WHY were the twin towers PURPOSELY collapsed?

[Rastaman]

Exactly AcidHead... I wonder why this undeniable fact is largely unspoken about. I mean, they were cought on video doing a classic bait-n-switch to keep the media away while Obama gets on a secret aircraft or limo and goes to meet Hillary...wherever..that...is...

[aquatus1]

we witnessed the buildings falling, it was completely like every single demolition i've ever seen in my life...

Too bad you have such limited experience. Implosions are the least frequent demolition type, and the least preferred by demolitionists.

Even at that, I find it amazing that you can look at a video of a genuine implosion, and one of the towers, and conclude that, because both buildings are falling in the same general direction, they are identical. You aren't much for detail.

And years back the guy who owned the WTC was on camera talking about building 7 and how he agreed to "Pull" the building, am i the only one who saw that?

No, pretty much everyone saw it. The difference is that not everyone is under the impression that the fire department listens to civilians regarding explosive demolitions. And fewer people bothered to ask what "Pull it" means to a firefighter (It means to withdraw and let the building burn down), as opposed to what "Pull it" means to a demolitionist (it means attach cables to the upper levels and pull the building down by force; incidentally, this is the most common form of demolition).

the jet fuel did not make the building collapse, i refuse to believe that...

So does pretty much anyone who has studied the collapse. In fact, the only people I here claiming that the fire was the sole cause of the collapse is the CT crowd when they try to attribute it to NIST.

on impact most of the fuel was in the explosion, remember when the FBI finally released the FDNY radio transmissions, the floor where they say was burning to much to even get to, well they have 1 FDNY worker on the radio on the same floor saying that they basically handle it...

Yes, most of the fuel went up in the explosion. That still left plenty to explode in the floors inside the building, and even some to go down the elevator shaft and explode there too.

And yes, there were firefighters who were exposed to parts of the fire that they were able to navigate in. That happens in fires. Some parts are not as bad as others.

workers of the Twin Towers, Police officers, FDNY workers and others have went on record of hearing explosions, saying that they heard bombs...

You are saying two things here. Yes, people heard explosions. Things blow up in fires. That does not mean that they heard bombs.

but a lot of people including you all on here think that the people who question or think theres more to it are crazy...its just funny

Not exactly. Questioning is wonderful. Thinking is great. The problem is that most people who are questioning aren't really thinking. They may be posing questions, but they are not interested in answers.

I think most CT'ers are "crazy" not because of what they believe, but because of the way they arrive at their beliefs.

[Q24]

In fact, the only people I here claiming that the fire was the sole cause of the collapse is the CT crowd when they try to attribute it to NIST.

NIST created extremely detailed computer models of the Twin Towers. When the best estimates were input for all variables including the impacts, buildings themselves and ensuing fires, the simulations showed no collapse. The case that allowed NIST to demonstrate collapse initiation had all variables set to their maximum possible severity. In doing this, the simulation still showed what was described as a “good” match with the observed damage but admitted to exceeding the actual impact damage seen present in photographic and video evidence. The conclusion is that not even one of the Twin Towers should have been expected to initiate collapse on 9/11 due to the conditions witnessed, let alone both of them.

[shaka5]

Yeah, they all start near the top of the building, and expose the explosive set-up to a big fire first.

huh?

[shaka5]

Too bad you have such limited experience. Implosions are the least frequent demolition type, and the least preferred by demolitionists.

I SAID IT "LOOKED" LIKE...BUT LIKE YOU SAID, YOU STATED THAT I HAVE LIMITED EXPERIENCE ON IT SO I CAN ONLY GO BY WHAT I SEE

Even at that, I find it amazing that you can look at a video of a genuine implosion, and one of the towers, and conclude that, because both buildings are falling in the same general direction, they are identical. You aren't much for detail.

UMM, OK

No, pretty much everyone saw it. The difference is that not everyone is under the impression that the fire department listens to civilians regarding explosive demolitions. And fewer people bothered to ask what "Pull it" means to a firefighter (It means to withdraw and let the building burn down), as opposed to what "Pull it" means to a demolitionist (it means attach cables to the upper levels and pull the building down by force; incidentally, this is the most common form of demolition).

