the test.

You are simply trying to avoid giving an answer, because you would have to accept that the predictions of your "theory" do not get closer to the experimental results than the mainstream theories, therefore excluding your "theory" as a viable alternative to the BBT.

So, which one requires the shorter inductive leap?
Time dilation (BBT) with ~1 sigma or No Time Dilation with ~2.5 sigma?

lyndonashmore
2006-Jun-23, 04:17 PM
How can it be off-topic? It is about the paper you introduced in the OP.
If the BBT gives a prediction which agrees with the experimental result within 1 sigma, then any undergraduate student can tell you that the BBT does not fail this test.
The paper was introduced not as the topic but as reference material in suppport of the OP - the Big Bang Idea cannot correctly predict the amount of time dilaton. In any case it is "not within 1 sigma", it is outside 1 sigma (1.03 sigma)


You obviously have no experience whatsoever of measurements and experimental tests of theoretical predictions. Otherwise you would not try to argue that a correspondence between prediction and experiment within the errors, means that the theory failed the test.
As I have posted before, this result does not represent success either - in fair minded people, the jury is still out on how well the BBi predicts "time dilation'.


You are simply trying to avoid giving an answer, because you would have to accept that the predictions of your "theory" do not get closer to the experimental results than the mainstream theories, therefore excluding your "theory" as a viable alternative to the BBT.
Off topic, nowhere in this thread have I mentioned my theory - but it is good isn't it?


So, which one requires the shorter inductive leap?
Time dilation (BBT) with ~1 sigma or No Time Dilation with ~2.5 sigma?

As I said before:

The OP was that the Big bang idea makes a wrong prediction and is therefore probably (7 to 3 against if it were a horse) not the correct interpretation. Two wrongs don't make a right regardless of which is nearest!

Cheers,
Lyndon
P.S. Did you finally get to the patent office?

JimTKirk
2006-Jun-23, 06:06 PM
<snip>
Off topic, nowhere in this thread have I mentioned my theory - but it is good isn't it?



<snip>

Sure, as long as you ignore Posts 42 and 45...:naughty:

lyndonashmore
2006-Jun-23, 06:47 PM
Sure, as long as you ignore Posts 42 and 45...:naughty:
these posts were made in response to popular demand. If people bring the matter up then I must reply. BTW you forgot about post 78!

JimTKirk
2006-Jun-23, 07:14 PM
these posts were made in response to popular demand. If people bring the matter up then I must reply. BTW you forgot about post 78!

Does that then mean that Papageno's question is not off-topic and you will answer it?


Edit: Since I was responding to Post 78, it wasn't included for obvious reasons.

papageno
2006-Jun-23, 07:14 PM
How can it be off-topic? It is about the paper you introduced in the OP.
If the BBT gives a prediction which agrees with the experimental result within 1 sigma, then any undergraduate student can tell you that the BBT does not fail this test.
The paper was introduced not as the topic but as reference material in suppport of the OP - the Big Bang Idea cannot correctly predict the amount of time dilaton. In any case it is "not within 1 sigma", it is outside 1 sigma (1.03 sigma)

And if you knew anything about measurements, you would know that a difference between prediction and experimental result of ~1 sigma does not mean that the prediction fails the test.
First semester undergraduate students would see that your reference does not disprove the prediction of the BBT.






You obviously have no experience whatsoever of measurements and experimental tests of theoretical predictions. Otherwise you would not try to argue that a correspondence between prediction and experiment within the errors, means that the theory failed the test.

As I have posted before, this result does not represent success either - in fair minded people, the jury is still out on how well the BBi predicts "time dilation'.

And for people who actually have an understanding of measurements, the time dilation predicted by the BBT fits the experimental result better than any theory predicting no time dilation.
You are the one subscribing to a black-or-white-like fallacy ("either the BBT agrees spot-on or it is disproven"), and failing to understand that science is about finding the explanation that fits the observation better than any other. That's why I quoted JayUtah: "The question -- the only proper question, that is -- is whether the inductive leap associated with Y is greater or lesser than X's leap."
And from the very reference you quoted in the OP, the inductive leap for the BBT predicted time dilation is definitely shorter that the one for no time dilation.






You are simply trying to avoid giving an answer, because you would have to accept that the predictions of your "theory" do not get closer to the experimental results than the mainstream theories, therefore excluding your "theory" as a viable alternative to the BBT.

