Merged Discussion of femr's video data analysis

That is funny.
Yes it is rather funnay that several members choose to utilise their time "investigating" the origins of a graph produced about 2 years ago. Bloomin' hilarious really.

Wrong filter
So you've said, many times. Yet you seem utterly incapable of suggesting why, nor what other filters you would use yourself. You have been saying the same thing for about two years. Any progress ? Any goal ?

Perhaps you'd care to address the use of Savitzky-Golay by W.D.Clinger and pgimeno too ?

funny curves
Er, riiiight. Good job beachnut ! Most impressive. Amazing.
 
Yes it is rather funnay that several members choose to utilise their time "investigating" the origins of a graph produced about 2 years ago. Bloomin' hilarious really...
Actually I find it sad. You probably meant to include that aspect also.

There are two bleedingly obvious questions jumping off the pages here:
1) Do femr's methods have errors? AND
2) Are those errors big enough to matter?

The first is a truism - any measurement and data derived from measurement has errors.

So the second one is the question which matters - split into two parts it becomes: "How big are the possible errors?" AND "Are the errors big enough to matter for whatever purpose the data is to be used?"

And nobody wants to venture into discussion of that second one despite my several not too subtle hints.

So why are people discussing in this thread?

A) To improve their understanding of the mathematical numeric methods? If so why pretend it is about femr allegedly being wrong? Or about use of the data?

B) A genuine concern that the results may not be accurate enough for a purpose? If so why not (i)nominate a purpose or range of purposes, (ii) quantify the range of errors AND test if "(ii)" is too big for valid use in "(i)" OR

C) Is it simply "prove femr wrong - the umpteenth attempt"?

Where falling flat on your face doesn't get included in the score? And the fact that this specific bit of data was provided in response to members requests? And that those members can make use of the data?

So I suppose femr is right with this response:
That is funny.
Bloomin' hilarious really.

Talk about "Forests versus trees" maybe its "forests versus (the leaves on the 43rd twig on branch xyz of.....)"

:boggled:
 
Last edited:
There are two bleedingly obvious questions jumping off the pages here:
1) Do femr's methods have errors? AND
2) Are those errors big enough to matter?

The first is a truism - any measurement and data derived from measurement has errors.

So the second one is the question which matters - split into two parts it becomes: "How big are the possible errors?" AND "Are the errors big enough to matter for whatever purpose the data is to be used?"

And nobody wants to venture into discussion of that second one despite my several not too subtle hints.

I can't really see a point to a discussion unless someone wants to propose a purpose other than 'just to see'. The analysis has obviously surpassed the quality of the input data to within any useful margin of error.

From an understanding why the building collapsed point of view, anything after the first signs of movement of the east penthouse is really of limited to no use. NIST took the "truther" bait when they started trying to explain the speed of the collapse. This added nothing of value to the report and frankly was a waste of time. This "fine" data is more or less the same thing. The collapse report would not have changed at all with this addition. NIST didn't "blow it", they used methods that was good enough for their purpose.

I know I will get slammed by femr2 for this view but, for him, have at it, keep up the excellent work. :)
 
I can't really see a point to a discussion unless someone wants to propose a purpose other than 'just to see'....
I think you and I are on the same track. The point of measurements is to measure things for some purpose. Not just for the sake of measuring. However the person who makes the ruler doesn't have to be the one who uses the ruler or the one who does the measurements.

I possess micrometers, vernier callipers, steel rulers up to one metre long, a couple of one meter carpenters folding rulers (the ones that fold four ways and used to be one yard/three feet) plus several tape measures up to 15 metres long, one ultrasonic distance measurer and one laser distance measure.

Main points are:
1) I didn't make them but I use them;
2) The people who made them neither need to use them for my purposes NOR have any obligation or right to tell me how I use them etc etc.

Take it one step further as an engineer I have often had a surveyor use "tools of trade" to get measurements for my projects. I then use and rely on his measurements.

That is the situation we have here - femr enjoys the speciality of measuring things related to 9/11 WTC collapses and obviously does a good job and an accurate job of it. Others of us use the data. So far so good.