YEA, BUT 1ST IT WAS JUST BECAUSE OF FIRE...BUT IF IT LOOKED LIKE A DEMOLITION, AND YOU HEAR A DEMOLITION TERM, WHAT ELSE ARE YOU SUPPOSE TO THINK

So does pretty much anyone who has studied the collapse. In fact, the only people I here claiming that the fire was the sole cause of the collapse is the CT crowd when they try to attribute it to NIST.

SORRY, NOT EVEN GONNA UNDERSTAND WHAT "NIST" STANDS FOR...BUT I'VE BEEN ON MANY SITES, I NEVER BOUGHT INTO THE FIRE THING

Yes, most of the fuel went up in the explosion. That still left plenty to explode in the floors inside the building, and even some to go down the elevator shaft and explode there too.

AGREE WITH THAT, DIDN'T MEAN TO IMPLY ALL THE FUEL WAS GONE, DON'T THINK I DID THO

And yes, there were firefighters who were exposed to parts of the fire that they were able to navigate in. That happens in fires. Some parts are not as bad as others.

NO, THE 1 SPOT OF THE BUILDING WHERE THEY SAID TO BE THE WORST WAS ACTUALLY NOT, I'LL TRY TO FIND ALL THE INFORMATION FOR YOU, HE'S ON RADIO STATING THAT ITS JUST A SMALL FIRE...WOULD DO MY BEST TO GET IT FOR YA

You are saying two things here. Yes, people heard explosions. Things blow up in fires. That does not mean that they heard bombs.

NO, I'M SAYING 1 THING, THE PEOPLE SAID IN CLEAR WORDS THAT THEY HEARD BOMBS GOING OFF, NOT COMPUTERS OR MICROWAVES EXPLODING

Not exactly. Questioning is wonderful. Thinking is great. The problem is that most people who are questioning aren't really thinking. They may be posing questions, but they are not interested in answers.

I AGREE WITH YOU COMPLETELY, BUT YOU MUST ADMIT THAT THEY'RE SKEPTICS THAT DO THE SAME, THEY LOOK FOR THE MOST OBVIOUS EXPLANATION AND JUST LEAVE IT AS THAT

I think most CT'ers are "crazy" not because of what they believe, but because of the way they arrive at their beliefs.

BUT TRUST ME, I'M NOT STATING ANYTHING AS FACT, JUST THINGS THAT JUST DOESN'T ADD UP TO ME

[Q24]

huh?

He said, “Yeah, they all start near the top of the building, and expose the explosive set-up to a big fire first.” In other words: the WTC demolitions appeared to initiate from the impact points high-up the buildings where the fires were. And that means: the Twin Towers were not taken down as conventional demolitions which initiate from the bottom.

There are features of controlled demolition in the WTC collapses – sudden initiation, virtually symmetrical, near freefall, complete collapses with visible explosive squibs (“smoke plumes” for aquatus’ benefit) far below the collapse zone. The difference is that they were necessarily unconventional and so would not appear exactly as other building demolitions.

[frenat]

No, pretty much everyone saw it. The difference is that not everyone is under the impression that the fire department listens to civilians regarding explosive demolitions. And fewer people bothered to ask what "Pull it" means to a firefighter (It means to withdraw and let the building burn down), as opposed to what "Pull it" means to a demolitionist (it means attach cables to the upper levels and pull the building down by force; incidentally, this is the most common form of demolition).

Also of note is that "Pull it" to a demolitionist does not mean anything to do with explosives. A demolitionist would likely say "Shoot it."

Note the red lined box in the lower left corner here

http://www.thestateonline.com/news/pdfs/implosion.pdf

"Pull" is only used if they bring something down mechanically (like they did with WTC 6) such as with a wrecking ball or pulling a chimney or tower over onto its side. Even then many companies speak of it as a "felling".

Check this demo companies list of projects. Do you see a pull anywhere? How about a felling, or a shot?

http://www.dykon-blasting.com/History/DemoJobList.htm

Even this kids DVD review mentions that buildings are "shot".

http://www.digitallyobsessed.com/showreview.php3?ID=395

with this quote, "they set a new world record for most buildings "shot" (industry term for bringing down a building or structure with explosives) at one time."