Off topic, nowhere in this thread have I mentioned my theory - but it is good isn't it?

No, it isn't. We showed that it is completely disproven by experimental results and the rest of physics in the old BABB thread.
And it is not off-topic, since you consider bashing the BBT as integral part of presenting your own "theory": from the "How good are the best alternatives to the Big Bang theories? (http://www.bautforum.com/showthread.php?t=42172)" thread, post #80 (http://www.bautforum.com/showpost.php?p=757452&postcount=80)


I thik that you realise that your Op ties everyones hands behind their backs and then expects them to fight.
Either we open up this OP to include what the BB cannot do compared to what ATM threads can or I am leaving.

Once you introduced your "theory" in this thread (posts #42 (http://www.bautforum.com/showpost.php?p=765632&postcount=42) and # 45 (http://www.bautforum.com/showpost.php?p=766058&postcount=45)), you untied our hands from behind our backs.






So, which one requires the shorter inductive leap?
Time dilation (BBT) with ~1 sigma or No Time Dilation with ~2.5 sigma?
As I said before:


The OP was that the Big bang idea makes a wrong prediction and is therefore probably (7 to 3 against if it were a horse) not the correct interpretation. Two wrongs don't make a right regardless of which is nearest!


It is not "two wrongs".
It is "one more accurate than the other".

So, which one requires the shorter inductive leap?




P.S. Did you finally get to the patent office?

My French was not good enough. But I am getting a paper accepted by Physical Review Letters.

lyndonashmore
2006-Jun-23, 08:07 PM
And if you knew anything about measurements, you would know that a difference between prediction and experimental result of ~1 sigma does not mean that the prediction fails the test.
First semester undergraduate students would see that your reference does not disprove the prediction of the BBT.
So, are you saying that a result over 1 sigma off proves the Big bang Idea?




And for people who actually have an understanding of measurements, the time dilation predicted by the BBT fits the experimental result better than any theory predicting no time dilation.
You are the one subscribing to a black-or-white-like fallacy ("either the BBT agrees spot-on or it is disproven"), and failing to understand that science is about finding the explanation that fits the observation better than any other. That's why I quoted JayUtah: "The question -- the only proper question, that is -- is whether the inductive leap associated with Y is greater or lesser than X's leap."
And from the very reference you quoted in the OP, the inductive leap for the BBT predicted time dilation is definitely shorter that the one for no time dilation.
Over 1 sigma off is hardly spot on. I don't believe in "inductive leaps". Ideas such as the BBI either make predictions that get it right (odds on favorite at least?) or they are "not proven".


No, it isn't. We showed that it is completely disproven by experimental results and the rest of physics in the old BABB thread.
You did no such thing

And it is not off-topic, since you consider bashing the BBT as integral part of presenting your own "theory": from the "How good are the best alternatives to the Big Bang theories? (http://www.bautforum.com/showthread.php?t=42172)" thread, post #80 (http://www.bautforum.com/showpost.php?p=757452&postcount=80)

Once you introduced your "theory" in this thread (posts #42 (http://www.bautforum.com/showpost.php?p=765632&postcount=42) and # 45 (http://www.bautforum.com/showpost.php?p=766058&postcount=45)), you untied our hands from behind our backs.
No. Sorry. I did not introduce it, I replied on a point of information.




It is not "two wrongs".
It is "one more accurate than the other".

So, which one requires the shorter inductive leap?
It's all about winning and losing. Neither of these wins. Ergo BBi is not proven.


My French was not good enough. But I am getting a paper accepted by Physical Review Letters.
Congratulatons on the paper. Well deserved. Serously, I always welcomed your contributions. You spend a lot of time and make us think. it is people like you that make this board worthwhile posting on.
My Fench often falls me on this board too!

korjik
2006-Jun-23, 08:32 PM
It isnt about winning and losing, it is about discovering the truth. Your assertion in the original post is wrong. The paper says that expansion fits fairly well and non-expansion dosent fit very well at all. No one here is taking it as proof. Expansion passed this hurdle, Non-expansion didnt. There is no proof here.

