Metatheory and the NIST report

take a look around the forum R.Mackey. Our young friends REV and Fates have stirred things up at another Moon Hoax thread, and most indearingly called all the astronauts LIARS.

TAM:)

Yes, I know. Kid's got quite a mouth on him.

As a practicing scientist working for NASA, I feel qualified to state unequivocally that anyone who denies the Moon Landings is beneath notice, and eminently lacks the intellectual capacity to contribute to any meaningful discussion. Such individuals may assume they are Ignored by default.
 
I figured as much...leave him to the others, he may actually survive in one piece...

TAM:)

Edit: I noticed he posted before me here, but I cannot read it...likely more trash talk.

TAM:)
 
If I'm not mistaken, NIST performed tests separate from the UL assembly certifications. Either way, you are claiming that the tests bore no relevance to NIST's collapse theory, right? So then, by what standards does the NIST theory appeal to in order to make sense? The simulations alone?

They do bear a degree of relevance, and can be seen to lend limited credence to parts of the NIST theory. Sagging of the floors was produced, and the test actually had to be halted for fear the assemblies would collapse if the diagonals buckled. The assemblies were also scaled down, meaning the effects of thermal expansion and other geometrical effects produced by fire demonstrated in the Cardington fire tests(conducted before 9/11) are less of a factor than they would have been on the full size assemblies on 9/11.

There is also a discussion of 'lessons from Cardington' at page 2 of this link:

http://www.era.lib.ed.ac.uk/bitstream/1842/1216/1/WTCpaper.pdf

Which is a full text of the Usmani et all abstract which I posted a few posts back, I still can't find a public full text for the other paper though.

So there is certainly experimental basis for the NIST claims of a collapse induced by fire. Though essentially they appeal to their simulations. However, these simulations are constructed from and based on observable and measurable events and structural responses, to ensure the highest possible degree of accuracy.

But that brings up one way to falsify the NIST theory which would be to demonstrate that their simulations did not represent observed events such as sagging of floor trusses and inward bowing of perimeter columns

It seems you are in agreement that the NIST theory is unable to explain these phenomena, correct?

yes, I think no stone should go unturned in regards to 9/11. like a poster above said, if there is a potential to weaken structures from phenomena such as this, it needs to be made clear to engineers. Even if it is not critical to collapse, there is no reason why they should not provide an explaination.
 
jay howard has no valid points. My last post on Page 7 remains completely unaddressed by him. In summary:

1. The NIST report is falsifiable, as described here over several pages

All that would be required to back this up is to add into the reply to this, WHAT criteria would falsify the NIST report.

Please keep in mind that falsifiability, as it has been explained since the OP, is not a relative quality: a theory either has it or it doesn't. To be a "good" theory, it must be falsifiable but not false. That is, there must exist some real criteria by which it COULD be false, but tests to demonstrate that criteria have failed, and thus, we have a "good" explanation for some phenomenon.

I'm sure you can see that no one, through these 12 pages or so, has been either able or willing to lay out what these criteria might be. It is my contention that they either do not exist (in which case the theory is non-falsifiable, and hence useless) or they are completely contingent on simulated data.

If they depend entirely on simulated data, and I find data in the simulations that is unrepresentative (by NIST's own admission) of the conditions in the towers that is used to corroborate their own theory, then why is that not a "valid point"?

2. The UL intact-truss testing has no direct bearing on the collapse model

Then has NIST ever tested its theory in the real world? No.

But this doesn't seem to bother you. If I made up a theory for how the towers could not have fallen from the planes and subsequent fires, and all I have to make my case are computer simulations, have I made a "good" case?

Likely you would say 'no.' Yet, when the official case is made from the same structure, you apparently have no trouble standing behind its conclusions. Why? Appealing to a theory based on simulations alone is not a good idea--regardless of the conclusions reached. We should be able to test the theory and confirm it or eliminate it. Why is that an unacceptable process in this case but not for competing theories?

3. He has conflated observations on the recovered steel to represent NIST's estimate of the entire volume of structural steel

My criticism is that they didn't use the recovered steel as a baseline for their estimates. Instead, they made up more severe cases based on no recovered evidence and categorized the recovered steel as uncharacteristic of the fires they theorize brought the towers down.

Structurally, there is a problem with giving the simulated data priority over the physical data.

Again, if you think I'm applying some kind of unfair standard here, imagine that I was making the opposite case with an argument that labeled the forensic evidence "uncharacteristic of the fires" in order to make the case, then I appealed entirely to a simulation. You would likely have a problem with that, right?