[aquatus1]

NIST created extremely detailed computer models of the Twin Towers. When the best estimates were input for all variables including the impacts, buildings themselves and ensuing fires, the simulations showed no collapse. The case that allowed NIST to demonstrate collapse initiation had all variables set to their maximum possible severity. In doing this, the simulation still showed what was described as a “good” match with the observed damage but admitted to exceeding the actual impact damage seen present in photographic and video evidence. The conclusion is that not even one of the Twin Towers should have been expected to initiate collapse on 9/11 due to the conditions witnessed, let alone both of them.

I've heard this argument before. Do you have a source that goes into more detail one it? I'd like to know where the NIST does this test, and what their conclusions were.

[Q24]

I've heard this argument before. Do you have a source that goes into more detail one it? I'd like to know where the NIST does this test, and what their conclusions were.

Obviously the NIST WTC investigation goes into full detail and this is the only source I am aware of. The following are the relevant sections: -

• NIST NCSTAR 1-2: Baseline Structural Performance and Aircraft Impact Damage Analysis of the World Trade Center Towers
  
• NIST NCSTAR 1-5: Reconstruction of the Fires in the World Trade Center Towers
  
• NIST NCSTAR 1-6: Structural Fire Response and Probable Collapse Sequence of the World Trade Center Towers
  

NIST’s conclusion as we all know is that the impact damage and fires caused the destruction of the Twin Towers – they pretty much make clear that this is going to be their conclusion from the beginning. It is what is revealed along the way and how exactly they manipulate the simulations to achieve their conclusion that is of interest – the aircraft was made faster and stronger, the buildings were made weaker, the fire temperatures were turned up. This exceeded the damage seen in visual evidence and caused collapse in the simulations. The case that used all of the best estimates and matched the actual damage situation most closely did not cause collapse in the simulations.

Flyingswan would say this is not a problem as the actual damage situation has been bracketed by NIST’s base and severe case. I say this is a problem as it does not prove that the actual damage situation on 9/11 would cause collapse initiation. Even more frustrating is that NIST could have definitively proved the case one way or another if they had run intermediary cases after having already done the hard work of creating the models.

[flyingswan]

Flyingswan would say this is not a problem as the actual damage situation has been bracketed by NIST’s base and severe case. I say this is a problem as it does not prove that the actual damage situation on 9/11 would cause collapse initiation. Even more frustrating is that NIST could have definitively proved the case one way or another if they had run intermediary cases after having already done the hard work of creating the models.

NIST also ran a less severe case. All three cases were consistent with the known parameters of the actual impact, which were subject to some measurement uncertainty. It was comparison of the three cases with the observed impact damage that suggested that the actual impact was intermediate between the baseline and severe cases. In other words, the baseline case was just as much an underestimate of actuality as the severe case was an overestimate.

Another point is that the way in which the collapses initiated matched the NIST models. The visible sagging of the floors and bowing of the outer walls is something that is easily explanable as a result of fire, but rather tricky to explain in terms of explosives.

[flyingswan]

huh?

Just pointing out that there were in fact major differences between the tower collapses and standard demolitions.

Perhaps since you were a witness, you could give us all a detailed description of what you saw and heard.

[flyingswan]

There are features of controlled demolition in the WTC collapses – sudden initiation, virtually symmetrical, near freefall, complete collapses with visible explosive squibs (“smoke plumes” for aquatus’ benefit) far below the collapse zone.

Every one has seen demolition films, but unexpected structural collapses don't get filmed, all you see is the debris. The initiations of the WTC collapses were not sudden, the walls visibly bowed beforehand. There is nothing about the speed or symmetry of the collapses that is unique to a demolition rather than a structural failure. The so-called "squibs" are easily explained as air compressed by the collapse finding weak points to blow out. They occur after the collapse starts, rather than before as they would if they were explosives.

The difference is that they were necessarily unconventional and so would not appear exactly as other building demolitions.

The all-purpose conspiracy explanation. Anything about the collapses that resembles a demolition is evidence for an inside job, but anything that doesn't resemble a demolition is evidence that the inside job was "necessarily unconventional". It's a demolition if it looks like a demolition, and it's still a demolition if it doesn't.

[Q24]

Flyingswan, I think we have been through near every general argument on the WTC buildings that it is possible to have. I know your views well, as you also know mine, so I will not address every point you raise to avoid going over a lot of old ground. There was just one suggestion you made that did concern me: -

In other words, the baseline case was just as much an underestimate of actuality as the severe case was an overestimate.