You would be better off trying to figure out why the data fits expansion better than trying to use this paper to tear down the BB

lyndonashmore
2006-Jun-23, 08:43 PM
It isnt about winning and losing, it is about discovering the truth. Your assertion in the original post is wrong. The paper says that expansion fits fairly well and non-expansion dosent fit very well at all. No one here is taking it as proof. Expansion passed this hurdle, Non-expansion didnt. There is no proof here.

You would be better off trying to figure out why the data fits expansion better than trying to use this paper to tear down the BB
Believe me korjk, science is about winnng and losing!
There was 'time dilation'.
the BBI said that there should not be any n a static universe.
The BBI said that there should be in an expanding one.
the Bbi found it to be true.
Bit based don't you think?
It was the same with CMB.
Steady state, at the time, said there should be no CMB, BBI said that there should be (mainly cos they knew it already existed)
ergo it was found!
but steady state theory and Big Bang idea, both assumed expansion. all the rediscovery of the CMB did was to separate between expansionists. it did nothing for other theories.
Same here.
BBI interpretation of Non expansion says no time dilation. BBI says there should be. BBI find it.
Does this prove the BBi correct?
No.
Just means that they are both wrong cos neither predict it right..
Cheers,
Lyndon

Celestial Mechanic
2006-Jun-23, 08:58 PM
[Snip!] Steady state, at the time, said there should be no CMB, BBI said that there should be (mainly cos they knew it already existed), ergo it was found! [Snip!]
Strange, the claim has been made that steady-state predicted a microwave background and gave a temperature closer to it than the first predictions in the big bang cosmologies. I'll have to look for the references.

papageno
2006-Jun-23, 10:47 PM
And if you knew anything about measurements, you would know that a difference between prediction and experimental result of ~1 sigma does not mean that the prediction fails the test.
First semester undergraduate students would see that your reference does not disprove the prediction of the BBT.
So, are you saying that a result over 1 sigma off proves the Big bang Idea?
No, I said It does not disprove the BBT prediction, and therefore your claim that the reference in the OP disproves the BBT time dilation is wrong.







And for people who actually have an understanding of measurements, the time dilation predicted by the BBT fits the experimental result better than any theory predicting no time dilation.
You are the one subscribing to a black-or-white-like fallacy ("either the BBT agrees spot-on or it is disproven"), and failing to understand that science is about finding the explanation that fits the observation better than any other. That's why I quoted JayUtah: "The question -- the only proper question, that is -- is whether the inductive leap associated with Y is greater or lesser than X's leap."
And from the very reference you quoted in the OP, the inductive leap for the BBT predicted time dilation is definitely shorter that the one for no time dilation.
Over 1 sigma off is hardly spot on.
It is more than enough to say that the BBT is not disproved by that result.

By claiming that the paper disproves the BBT prediction, you imply in a not-so-subtle manner that only spot-on predictions would be acceptable as support for the BBT. And this is a tell-tale of your "either the BBT agrees spot-on or it is disproved"-attitude.




I don't believe in "inductive leaps". Ideas such as the BBI either make predictions that get it right (odds on favorite at least?) or they are "not proven".
There you go: "either the BBT agrees spot-on or it is disproven".

This only shows that you have no experience in experimental science. Otherwise you would not expect exact experimental results, ignoring the errors that are unavoidable in measurements.
You clearly threat the experimental results as having infinitely high accuracy, so that if a theoretical prediction is not spot-on then it must be wrong. Well, that's not how the Universe works.

You know what my first experiment as an undergraduate student was? Measuring the internal angles of triangles we had drawn on paper, and test whether the sum the angles gave 180 degrees. Guess what happened? We got (179 +/- 2) degrees.
Try it yourself, and then tell us with a straight face that if the result is not spot-on (180 +/- 0) degrees, then such an experiment disproves Euclidean geometry.

If you do not accept inductive leaps when comparing theory and experiments, then you are not doing science.






No, it isn't. We showed that it is completely disproven by experimental results and the rest of physics in the old BABB thread.
You did no such thing
As usual you stick your head in the sand.






Once you introduced your "theory" in this thread (posts #42 (http://www.bautforum.com/showpost.php?p=765632&postcount=42) and # 45 (http://www.bautforum.com/showpost.php?p=766058&postcount=45)), you untied our hands from behind our backs.
No. Sorry. I did not introduce it, I replied on a point of information.
Read again: you argued about your "theory" over several posts. Thus your "theory" became fair game.