Then we see another legitimate objective. If I understand him correctly pgimeno wants to verify for himself one way or another that femr's methods are right or wrong and he wants to do it from the specific focus of the mathematical modelling. He is entitled to do that provided he doesn't make claims outside the area of his reasoning. I am more than willing to look at femr's accuracy from several other methods which are available other than criticising his maths. If the results are either "right" or "good enough for purpose" (those may not be the same) I will validly accept that the maths are not introducing big errors. For me then whether the maths are theoretically perfect matters not - I can use the results.

Then, of course, the needed accuracy depends on the purpose. I don't use the micrometer when carpentering concrete formwork. I once met a fellow who did cabinet making woodwork to metalworkers precision but that is another story. He was very good. :) (BTW I didn't tell him he was wrong which would happen if he was posting here :boxedin: )

The analysis has obviously surpassed the quality of the input data to within any useful margin of error.
Quite likely.
From an understanding why the building collapsed point of view, anything after the first signs of movement of the east penthouse is really of limited to no use....
True within that limited purpose. BUT remember that several members were interested is truther claims related to "Acceleration near 'G' always means CD" and a consequential subset of "Momentary accelerations in excess of 'G'" - the latter something that some of the debunkers did not believe. So femr's data was of use to members whose legitimate for them needs went further than "collapse initiation and EPH falling".

Also don't miss the significance of the fact which femr and I have noted many times - that early movement was shown for significant time before the release to rapid collapse. Both femr and I have not stated explicitly why that fact is significant. Something about moving horses into the proximity of water without being able to force imbibing. If folk cannot work it out for themselves......:boggled:

NIST took the "truther" bait when they started trying to explain the speed of the collapse. This added nothing of value to the report and frankly was a waste of time.
Yes - I think they would have been better not responding to Chandler's trickery. There was no way he would be honest with the findings. But it was their call - some of us forumites who are used to truther mendacity may have taken a different line. But I doubt that the majority would have resisted the temptation - just look at the massive amount of troll feeding that occupies these threads. Was NIST responding to Chandler any different than debunkers feeding trolls?

This "fine" data is more or less the same thing. The collapse report would not have changed at all with this addition. NIST didn't "blow it", they used methods that was good enough for their purpose.
It is only "more or less the same thing" if you limit to your own need for information. Others have different needs. Some I agree with, some I disagree with but they are mostly legitimate needs for the purposes of those wanting the info.

Yes NIST was "good enough for their purpose".

...I know I will get slammed by femr2 for this view but, for him, have at it, keep up the excellent work. :)
I doubt it provided you don't make him think that you are trying to impose your standards and your purposes on him. Just look at the repeated nonsense from several members who want to force femr to adopt their own goals, in Beachnut's case to adopt not only his goal but his limited military style definition of goal. Or to comment on things such as CD when that is not the topic under discussion. femr has always (AFAICS) been clear that his objective is "measure to the best of his ability" or similar words. AND what other people want to use data for is their call not his.
 
Last edited:
I think you and I are on the same track. The point of measurements is to measure things for some purpose. Not just for the sake of measuring. However the person who makes the ruler doesn't have to be the one who uses the ruler or the one who does the measurements.

I possess micrometers, vernier callipers, steel rulers up to one metre long, a couple of one meter carpenters folding rulers (the ones that fold four ways and used to be one yard/three feet) plus several tape measures up to 15 metres long, one ultrasonic distance measurer and one laser distance measure.

Main points are:
1) I didn't make them but I use them;
2) The people who made them neither need to use them for my purposes NOR have any obligation or right to tell me how I use them etc etc.

Take it one step further as an engineer I have often had a surveyor use "tools of trade" to get measurements for my projects. I then use and rely on his measurements.

That is the situation we have here - femr enjoys the speciality of measuring things related to 9/11 WTC collapses and obviously does a good job and an accurate job of it. Others of us use the data. So far so good.

Then we see another legitimate objective. If I understand him correctly pgimeno wants to verify for himself one way or another that femr's methods are right or wrong and he wants to do it from the specific focus of the mathematical modelling. He is entitled to do that provided he doesn't make claims outside the area of his reasoning. I am more than willing to look at femr's accuracy from several other methods which are available other than criticising his maths. If the results are either "right" or "good enough for purpose" (those may not be the same) I will validly accept that the maths are not introducing big errors. For me then whether the maths are theoretically perfect matters not - I can use the results.