You claim the base case to be “just as much” from the observable evidence as the severe case was. You must know this is not true as we have been over it many times. It isn’t even something that’s up for debate because it’s printed in black and white in the NIST investigation NCSTAR1-2.

Discussing comparison between the base case and severe case for WTC1: -

“The overall agreement with the observed damage to the north wall was good for the base case and the more severe case, with the base case analysis providing better match to the observed damage.”

Discussing the same for WTC2: -

“As was the case for WTC 1, there were small differences in the damage estimates for the south wall of WTC 2 from the base case and the more severe case scenarios. Overall, the agreement with the observed damage from photographs was very good for both cases.”

In the WTC1 comparison, although agreement with both cases was “good”, it was the base case that provided a “better match” to the actual damage observed in visual evidence. The WTC2 comparison is not so specific in which case is the best match though the introduction “As was the case for WTC 1…” obviously suggests it is the same situation as previous.

Anyone interested should take a look at all of the variables that were moved away from the best estimates to achieve collapse initiation in the simulations. The building models alone had their failure strain reduced to 80% of that which was expected and combined with other similar manipulations the severe case was simply not a realistic representation of events on 9/11.

[flyingswan]

You claim the base case to be “just as much” from the observable evidence as the severe case was. You must know this is not true as we have been over it many times.

I've no more desire to reanimate our discussion than you. However, if you want to just limit things to the one point:

You are ignoring the less severe case, which proved to be a worse match than the other two. With three cases rated in that way, the actual result must lie between the two cases that are the best matches, somewhere near the middle. That makes the baseline case an underestimate and the severe case an overestimate. You read too much into my post when you say I claim the actual case was exactly in the middle.

I read “As was the case for WTC 1…” as saying that there were again small differences in predicted damage, but I think you again read too much when you interpret that as meaning the base case was better.

Edited September 5, 2009 by flyingswan

[Q24]

I've no more desire to reanimate our discussion than you. However, if you want to just limit things to the one point:

Ok.

Of course none of this actually affects the main body of my post #103: -

• Base case (expected) equals no collapse initiation
  
• Severe case equals collapse initiation
  

I know you would say that the severe case uses ‘possible’ parameters, but they were still beyond what were found to be the best estimates using all available evidence. Seriously, using the NIST method and adjusting some variables I could prove that I would beat Usain Bolt in a sprint. And whilst that would be technically possible, it’s never going to happen and would be completely unexpected if it did.

I would be interested to hear aquatus’ view.

[flyingswan]

Of course none of this actually affects the main body of my post #103: -

• Base case (expected) equals no collapse initiation
  
• Severe case equals collapse initiation
  

I know you would say that the severe case uses ‘possible’ parameters, but they were still beyond what were found to be the best estimates using all available evidence.

But after the comparison with the actual impact damage, we have base case and severe case giving similarly good results, the base case a bit better for WTC1, both about the same for WTC2. That means that your characterisation of the severe case as well beyond anything that would be expected is incorrect, the actual impact departure from the base case was more severe by about half of the maximum expected when picking parameters for the severe case.

I can't see your point about NIST needing to run more cases. Say they did a few more cases to fill the gap between the base and severe cases. We would then have a better handle on what parameters would result in collapse initiation, but you still wouldn't know where the actual impact parameters were. What you would have is somewhere in the range a new case with no collapse and next up a more severe new case with collapse, and you would be arguing for more cases in the gap between these. It's like creationists asking science for "missing links". When science finds one, the creationists just ask for something in the gaps between the new discovery and the old gap markers.

Given the small changes in the matches to the impact damage with the existing base and severe cases, there is no reason to think that the intermediate cases would give a noticably better match. For each tower, NIST have a case with collapse initiation, it matches the impact damage and it matches the observable evidence for the way that the collapses actually started, what more can they reasonably do?

The building models alone had their failure strain reduced to 80% of that which was expected

Just picked this up from your earlier post. Why do you think a 20% uncertainty in failure strain is unreasonable?

Edited September 6, 2009 by flyingswan

[Q24]

That means that your characterisation of the severe case as well beyond anything that would be expected is incorrect…

I’m sure you mean to say “possible” rather than “expected” - whilst the severe case inputs were set to their maximum possible limits they certainly were not expected. As is made very clear in NCSTAR1-2, the base case was expected whilst the severe case was beyond that expected.