So, show us that your "theory" yields predictions that are spot-on.






It is not "two wrongs".
It is "one more accurate than the other".

So, which one requires the shorter inductive leap?
It's all about winning and losing. Neither of these wins. Ergo BBi is not proven.
So, you are not doing science.
You are just trying to play a modern-day Galileo. Then show us that your "theory" 'wins'.

Nereid
2006-Jun-24, 02:53 AM
Here (http://www2.slac.stanford.edu/tip/special/cp.htm) is a recent announcement concerning an observation of "direct CP violation", involving B and B-bar mesons.

The relevant phrase, wrt my question, is:
According to theory, CP symmetry would dictate that the two events have the same odds of happening. Hence, by starting with equal numbers of ** and anti-** one should end with equal numbers of K+pi- and K-pi+ pairs. However, the BaBar collisions produced 910 K+pi- pairs but only 696 K-pi+ pairs.Applying the "lyndonashmore method" to this, how did the horses finish?

lyndonashmore
2006-Jun-24, 04:14 AM
You know what my first experiment as an undergraduate student was? Measuring the internal angles of triangles we had drawn on paper, and test whether the sum the angles gave 180 degrees. Guess what happened? We got (179 +/- 2) degrees.
Try it yourself, and then tell us with a straight face that if the result is not spot-on (180 +/- 0) degrees, then such an experiment disproves Euclidean geometry.

If you do not accept inductive leaps when comparing theory and experiments, then you are not doing science.

And was this result 1.03 sigma out? Was the chance of getting 179 degrees by chance 30%?
No it was not, Ergo we accept the result as consistent - particularly because the rest of the class of '69 got similar results too.
In the 2005 paper on time dilation, the result was 1.03 sigma out one result, proves nothing.
The BB idea is not proven by time dilation results, and never comes close.
Cheers,
Lyndon.
P.S.

You know what my first experiment as an undergraduate student was? Measuring the internal angles of triangles we had drawn on paper,
Is this really true? I knew that standards had fallen but not by this much.

lyndonashmore
2006-Jun-24, 04:18 AM
Here (http://www2.slac.stanford.edu/tip/special/cp.htm) is a recent announcement concerning an observation of "direct CP violation", involving B and B-bar mesons.

The relevant phrase, wrt my question, is:Applying the "lyndonashmore method" to this, how did the horses finish?
We don't know yet. This is only a qualifier - and a totally different racecourse.
Remember the three R's of Physics.
Repeat, repeat, repeat.
Cheers,
Lyndon

Tobin Dax
2006-Jun-24, 05:12 AM
And was this result 1.03 sigma out? Was the chance of getting 179 degrees by chance 30%?
No it was not, Ergo we accept the result as consistent - particularly because the rest of the class of '69 got similar results too.
Wait, his average value was 179 degrees. His error was 2 degrees. The error was based on his measurements, so there was likely at least one triangle that was 177 or 178 degrees for the sum. That's one sigma away from 180, so papageno has disproven Euclidean geometry. His experiment failed at least once, so it's obviously been disproven.

Did I get your logic right?

lyndonashmore
2006-Jun-24, 06:55 AM
Wait, his average value was 179 degrees. His error was 2 degrees. The error was based on his measurements, so there was likely at least one triangle that was 177 or 178 degrees for the sum. That's one sigma away from 180, so papageno has disproven Euclidean geometry. His experiment failed at least once, so it's obviously been disproven.

Did I get your logic right?
In a word, No.
In this paper the result is 1.03 sigma away from the observed value. 30% by chance is quoted. That means that by chance, you would get a nearer match 70% of the time?
So with Papageno on his way to a degree in degrees (I wonder if that is why they call it a BA degree?) using lets say a 180 degree protractor, 70% of 180 is 126. The paper uses a two tailed test (63 degrees each side) so to be as far out as the Big Bang Idea, when Papageno measures his angle of 60 degree (say) he must measure it as either -3 degree or + 123 degree.
Would he still feel sure that the angles in his triangle add up to 180 degree?
I think not.
Nor do we feel that this results adds up to the BBi being correct.
Cheers,
Lyndon

Thanatos
2006-Jun-24, 09:09 AM
That falls squarely between the hoofprints, lyndanashmore.