4. He did not, but may finally, understand that the WPI paper -- concerning only a single piece of steel from the Towers -- shows that the steel never even approached melting temperature

There were at least 2 pieces of steel in question: one from building 7 and one from either building 1 or 2.

The temperature is a concern, but not the primary concern for the steel in question. How did a eutectic reaction take place? One way or another, we must explain "high temperature corrosion and sulfidation" of the steel.

That is an issue that no official theory is willing or capable of explaining with any testability. (I haven't gone over Dr. Greening's work on this yet--mostly because I can't seem to find it. If anyone has a link, please post it.)

Any way you slice it, my original claim, that the NIST theory is inadequate to deal with this evidence, stands without challenge.

Basically all he's doing now is saying "nuh-uh." That's not useful. You'll also note that he declined to respond to my question of where he's copying his information from. Once in a while he still sticks in a tell-tale footnote, and those notes aren't from NIST.

I'm not sure why you're saying this. If anyone doubts that I'm in fact drawing from the NIST report, take a bit of a quote and use your document search feature to find it. If it doesn't turn up after that, then let me know.
 
Anyway, regarding the Ove Arup paper, this is one of the few legitimate criticisms of the NIST report in existence. I wholeheartedly agree that it has the potential to be the "best" theory, and would falsify the NIST report if found to be true. However, I question its accuracy.

If we put any stock in the NIST fire and structural heat transfer models, they predict burnout on the impact floors after about two hours. Had there been no fireproofing damage, given what we saw in the truss tests, the maximum temperature in the floor elements would be expected to peak at under 500 oC and lead to deflections within the normal range of elasticity, thus the thermal expansion and subsequent contraction would not materially increase the strain energy in the support columns. The NIST models may be wrong, but this would not be a simple mistake, and we need much more evidence to draw such a conclusion.

If we assume no impact damage at all, i.e. no broken windows and the fire ventilated mainly by stack effect, then we can expect an even cooler if longer-duration fire. This is not, of course, what the Arup paper proposes, but we need to be clear what we mean when we say "no impact damage."

It's definitely worth comparing the two side-by-side. I still put more stock in the NIST result, but if you want to criticize NIST, this is the way to do it. Dr. Usmani has done his homework, as few others have.

by truss tests are you also referring to the UL tests? I would say that this is a poor representation of the Usmani/Arup theory as well.

only one floor assembly was tested at UL whereas the U/A theory involves the interaction of several assemblies to produce the buckling mechanism for the floor trusses.

I actually emailed Arup fire a few days ago to see if they are still working on this, since all of this work was produced before the publication of the NIST final report.
 
by truss tests are you also referring to the UL tests? I would say that this is a poor representation of the Usmani/Arup theory as well.

only one floor assembly was tested at UL whereas the U/A theory involves the interaction of several assemblies to produce the buckling mechanism for the floor trusses.

I actually emailed Arup fire a few days ago to see if they are still working on this, since all of this work was produced before the publication of the NIST final report.

I'm aware of Dr. Usmani's theory, having read the paper some time ago. I'm only referring to the expected fireproofing of each truss, not trying to extrapolate structural interaction or localized fire effects from the NIST NCSTAR1-6B testing.

To work out the total structural response, you need a global model, such as NCSTAR1-6D. As you state, you cannot predict the interaction from a single floor test, and that was not my intention.
 
Last edited:
All that would be required to back this up is to add into the reply to this, WHAT criteria would falsify the NIST report.

I'm sure you can see that no one, through these 12 pages or so, has been either able or willing to lay out what these criteria might be.

Are you even reading this thread? Do you have me on ignore?

I've given you blatant and obvious examples of falsifiability on two separate occasions and you've seemed to ignore me at every turn and continue to repeat that you are waiting for answers. How long do you plan on waiting for an answer whilst ignoring it? That's a bad combination, to be sure.
 
Last edited:
Are you even reading this thread? Do you have me on ignore?

I've given you blatant and obvious examples of falsifiability on two separate occasions and you've seemed to ignore me at every turn and continue to repeat that you are waiting for answers. How long do you plan on waiting for an answer whilst ignoring it? That's a bad combination, to be sure.
If Jay can not show the NIST theory to be falsifiable, he is not qualified to post in his own thread.
 