For each tower, NIST have a case with collapse initiation, it matches the impact damage and it matches the observable evidence for the way that the collapses actually started, what more can they reasonably do?

You know what NIST needed to do to prove their case, you are the one who first brought it up in our discussion - the crossover point in the simulations from ‘no collapse’ to ‘collapse’ had to be established. You actually said that this is what I needed to ascertain to prove the buildings should not have initiated collapse and yet ironically this is also the same point NIST needed to determine to prove they should have initiated collapse (and they are the ones provided a multi-million dollar budget to carry out the investigation after all). By finding the last case at the end of the ‘no collapse’ range and the first case at the beginning of the ‘collapse’ range NIST could have concluded which situation was the best match to the actual damage seen in visual evidence.

As it stands, the last simulation prior to collapse is the base case and the next simulation after collapse is the severe case. NIST have stated the actual damage seen in visual evidence, certainly in the case of WTC1, is closer to the base case. If any conclusion is to be drawn from this, it is that the simulation tells us the building should not have initiated collapsed.

The very least that can be said is that NIST did not prove the impacts and fires would cause collapse initiation and when I have previously asked the question, “Did NIST simulate the reality of the situation on 9/11?” the answer even you gave was, “No”. Further than this involves speculation and I don’t see why anyone should find NIST’s indefinite investigation to be acceptable.

Edit: -

Just picked this up from your earlier post. Why do you think a 20% uncertainty in failure strain is unreasonable?

I do not think a 20% uncertainty in failure strain is “unreasonable” per se, but it is “unexpected” in relation to the building design properties and combined with the other variables maximised to bring about collapse initiation in the simulations culminates in NIST’s severe case being entirely beyond what was expected. The fact that the actual damage was a better match to the base case, certainly in the case of WTC1, proves this conclusively.

Edited September 6, 2009 by Q24

[aquatus1]

Q24, I would like to ask you what should be a "yes" or "No" question.

Do you believe that failure models, such as the ones used by NIST for the WTC towers collapse, represent a mathematical reality, in which given a set of variables, the result will always be the same?

[Q24]

Q24, I would like to ask you what should be a "yes" or "No" question.

Do you believe that failure models, such as the ones used by NIST for the WTC towers collapse, represent a mathematical reality, in which given a set of variables, the result will always be the same?

All variables were fixed inputs by NIST specific to each case they ran and with the computer simulations being based on physics this should produce the same result each time for any given set of values. The answer to your question in regard to NIST’s methodology in the WTC cases is then “yes”.

I understand that some computer simulations will set the possible parameters for a given situation and let the variables be randomly generated within them to give a range of possible results over a number of runs. The answer to your question in regard to a simulation of this type is “no”.

It would have been extremely interesting if NIST had used the second method, ie let the computer randomly generate the variables within each parameter, running the simulation a number of times and recording the percentage of occasions ‘no collapse’/‘collapse’ initiation was produced. Obviously NIST did not do this and here is a possible reason why – it would hardly be supportive of their conclusion if the models determined that 99% of the time the buildings did not collapse!

Flyingswan would tell you that NIST didn’t carry out further runs like this because of money and/or time constraints. I do not find this acceptable as NIST were provided with a $23.4million budget exactly to provide these sorts of answers. No, I’m being unfair – this was never honestly a political quagmire that NIST were willing to traverse in the first place; only one answer was acceptable.

[aquatus1]

So, in summary, "yes".

Thank you for answering.

[flyingswan]

I’m sure you mean to say “possible” rather than “expected” - whilst the severe case inputs were set to their maximum possible limits they certainly were not expected. As is made very clear in NCSTAR1-2, the base case was expected whilst the severe case was beyond that expected.

Oh dear, we are back to the old familiar situation. You ignore my main points and nit-pick about minor things like this. The base case used the best estimate of a number of impact parameters, but these parameters all had ranges of possibility. Anywhere in that range would be "possible", all that would be "expected" is that the actual impact would be contained in that range. To claim that only the base case is "expected" is like claiming that every 10 coin tosses will give 5 heads and 5 tails.