Tobin Dax
2006-Jun-24, 09:48 AM
In a word, No.
In this paper the result is 1.03 sigma away from the observed value. 30% by chance is quoted. That means that by chance, you would get a nearer match 70% of the time?
So you do understand. Wonderful.

lyndonashmore
2006-Jun-24, 09:55 AM
So you do understand. Wonderful.
Ergo, if the BBi were a horse it would be 7 to 3 against - and put down!

papageno
2006-Jun-24, 10:31 AM
You know what my first experiment as an undergraduate student was? Measuring the internal angles of triangles we had drawn on paper, and test whether the sum the angles gave 180 degrees. Guess what happened? We got (179 +/- 2) degrees.
Try it yourself, and then tell us with a straight face that if the result is not spot-on (180 +/- 0) degrees, then such an experiment disproves Euclidean geometry.

If you do not accept inductive leaps when comparing theory and experiments, then you are not doing science.
And was this result 1.03 sigma out? Was the chance of getting 179 degrees by chance 30%?
You completely missed the point.
My example was not about the ~1 sigma, but about experimental errors. And you are supposed to try the experiment yourself, so you can get a minimum of experience in measurements.

If you feel it more appropriate I can change the result to (179 +/- 1) degrees.
So, would you consider Euclidean geometry disproved in this case?




No it was not, Ergo we accept the result as consistent - particularly because the rest of the class of '69 got similar results too.
In the 2005 paper on time dilation, the result was 1.03 sigma out one result, proves nothing.
The BB idea is not proven by time dilation results, and never comes close.
You completely ignored the rest of my post.
~1 sigma off does not disprove the BBT, and this is all we need to point out to prove your claim wrong.
And now show us that your "theory" gives spot-on predictions; if it does not, then you'll have to accept that it is disproven.







You know what my first experiment as an undergraduate student was? Measuring the internal angles of triangles we had drawn on paper, and test whether the sum the angles gave 180 degrees.
Is this really true? I knew that standards had fallen but not by this much.
You clearly have no clue.
What would you do in the first lesson in the first year teaching lab, since not all of the students came from scientifically-oriented high schools? (This isn't the American or British system.)
A simple experiment like this is very good if you want to focus on the concept of random fluctuations due to measurement errors and their statistical distribution.

Nereid
2006-Jun-24, 10:52 AM
Here is a recent announcement concerning an observation of "direct CP violation", involving B and B-bar mesons.

The relevant phrase, wrt my question, is:Applying the "lyndonashmore method" to this, how did the horses finish?We don't know yet. This is only a qualifier - and a totally different racecourse.
Remember the three R's of Physics.
Repeat, repeat, repeat.
Cheers,
LyndonInteresting.

Now suppose we could get our hands on BaBar's "old" result* ... how many "repeat"s would we now have? How would the race have been called?

Now suppose we add the results from the experiments on K mesons ... how many "repeat"s would we now have? How would the race have been called?

*"This fact enabled BaBar – and the Belle experiment at the KEK laboratory in Japan – to measure indirect CP violation for B mesons in a landmark 2001 result."

lyndonashmore
2006-Jun-24, 11:01 AM
Interesting.

Now suppose we could get our hands on BaBar's "old" result* ... how many "repeat"s would we now have? How would the race have been called?

Now suppose we add the results from the experiments on K mesons ... how many "repeat"s would we now have? How would the race have been called?

*"This fact enabled BaBar – and the Belle experiment at the KEK laboratory in Japan – to measure indirect CP violation for B mesons in a landmark 2001 result."
As I said, this is a totally different racecourse so we are not there to find out. Why not come back and join us on the Big Bang Derby and the supernovae time dilation gold cup?
Cheers,
Lyndon

Grey
2006-Jun-24, 11:27 AM
In a word, No.
In this paper the result is 1.03 sigma away from the observed value. 30% by chance is quoted. That means that by chance, you would get a nearer match 70% of the time?
So with Papageno on his way to a degree in degrees (I wonder if that is why they call it a BA degree?) using lets say a 180 degree protractor, 70% of 180 is 126. The paper uses a two tailed test (63 degrees each side) so to be as far out as the Big Bang Idea, when Papageno measures his angle of 60 degree (say) he must measure it as either -3 degree or + 123 degree.
Would he still feel sure that the angles in his triangle add up to 180 degree?
I think not.Huh? Why on earth are you multiplying 70% (the chance that a result lies within one standard deviation of the mean) by the actual value measured? That doesn't make any sense at all.