Then, of course, the needed accuracy depends on the purpose. I don't use the micrometer when carpentering concrete formwork. I once met a fellow who did cabinet making woodwork to metalworkers precision but that is another story. He was very good. :) (BTW I didn't tell him he was wrong which would happen if he was posting here :boxedin: )

Quite likely.
True within that limited purpose. BUT remember that several members were interested is truther claims related to "Acceleration near 'G' always means CD" and a consequential subset of "Momentary accelerations in excess of 'G'" - the latter something that some of the debunkers did not believe. So femr's data was of use to members whose legitimate for them needs went further than "collapse initiation and EPH falling".

Also don't miss the significance of the fact which femr and I have noted many times - that early movement was shown for significant time before the release to rapid collapse. Both femr and I have not stated explicitly why that fact is significant. Something about moving horses into the proximity of water without being able to force imbibing. If folk cannot work it out for themselves......:boggled:

Yes - I think they would have been better not responding to Chandler's trickery. There was no way he would be honest with the findings. But it was their call - some of us forumites who are used to truther mendacity may have taken a different line. But I doubt that the majority would have resisted the temptation - just look at the massive amount of troll feeding that occupies these threads. Was NIST responding to Chandler any different than debunkers feeding trolls?

It is only "more or less the same thing" if you limit to your own need for information. Others have different needs. Some I agree with, some I disagree with but they are mostly legitimate needs for the purposes of those wanting the info.

Yes NIST was "good enough for their purpose".


I doubt it provided you don't make him think that you are trying to impose your standards and your purposes on him. Just look at the repeated nonsense from several members who want to force femr to adopt their own goals, in Beachnut's case to adopt not only his goal but his limited military style definition of goal. Or to comment on things such as CD when that is not the topic under discussion. femr has always (AFAICS) been clear that his objective is "measure to the best of his ability" or similar words. AND what other people want to use data for is their call not his.

Why do you refuse to say why you think it's significant?
 
Why do you refuse to say why you think it's significant?
For the reasons I outlined in the two sentences following....

So far IIRC only two members have spotted the significance. They also kept the "secret". :) So it is a bit of a game.

HINT
There are only two main claims for why the WTC buildings collapsed. One of them is a bit shaky if the collapse built up slowly over minutes before the "big moment of release".

PM me if you want more.
 
Last edited:
Actually I find it sad. You probably meant to include that aspect also.
Indeed. Funnay :)

Do femr's methods have errors?
There is certainly noise in the data itself, which is unavoidable. However, given the superior method by which the data is extracted from video, each datapoint has demonstrably much LESS noise/error than both NIST and Chandler data.

Some members here choose to constantly question methods, without actually identifying errors.

I have withheld exact details of procedure for one acceleration plot, and the responses (over the last two YEARS) have been very enlightening about the purpose and intent of those commenting.

Given that I have provided full details for the 1-pass SG plot, which has been ignored, it is clear that focus is not really femr2 methods, but femr2.

Are those errors big enough to matter?
I suggest step 1 would be for those questioning methods to actually state what "errors" they are referring to. The silly assertions about "self consistency" are demonstrably wrong, and borne from such folk making assumptions about steps in plot generation...moral being "don't say it's wrong until you know what it is" ;) (that's just as bad as folk who think something looks like a duck calling it a duck)

any measurement and data derived from measurement has errors
Of course, though again, when compared to the amount of error contained in data extracted by manual eye-balling... :)
 
I can't really see a point to a discussion unless someone wants to propose a purpose other than 'just to see'.
What is actually being discussed ? Folk veered away from WTC7 itself long ago, and instead chose to look at femr2. There's been no "new" assertions (or analysis) for quite some time.

A few new graphs from WDC and pgimeno, but they change nothing...same general trend on a focus on behaviour during descent.