You know what NIST needed to do to prove their case, you are the one who first brought it up in our discussion - the crossover point in the simulations from ‘no collapse’ to ‘collapse’ had to be established. You actually said that this is what I needed to ascertain to prove the buildings should not have initiated collapse and yet ironically this is also the same point NIST needed to determine to prove they should have initiated collapse (and they are the ones provided a multi-million dollar budget to carry out the investigation after all). By finding the last case at the end of the ‘no collapse’ range and the first case at the beginning of the ‘collapse’ range NIST could have concluded which situation was the best match to the actual damage seen in visual evidence.

As it stands, the last simulation prior to collapse is the base case and the next simulation after collapse is the severe case. NIST have stated the actual damage seen in visual evidence, certainly in the case of WTC1, is closer to the base case. If any conclusion is to be drawn from this, it is that the simulation tells us the building should not have initiated collapsed.

The very least that can be said is that NIST did not prove the impacts and fires would cause collapse initiation and when I have previously asked the question, “Did NIST simulate the reality of the situation on 9/11?” the answer even you gave was, “No”. Further than this involves speculation and I don’t see why anyone should find NIST’s indefinite investigation to be acceptable.

I will just repeat the points from my last post which you have ignored:

Say they did a few more cases to fill the gap between the base and severe cases. We would then have a better handle on what parameters would result in collapse initiation, but you still wouldn't know where the actual impact parameters were.

Given the small changes in the matches to the impact damage with the existing base and severe cases, there is no reason to think that the intermediate cases would give a noticably better match.

I do not think a 20% uncertainty in failure strain is “unreasonable” per se, but it is “unexpected” in relation to the building design properties...

I have no idea what you mean by this. What point are you making about the failure strain numbers if you don't think it unreasonable for there to be a 20% uncertainty?

[Q24]

The base case used the best estimate of a number of impact parameters, but these parameters all had ranges of possibility. Anywhere in that range would be "possible", all that would be "expected" is that the actual impact would be contained in that range. To claim that only the base case is "expected" is like claiming that every 10 coin tosses will give 5 heads and 5 tails.

I like this example – it supports exactly that some areas in the “possible” range are “expected” whilst some are “unexpected”.

If we toss a coin 10 times then the most expected result would indeed be 5 heads and 5 tails – this is exactly equivalent of the way NIST input variables to the base case. If we toss a coin 10 times then the most unexpected results would be 10 heads or 10 tails – this is exactly equivalent of the way NIST input variables to the less severe and severe cases.

If we toss 10 coins, here are the chances of achieving a given number of heads: -

0 heads =

0.10%

(equivalent to less severe case)

1 head = 0.98%

2 heads = 4.39%

3 heads = 11.72%

4 heads = 20.51%

5 heads =

24.61%

(equivalent to base case)

6 heads = 20.51%

7 heads = 11.72%

8 heads = 4.39%

9 heads = 0.98%

10 heads =

0.10%

(equivalent to severe case)

In summary: the base case is the most expected outcome, the cases adjacent are still highly expected but the further we depart from that the less expected the situation becomes until we reach the less severe and severe cases at the extreme ends as the most unexpected of outcomes. Of course the variables in NIST’s computer simulations were far more complex than the toss of a coin which can only give one of two results. This would in fact mean that the extreme cases had a lower probability of occurrence than even the example above.

And remember: -

• Base case (expected) equals no collapse initiation
  
• Severe case (unexpected) equals collapse initiation
  

I will just repeat the points from my last post which you have ignored:

Say they did a few more cases to fill the gap between the base and severe cases. We would then have a better handle on what parameters would result in collapse initiation, but you still wouldn't know where the actual impact parameters were.

Given the small changes in the matches to the impact damage with the existing base and severe cases, there is no reason to think that the intermediate cases would give a noticably better match.

As I said, NIST could have carried out a detailed comparison of the visual evidence against intermediary simulations to determine which side of the ‘no collapse’/‘collapse’ crossover point the actual damage fell. You speculate changes in the simulated damage would perhaps be too small to observe. I disagree – the models were detailed enough to do this if sufficient effort had been given to the task.

The least NIST should have done to support their conclusion is provide a simulated case that was the best match to the actual damage and caused collapse initiation. What NIST actually provided was a best match to the actual damage that did not cause collapse initiation! It’s madness – they need to change their conclusion to state that the collapses were unexpected due to the impacts/fires and renew the investigation to explore other initiators.