Nereid
2006-Jun-24, 11:48 AM
As I said, this is a totally different racecourse so we are not there to find out. Why not come back and join us on the Big Bang Derby and the supernovae time dilation gold cup?
Cheers,
LyndonAu contraire, mon ami, au contraire.

It may be that I have misunderstood "the lyndonashmore method" for interpreting confidence limits, doing probability calculations, and so on. However, what I do understand, from what I've read so far, in this thread, is that it is an "alternative" method - i.e. an approach to these aspects of science and math that one does not find in textbooks or papers.

Further, it is quite fundamental - it applies to all branches of science where quantitative data are obtained, analysed, and interpreted.

(If it's not fundamental, and applies only to astronomy, or astrophysics, or cosmology - or even, only to the analysis of data about distant SNe - please clarify).

Ergo, questions to the proponent of this ATM idea, concerning its scope, application, internal consistency, etc are entirely reasonable.

But then, I think we've the pleasure of your company for only 4 more days, n'est pas?

So perhaps an exploration of this new, alternative approach to handling quantitative data in science can wait until the return?

lyndonashmore
2006-Jun-25, 07:00 PM
Au contraire, mon ami, au contraire.

It may be that I have misunderstood "the lyndonashmore method" for interpreting confidence limits, doing probability calculations, and so on. However, what I do understand, from what I've read so far, in this thread, is that it is an "alternative" method - i.e. an approach to these aspects of science and math that one does not find in textbooks or papers.

Further, it is quite fundamental - it applies to all branches of science where quantitative data are obtained, analysed, and interpreted.

(If it's not fundamental, and applies only to astronomy, or astrophysics, or cosmology - or even, only to the analysis of data about distant SNe - please clarify).

Ergo, questions to the proponent of this ATM idea, concerning its scope, application, internal consistency, etc are entirely reasonable.

But then, I think we've the pleasure of your company for only 4 more days, n'est pas?

So perhaps an exploration of this new, alternative approach to handling quantitative data in science can wait until the return?
Hi Nereid,
welcome back!
I assume that someone posted some nooky spam with the result that this site has been blocked for the last 24hrs as being culturally and .... whatever.
Thankfully the internet here is censored and stops this sort of thing.
So where are we?
What about the rules?

1. Posting Topics

First and foremost, this discussion forum focuses upon space and astronomy. Please stay within these areas when creating topics and participating in existing conversations.
your post breaks this rule. I don't know any moderators, maybe you do. Please report it (joking really).
Particle Physics must be the most boring area of science and so i am not going onto that area on a space astronomy site.
Yep. going on my hols.
In the meantime let me leave you all with a little problem.
Papageno is trying to measure an angle of 90 degrees with his 180 degree protractor.
meanwhile, next door is Einstein who is guessing numbers.
Einstein guesses 30 degrees we say no.
Einstein guesses 89 degrees we say great.
Over a period of time, Einstein gets his random number answers closer than Papageno's 70% of the time. What number did Papageno decide upon?
Cheers,
Lyndon.

papageno
2006-Jun-25, 11:50 PM
Particle Physics must be the most boring area of science and so i am not going onto that area on a space astronomy site.
Yep. going on my hols.
In the meantime let me leave you all with a little problem.
Papageno is trying to measure an angle of 90 degrees with his 180 degree protractor.
meanwhile, next door is Einstein who is guessing numbers.
Einstein guesses 30 degrees we say no.
Einstein guesses 89 degrees we say great.
Over a period of time, Einstein gets his random number answers closer than Papageno's 70% of the time. What number did Papageno decide upon?
Let me guess: you finally realized that your initial claim is blatantly wrong. And it hurts so much that you cannot face it.
So you stick your head in the sand hoping it will all go away. As you usually do.

Of course, you could save your face if you admitted your error and finally answered my question:


So, which one requires the shorter inductive leap?
Time dilation (BBT) with ~1 sigma or No Time Dilation with ~2.5 sigma?

Why don't you try that little experiment with triangles during your holidays?