I'm aware of Dr. Usmani's theory, having read the paper some time ago. I'm only referring to the expected fireproofing of each truss, not trying to extrapolate structural interaction or localized fire effects from the NIST NCSTAR1-6B testing.

To work out the total structural response, you need a global model, such as NCSTAR1-6D. As you state, you cannot predict the interaction from a single floor test, and that was not my intention.

Fair enough, I agree Usmani's 2D models, as he admits as well, don't give the full picture. Arup leaves a cliffhanger as well:

Our views presented here will be amended based on our own
work progress and when NIST publishes its thermo-mechanical
response analyses which are not currently in the public
domain.

which they still haven't published as far as I know, making it difficult to find some standard to compare the two at this point.
 
Fair enough, I agree Usmani's 2D models, as he admits as well, don't give the full picture.

Yup. He is a responsible scientist...

As I said before, I still consider this a valid alternative, and one of the very few pieces of legitimate criticism of the NIST report. I also think, for the reasons I gave above, that the NIST report is more likely to be correct. But it will take more work to answer this for sure. If Dr. Usmani has followed up on this, and if you get an answer, I'd be interested to see what they have to say.
 
People who really believe man went to the moon are deranged.

The space shuttle has never gone further than 400 miles up. When it did, they beat a hasty retreat due to the radiation.


Rev, please--just stop.

Moon-landing deniers are the dumbest conspiracy loons in existence and that's saying a lot! They have absolutely nothing that supports their idiotic beliefs. The evidence that Americans (there's that dirty word) went to the moon several times is overwhelming. You'll note that the Russians don't dispute our achievement.
 
Are you even reading this thread? Do you have me on ignore?

I've given you blatant and obvious examples of falsifiability on two separate occasions and you've seemed to ignore me at every turn and continue to repeat that you are waiting for answers. How long do you plan on waiting for an answer whilst ignoring it? That's a bad combination, to be sure.


So I went back and found this:

Let's presume tomorrow evidence came to light that the WTC towers were still, in fact, standing. Under this new piece of evidence, do you think we could safely conclude that the NIST report was false?

Fair enough. Sorry I missed it. Probably missed a few others too.

Yes. That is a perfect example of the kind of evidence that would falsify the report. Now, we all agree the towers are not standing, so what is an example that would put this debate to rest? Like "steel truss sections of x design and y quality were heated for m minutes at t temp. With expected loads, they failed within these parameters."

Then we can agree to what counts as a good test of failure.

There are plenty of accompanying NIST documents I haven't read--mostly the evacuation and response accounts. A good chunk of the engineering is beyond me. But ultimately, the theory is given in these generalized ways, like these (with my emphases):

“Following the impact of the aircraft, the jet-fuel-ignited fires created the sustained and elevated temperatures that heated the remaining building structure to the point of collapse initiation.”(118)

and

“… the fires weakened the core columns and caused the floors on the south side of the building to sag. The floors pulled the heated south perimeter columns inward, reducing their capacity to support the building above. Their neighboring columns quickly became overloaded as columns on the south wall buckled. The top section of the building tilted to the south and began its descent. The time from aircraft impact to collapse initiation was largely determined by how long it took for the fires to weaken the building core and to reach the south side of the building and weaken the perimeter columns and floors.”(xliii, xliv)

and

“The steady burning fires on the east side of the building caused the floors there to sag. The floors pulled the heated east perimeter columns inward, reducing their capacity to support the building above. Their neighboring columns quickly became overloaded as columns on the east wall buckled. The top section of the building tilted to the east and to the south and began its descent. The time from aircraft impact to collapse initiation was largely determined by the time for the fires to weaken the perimeter columns and floor assemblies on the east and the south sides of the building. WTC 2 collapsed more quickly than WTC 1 because there was more aircraft damage to the building core and there were early and persistent fires on the east side of the building, where the aircraft had extensively dislodged insulation from the structural steel.”(xliv)

But all of this is based on the simulations. They aren't basing their case on the available evidence--or at least they are ignoring its centrality to what I think we can all agree was an act of arson. We do agree on that, right?

If this is an arson investigation, it should be handled carefully--not with quiet leaps of logic to use the simulations as more relevant than the physical evidence.

If you agree with NIST, you've got to concede that the simulations are more relevant than the empirical data. Why, in this of all cases, is the simulated data "more real" than the steel which in the NIST case, doesn't get above 600C for more than 15 minutes?
 