The analysis has obviously surpassed the quality of the input data to within any useful margin of error.
Could you be more specific ? My "analysis" (conclusion) has been conservative. No great bold assertions from me (unlike those from others such as tfk)

From an understanding why the building collapsed point of view, anything after the first signs of movement of the east penthouse is really of limited to no use.
Once the acceleration trend behaviour was revealed in reasonable detail, sure. Detail in both NIST and Chandler was severely lacking.

Whilst I have responded to discussion about acceleration behaviour I have made it clear long ago that my personal focus was on displacement data behaviour (early motion), which is much more "interesting" imo.

This "fine" data is more or less the same thing. The collapse report would not have changed at all with this addition.
They could have presented early motion. Hugely significant. They could have rejected "FFA for n seconds". They could have...

NIST didn't "blow it", they used methods that was good enough for their purpose.
I suggest you make such assertions in the appropriate thread, where you will find rather lengthy critiques from me showing the many issues and errors in the NIST analysis. Your statement above is particularly ill-advised.

I know I will get slammed by femr2 for this view but, for him, have at it, keep up the excellent work. :)
I suggest the poor part of your post is that last assertion "NIST didn't "blow it", they used methods that was good enough for their purpose.". That was a silly thing to say.
 
Indeed. Funnay :)
Yes.
There is certainly noise in the data itself, which is unavoidable.
I have been using "error" in the sense of "difference between the measurement" and" Reality" - recognising that even "reality" is hard to define. Also my deliberate simplification of "error" into "noise" and "procedural artefacts". I actually don't care much if someone would identify one or two others that my categories miss BUT the concept is too complex or too broad for those who want a narrow focus. Too dangerous to the stability of the "house of cards" criticisms built on mathematical theory without quantification PLUS all the selective reference to graphs which suit.
However, given the superior method by which the data is extracted from video, each datapoint has demonstrably much LESS noise/error than both NIST and Chandler data.
There is really no doubt about that...hence the avoidance of either quantifying the error range in your results OR doing comparison with NIST or Chandler. Hence my emphasis of that need for quantification as a basis for comparison...and the resultant reluctance to follow my challenges. It's not exactly a complicated concept. And my writing is not all that dense either. As pedantic for logic as I can make it without spending hours on each post but....:rolleyes:

Some members here choose to constantly question methods, without actually identifying errors.
Exactly the thrust of my recent posts. And why I have deliberately come at it from the opposite direction viz "If the results are accurate enough then the maths cannot be too far out. (That is a bit shortcut and a couple of holes in the logic I will plug if anyone gets game enough to join in discussion. e.g. I would have to deal with the "Right for the wrong reasons" issues but....)
Given that I have provided full details for the 1-pass SG plot, which has been ignored, it is clear that focus is not really femr2 methods, but femr2.
Likely but don't discount reasoning skill issues.

I suggest step 1 would be for those questioning methods to actually state what "errors" they are referring to.
Again my theme stated differently.

...Of course, though again, when compared to the amount of error contained in data extracted by manual eye-balling... :)
The total avoidance of any quantification or any basis for comparison is so evident.
 
What is actually being discussed ? Folk veered away from WTC7 itself long ago, and instead chose to look at femr2.
I think what is being discussed is video analysis and graph production.

A few new graphs from WDC and pgimeno, but they change nothing...same general trend on a focus on behaviour during descent.
Yup, who can make the nicest looking graph. I can give a useful trend looking with the naked eye. ;)

Could you be more specific ? My "analysis" (conclusion) has been conservative. No great bold assertions from me (unlike those from others such as tfk)
Perhaps I shouldn't have used the word surpassed. "Reached" would have been better. Naturally this is in how the data relates to real world movement and not just trends.

Once the acceleration trend behaviour was revealed in reasonable detail, sure. Detail in both NIST and Chandler was severely lacking.

Sure, but good enough for their objectives.

Whilst I have responded to discussion about acceleration behaviour I have made it clear long ago that my personal focus was on displacement data behaviour (early motion), which is much more "interesting" imo.
"Interesting" to what end? If your talking about what it looked like, sure. From an understanding of the mechanics of the collapse, I would say not so much.

They could have presented early motion. Hugely significant. They could have rejected "FFA for n seconds". They could have...
Hugely significant how (in the fine detail you are looking for)? Early motion would be expected, I would be surprised if the building wasn't shaking, shifting and swaying all day. Unless this can be translated to a specific piece or event, it's of little use.