Eta C
2006-Jun-26, 12:49 AM
Particle Physics must be the most boring area of science and so i am not going onto that area on a space astronomy site.
Yep. going on my hols.
In the meantime let me leave you all with a little problem.
Papageno is trying to measure an angle of 90 degrees with his 180 degree protractor.
meanwhile, next door is Einstein who is guessing numbers.
Einstein guesses 30 degrees we say no.
Einstein guesses 89 degrees we say great.
Over a period of time, Einstein gets his random number answers closer than Papageno's 70% of the time. What number did Papageno decide upon?
Cheers,
Lyndon.

I'll break my ban on this topic to say that this comment simply shows Lyndon's ignorance of the status of cosmology. And I don't say this simply as a particle physicist who is feeling insulted. Lyndon, if you don't take particle physics into account you can say nothing about early universe cosmology. The topics of barygenesis, the formation of elements, the nature of the early universe, and others all require knowledge of particle physics. Or don't you think these are important issues in a cosmological theory? Particle physics and cosomology are more linked than you would seem to know. Statements like this simply display the vapidity of your arguments and the shallowness of your knowledge. I look forward to dissecting your "derivation" of the hubble constant. Until then, enjoy your vacation. I'll be waiting.


P.S. You still are wrong in your statistics. Nothing you've posted since I last commented has shown otherwise.

Nereid
2006-Jun-26, 01:03 AM
[snip]

Particle Physics must be the most boring area of science and so i am not going onto that area on a space astronomy site.
[snip]Perhaps, in time for your return, someone will have started a Q&A thread, on the relationship between particle physics and astronomy, astrophysics, and cosmology.

However, whether they will or not, I will state that an ATM idea which is (apparently) based on a substantial revision (shall we say) of confidence limits, probability, statistics, and so on automatically opens that idea to challenges in those areas (per BAUT rules).

Specifically, if you continue to insist on your interpretation of confidence limits, etc, per your posts in this thread so far, then questions to you about the application of your methods (as outlined in your posts in this thread) to aspects of particle physics that are pertinent to astronomy, astrophysics, and cosmology are quite OK ... and you will be expected to answer such questions in a timely fashion (per the rules).

Enjoy your hols!

Tim Thompson
2006-Jun-26, 11:42 AM
Ergo, if the BBi were a horse it would be 7 to 3 against - and put down!
I refer the reader to my post 51 (http://www.bautforum.com/showpost.php?p=766364&postcount=51) as a definitive proof that this is nonsensical, silly, wrong, and in all aspects deserving of a hearty laugh.

upriver
2006-Jun-28, 10:27 PM
on the relationship between particle physics and astronomy, astrophysics, and cosmology


There is no way to separate any of them.
They are all intertwined.

Sounds like ancient chinese proverb, its true, but to understand what you are seeing you must first understand what you are looking at.

OR

How do you know what you are looking at unless you know what you are looking at........

Nereid
2006-Jun-28, 11:04 PM
There is no way to separate any of them.
They are all intertwined.

Sounds like ancient chinese proverb, its true, but to understand what you are seeing you must first understand what you are looking at.

OR

How do you know what you are looking at unless you know what you are looking at........Shh! :shhh: Don't mention that to lyndonashmore!

Jerry
2006-Jul-10, 05:57 PM
I'll break my ban on this topic to say that this comment simply shows Lyndon's ignorance of the status of cosmology. And I don't say this simply as a particle physicist who is feeling insulted. Lyndon, if you don't take particle physics into account you can say nothing about early universe cosmology. The topics of barygenesis, the formation of elements, the nature of the early universe, and others all require knowledge of particle physics. Or don't you think these are important issues in a cosmological theory?
That depends upon the theory. Obviously it is important to the BB, but less so in a steady state scenario. However, to ignore particle physics, a SS needs to address hydrogen synthesis without a sub-atomic synthesis.

I was watching a NASA TV presentation on the BB the other day and the narration went something like this: Beyond the galaxies and structures that we see today, there was a period that is not well understood, it might have been inflation, but we do know exactly what happened before that episode.

I have a real problem with the logic here. Particle phycists have worked out a strong genesis, but no exodus, and without an exodus, for all we know the ten tribes came from Canada. All the pieces have to be in place, before we can be certain of the beginning. Sometimes 'we do not know' is as good of an answer as there is.