Yes. That is a perfect example of the kind of evidence that would falsify the report

Alright. So we can put to bed the issue of the report being non-falsifiable, right?

It is very obviously falsifiable. Of this, we must certainly agree.

Like "steel truss sections of x design and y quality were heated for m minutes at t temp. With expected loads, they failed within these parameters."
Something along those lines would certainly qualify as an experiment that could falsify the NIST report. In the extreme it would be "rebuild the entire wtc and set N floors on fire of temp K for time T". For example, if you set 15 low floors of a replica WTC on fire and held it for 2 hours at 1200-C... I'd say you pretty much falsified the NIST report in the extreme.

If you agree with NIST, you've got to concede that the simulations are more relevant than the empirical data.
I do not agree, in the slightest. Simulations are only useful when and where we lack empirical data.

Why, in this of all cases, is the simulated data "more real" than the steel which in the NIST case, doesn't get above 600C for more than 15 minutes?
I guess this is the 'point' of your previous statement. You are trying to argue that NIST is using simulation to override empirical data. I disagree vehemently.

The argument that the maximum temperatures of the fires was 600C is faulty and is not based on empirical data. The empirical data you are referencing found 600C to be a lower bound of the maximum temperature, not an upper bound. The sample size was nowhere near what would be required to make an educated guess on the upper bound. We do not have enough empirical evidence to make that claim. We don't have anything of the sort. The physical evidence was heavily contaminated by the post-collapse environment to make any evidence of the sort reliable.

Similarly, the temperature experiment NIST did perform was certainly not a simulation. That experiment had actual empirical data revealing a temperature far higher then 600C. That is empirical evidence. Evidence NIST used to shape their simulations.
 
Last edited:
Alright. So we can put to bed the issue of the report being non-falsifiable, right?

It is very obviously falsifiable. Of this, we must certainly agree.

It is falsifiable. So, what are the criteria which would demonstrate its falsity? How do we test to confirm it?

Something along those lines would certainly qualify as an experiment that could falsify the NIST report. In the extreme it would be "rebuild the entire wtc and set N floors on fire of temp K for time T". For example, if you set 15 low floors of a replica WTC on fire and held it for 2 hours at 1200-C... I'd say you pretty much falsified the NIST report in the extreme.

So we look at the controlled heating of the floor assemblies and see if we can get any useful data from those tests.

Many here only take those results as far as they say something corroborative or ambivalent about the official theory. When the truss systems do not fail, the report calls their failure "immanent". That's a loaded concept. They could as truthfully, (and do elsewhere) say "none of the trusses failed in the time and temperature restraints. These temperatures are hotter than the forensic evidence indicates."

I do not agree, in the slightest. Simulations are only useful when and where we lack empirical data.

On what grounds are they claiming that the hottest recovered beams were not representative of the steel in the collapse initiation? The simulations. If you agree with NIST that “none of the recovered steel samples showed evidence of exposure to temperatures above 600ºC for as long as 15 min," then how is it determined that these pieces corroborate the NIST theory?

“These results provide some confirmation of the thermal modeling of the structures, since none of the samples were from zones where such heating was predicted.”(176)

How do they know that? It was "predicted" in the "thermal modeling". This means someone had to make up data from which to determine the baseline temperatures of the failed steel--in order to determine that the 600C steel "confirmed" their hypothesis.

Do they have any specific forensic data that would solidify this case? A floor truss knuckle that displayed predicted deformations that also indicates a temperature of 700C for 30 minutes? If they do, please wheel it out.

The argument that the maximum temperatures of the fires was 600C is faulty. We do not have enough empirical evidence to make that claim.

So where is the empirical evidence that shows those 600C samples were on the low side of the temperatures that day? The simulations.

The physical evidence was heavily contaminated by the post-collapse environment to make any evidence of the sort reliable.

It doesn't change the fact that the baseline is a simulation and the physical evidence is secondary.

Similarly, the temperature experiment NIST did perform was certainly not a simulation. That experiment had actual empirical data revealing a temperature far higher then 600C. That is empirical evidence.

Precisely. And found:

“All four test assemblies were able to withstand standard fire conditions for between 3⁄4 hour and 2 hours without exceeding the limits prescribed by ASTM E 119.”(141)

And, more generally,

“All four test specimens sustained the maximum design load for approximately 2 hours without collapsing.”(141)

And we know it wasn't likely the result of faulty steel:

“Material substitutions of higher strength steels were not uncommon in the perimeter columns and floor trusses.”(176)

and

“87 percent of the tested steel specimens (columns, trusses and bolts) met or exceeded the required yield strengths specified in design documents.”(176)

And the other nearly 13 percent, NIST claims, can be accounted for by virtue of the destruction of the collapse. So this isn't a steel quality issue.
 