I have no idea what you mean by this. What point are you making about the failure strain numbers if you don't think it unreasonable for there to be a 20% uncertainty?

To oversimplify in the hope of explaining, let’s go back to the coins. It is not unreasonable to expect any given coin to provide a head. It is more unreasonable to expect 10 coins to provide 10 heads. Even putting aside that a fifth weakening of the buildings from the most expected case appears to be rather significant, it is combining with the other variables that the case becomes altogether unexpected.

And finally, when I am talking about unexpected situations here, that is for a single building to collapse and yet… it happened three times at one site, on one day.

Unbelievable.

[flyingswan]

I like this example – it supports exactly that some areas in the “possible” range are “expected” whilst some are “unexpected”.

If we toss a coin 10 times then the most expected result would indeed be 5 heads and 5 tails – this is exactly equivalent of the way NIST input variables to the base case. If we toss a coin 10 times then the most unexpected results would be 10 heads or 10 tails – this is exactly equivalent of the way NIST input variables to the less severe and severe cases.

If we toss 10 coins, here are the chances of achieving a given number of heads: -

0 heads =

0.10%

(equivalent to less severe case)

1 head = 0.98%

2 heads = 4.39%

3 heads = 11.72%

4 heads = 20.51%

5 heads =

24.61%

(equivalent to base case)

6 heads = 20.51%

7 heads = 11.72%

8 heads = 4.39%

9 heads = 0.98%

10 heads =

0.10%

(equivalent to severe case)

In summary: the base case is the most expected outcome, the cases adjacent are still highly expected but the further we depart from that the less expected the situation becomes until we reach the less severe and severe cases at the extreme ends as the most unexpected of outcomes. Of course the variables in NIST’s computer simulations were far more complex than the toss of a coin which can only give one of two results. This would in fact mean that the extreme cases had a lower probability of occurrence than even the example above.

And remember: -

• Base case (expected) equals no collapse initiation
  
• Severe case (unexpected) equals collapse initiation
  

From your own calcs, 5-5 only happens a quarter of the time, so you actually expect something else.

As I said, NIST could have carried out a detailed comparison of the visual evidence against intermediary simulations to determine which side of the ‘no collapse’/‘collapse’ crossover point the actual damage fell. You speculate changes in the simulated damage would perhaps be too small to observe. I disagree – the models were detailed enough to do this if sufficient effort had been given to the task.

The models may be detailed, but most of the actual damage isn't easy to assess. There could be big differences in interior damage for similar exterior damage, and only the latter is visible.

The least NIST should have done to support their conclusion is provide a simulated case that was the best match to the actual damage and caused collapse initiation. What NIST actually provided was a best match to the actual damage that did not cause collapse initiation! It’s madness – they need to change their conclusion to state that the collapses were unexpected due to the impacts/fires and renew the investigation to explore other initiators.

There you go again, ignoring the fact that the two cases gave similar matches to the actual damage, getting identical "very good" ratings for WTC2, and also ignoring the match between the predicted and observed collapse initiation, floors sagging and walls bowing.

To oversimplify in the hope of explaining, let’s go back to the coins. It is not unreasonable to expect any given coin to provide a head. It is more unreasonable to expect 10 coins to provide 10 heads. Even putting aside that a fifth weakening of the buildings from the most expected case appears to be rather significant, it is combining with the other variables that the case becomes altogether unexpected.

That wasn't the point you were making initially. You specifically picked out the 20% uncertainty in failure strain, as if it was somehow unreasonable, then when I queried it you back-pedalled. Now you are equating it with "a fifth weakening of the buildings". Do you even know what failure strain is?

And finally, when I am talking about unexpected situations here, that is for a single building to collapse and yet… it happened three times at one site, on one day.

Unbelievable.

It might be unbelievable if two airliners hadn't been crashed into them. Seeing that two of the buildings were of the same construction and suffered very similar damage and fires, the fact that they collapsed in similar ways is hardly surprising. The third was of a different construction and collapsed in a different way after a long and unfought fire.

Hardly surprising.

[Q24]

From your own calcs, 5-5 only happens a quarter of the time, so you actually expect something else.