There's those footnotes again.

Who are you quoting?

Whomever it is, it isn't the NIST report. Whomever it is is cherry-picking quotes from hither and yon within the 10,000 pages, trying to stick together some stilted narrative that doesn't represent the actual thought process.

You haven't read the NIST report itself, just other people's mystifications of it. That is why you are confused.
 
There's those footnotes again.

Who are you quoting?

Whomever it is, it isn't the NIST report. Whomever it is is cherry-picking quotes from hither and yon within the 10,000 pages, trying to stick together some stilted narrative that doesn't represent the actual thought process.

You haven't read the NIST report itself, just other people's mystifications of it. That is why you are confused.

This is what's written at the bottom:

"NIST NCSTAR 1, WTC Investigation"

Here's the title:

"Final Report of the
National Construction Safety Team
on the Collapses of the
World Trade Center Towers (Draft)





For Public Comment"

Maybe my pagination is off for some reason? Are you not finding these excerpts in your copy?
 
... trying to stick together some stilted narrative that doesn't represent the actual thought process.

You haven't read the NIST report itself, just other people's mystifications of it. That is why you are confused.

That's just not true.

If you think I'm doing violence to their argument, please clarify. I've been as clear as possible in pointing why my interpretation is not an editorial, just a reordering of their argument in a sequential way.

Show me where I'm interpreting their theory unfairly.
 
This is what's written at the bottom:

"NIST NCSTAR 1, WTC Investigation"

Here's the title:

"Final Report of the
National Construction Safety Team
on the Collapses of the
World Trade Center Towers (Draft)





For Public Comment"

Maybe my pagination is off for some reason? Are you not finding these excerpts in your copy?


I find the excerpts, but like I said, you have consistently -- from your first post -- connected parts of the report that should not be connected.

I do not find that pagination, but I'm beginning to suspect you are looking at the draft report, and that you've only looked at NCSTAR1, not the nine project reports. The updated reports can be found here. Is this what's going on with your footnotes, or is it something else?

Tellingly, Page 141 of the actual NCSTAR1 report (the Final) shows two tables, titled "Regions in which temperatures of structural steel exceeded 600 oC." One for each Tower. This ought to quell your repeated and incorrect statements that NIST said none of the steel reached this temperature.
 
Last edited:
So where is the empirical evidence that shows those 600C samples were on the low side of the temperatures that day? The simulations.

No, the actual experiment conducted demonstrated this, not the simulation. You do know about the temperature experiment, do you not?

You've still missed the point. Finding 600C steel doesn't make 600C the upper bound of the maximum temperature. That is a fallacious argument. Do you understand this?

Unless you can demonstrate adequately that the samples were statistically randomly selected (which we know, for certain, they were not), it's impossible to make any inferences about the upper bound of the temperature by testing specific pieces, beyond the fact that the maximum must be greater then or equal to the maximum of the samples found.

All this 600-C piece of evidence tells us is that the maximum temperature attained was at least 600-C. To find out how much, if any, above that, we need to turn to other methods. NIST chose both a recreation experiment and a simulation.

“All four test assemblies were able to withstand standard fire conditions for between 3⁄4 hour and 2 hours without exceeding the limits prescribed by ASTM E 119.”(141)

And, more generally,

“All four test specimens sustained the maximum design load for approximately 2 hours without collapsing.”(141)

And we know it wasn't likely the result of faulty steel:

“Material substitutions of higher strength steels were not uncommon in the perimeter columns and floor trusses.”(176)

and

“87 percent of the tested steel specimens (columns, trusses and bolts) met or exceeded the required yield strengths specified in design documents.”(176)

And the other nearly 13 percent, NIST claims, can be accounted for by virtue of the destruction of the collapse. So this isn't a steel quality issue.
None of these tests represent a falsification of NIST as none of them were meant to mirror the conditions found that day.

The argument that any of these tests adequately test the NIST collapse initiation theory implies that you do not understand either the experiment or the theory.
 
Last edited:

ISF - Join now!

Every member here is approved by hand. No bots, no spam, just people who care about evidence and honest debate.

Membership is free!

Create your free account

Back
Top Bottom