Yes, combining a range of results against the single 5 heads case means we would expect an unspecified ‘something else’. The problem is that the actual event on 9/11 was not a range but a single outcome, and the single most expected outcome in our example is 5 heads. If you want to look at the most expected range then this would be 4-6 heads at 66%, ie still close to what would be the equivalent of the base case.

The models may be detailed, but most of the actual damage isn't easy to assess. There could be big differences in interior damage for similar exterior damage, and only the latter is visible.

Correct, we know that small exterior changes in the damage carried through the building causing larger internal changes in the damage – for example, in going from base case to severe case double the number of core columns were severed in the simulations. It doesn’t matter that only the external damage is visible in photographic evidence because if NIST’s models are accurate and got a match externally then the internal damage should be correctly simulated to match the actual damage also.

There you go again, ignoring the fact that the two cases gave similar matches to the actual damage, getting identical "very good" ratings for WTC2, and also ignoring the match between the predicted and observed collapse initiation, floors sagging and walls bowing.

You earlier said you wanted to stick to one point. If so, please drop the nonsense about the models predicting the bowing in the walls. We have discussed more than once that this was an input by NIST rather than a prediction of the models in this instance – I don’t understand why you keep repeating it. Would you like me to quote the relevant parts of NIST’s investigation again?

That wasn't the point you were making initially. You specifically picked out the 20% uncertainty in failure strain, as if it was somehow unreasonable, then when I queried it you back-pedalled. Now you are equating it with "a fifth weakening of the buildings". Do you even know what failure strain is?

The initial point was just to give one example of a parameter that NIST optimised to cause collapse in the severe case simulation. Even if not entirely unreasonable, I find a reduction of the building strength by 20% from the most expected case to be significant. The term “failure strain” is self-explanatory – the strain a material can withstand before failure.

It might be unbelievable if two airliners hadn't been crashed into them.

NIST accounted for the airliner impacts (I’m not sure how you missed that) and still demonstrated the collapses to be unexpected.

[flyingswan]

Yes, combining a range of results against the single 5 heads case means we would expect an unspecified ‘something else’. The problem is that the actual event on 9/11 was not a range but a single outcome, and the single most expected outcome in our example is 5 heads. If you want to look at the most expected range then this would be 4-6 heads at 66%, ie still close to what would be the equivalent of the base case.

Your claim that only the base case was "expected" is similar to claiming that only 5-5 is "expected". In fact, three times out of four you get something else. On your analogy, a more severe case of some sort is more probable that baseline, for instance.

Your actual probabilities are not relevant. Just because I picked ten tosses, you claim that the severe case is as probable as ten heads. If I'd said six tosses or twenty tosses you'd have come up with different numbers.

Correct, we know that small exterior changes in the damage carried through the building causing larger internal changes in the damage – for example, in going from base case to severe case double the number of core columns were severed in the simulations. It doesn’t matter that only the external damage is visible in photographic evidence because if NIST’s models are accurate and got a match externally then the internal damage should be correctly simulated to match the actual damage also.

But if both the baseline and severe cases give very good matches to the external damage, and you can't see the internal damage, how do you assess which case is better?

You earlier said you wanted to stick to one point. If so, please drop the nonsense about the models predicting the bowing in the walls. We have discussed more than once that this was an input by NIST rather than a prediction of the models in this instance – I don’t understand why you keep repeating it. Would you like me to quote the relevant parts of NIST’s investigation again?

Would you like me to explain what a calibration test is again?

The bowing was indeed observed, but it was also predicted as a consequence of the sagging floors pulling the walls inwards. This is a combination of prediction and observation that none of the demolition theories can match.

I agree to drop this, anyone interested can look up the old thread. I'd hate to distract you from your debates with Scott G, which I find immensely entertaining. You've found someone even better at confirmation bias than you are, and as for you getting banned from those forums...priceless.

The initial point was just to give one example of a parameter that NIST optimised to cause collapse in the severe case simulation. Even if not entirely unreasonable, I find a reduction of the building strength by 20% from the most expected case to be significant. The term “failure strain” is self-explanatory – the strain a material can withstand before failure.

And "strain" in an engineering sense means?

Hint, it doesn't mean strength, as you again seem to think.

NIST accounted for the airliner impacts (I’m not sure how you missed that) and still demonstrated the collapses to be unexpected.

Only on your personal definition of "unexpected".

Edited September 9, 2009 by flyingswan