As far as FFA for n seconds goes, interesting but, really not all that important in the final review. (as seen by the fact explaining/expanding on it didn't really change anything in the report). These are things that if this was to happen a hundred times, they would all be slightly different due to any number of variables. It all boils down to what you want to use the information for, more detail is not always better.

I suggest you make such assertions in the appropriate thread, where you will find rather lengthy critiques from me showing the many issues and errors in the NIST analysis. Your statement above is particularly ill-advised.

NIST didn't "blow it", they used methods that was good enough for their purpose.

I suspect you just don't agree with the bold ;)


I suggest the poor part of your post is that last assertion "NIST didn't "blow it", they used methods that was good enough for their purpose.". That was a silly thing to say.

Not really. They didn't produce the level of accuracy you wanted but none of this really changes to overall result they hoped to achieve.


Now back to the regularly scheduled discussion of how many anglesAngels can fit on the head of a pin. :)
 
Last edited:
Or Angels, even.

The "early motion" was actually first measured on September 11, 2001 by the FDNY. Probably not to the precision desired here, but it undoubtedly saved some lives.
 
I think what is being discussed is video analysis and graph production.
Really ? There's next to zero video analysis within the approaching 3000 posts in this thread. There is discussion of data extraceted from video, but that's a different thing. The focus upon one graph produced nearly 2 years ago is hilarious. (aka incredibly sad)

Yup, who can make the nicest looking graph.
There's a page in this thread of folk complaining (and utterly misunderstanding) the inclusion of the perfectly acceptible term "datapoint pair" on an axis of one of my many graphs. Anything to try and "prove femr2 wrong version the umpteenth".

I can give a useful trend looking with the naked eye. ;)
:( Silly thing to say.

Sure, but good enough for their objectives.
Defending Chandler now ? Interesting.

"Interesting" to what end? If your talking about what it looked like, sure. From an understanding of the mechanics of the collapse, I would say not so much.
Oh dear.

Hugely significant how (in the fine detail you are looking for)? Early motion would be expected, I would be surprised if the building wasn't shaking, shifting and swaying all day. Unless this can be translated to a specific piece or event, it's of little use.
Oh dear (2)

As far as FFA for n seconds goes, interesting but, really not all that important in the final review.
:rolleyes: There was no period of at FFA. Chandler was wrong. NIST replicated his error. Dumb.

I suspect you just don't agree with the bold ;)
You suspect wrong. There is a full thread on that topic, which has nothing to do with femr2 trace data or analysis.

Not really. They didn't produce the level of accuracy you wanted but none of this really changes to overall result they hoped to achieve.
They replicated Chandlers erronious results. Dumb.
 
The "early motion" was actually first measured on September 11, 2001 by the FDNY.
No, Peter Hayden mentioned a transit was placed near a bulge in the southwest corner between floors 10 and 13. There is no mention of change...no motion recorded.
 
Really ? There's next to zero video analysis within the approaching 3000 posts in this thread. There is discussion of data extraceted from video, but that's a different thing. The focus upon one graph produced nearly 2 years ago is hilarious. (aka incredibly sad)

Potato......patato.


There's a page in this thread of folk complaining (and utterly misunderstanding) the inclusion of the perfectly acceptible term "datapoint pair" on an axis of one of my many graphs. Anything to try and "prove femr2 wrong version the umpteenth".

Still only producing nice looking graphs.

:( Silly thing to say.

Let me show you. The east penthouse was seen falling into the building. There was a kink forming and windows were seen breaking along the face of the building (along with several other detail. This trend leads to a belief that what ever was holding up the penthouse failed. The exterior of the building falling all at one leads us to believe that the exterior columns were left un-braced by the collapse of the core. Useful stuff. ;)


Defending Chandler now ? Interesting.

I don't know where you got this. I said good enough for their objectives. That's not defending them.



Oh dear (2)

You can "oh dear" all you like. Unless its put to a practical use, it's just data. Are you waiting for someone to do this? :rolleyes:

:rolleyes: There was no period of at FFA. Chandler was wrong. NIST replicated his error. Dumb.

Who cares. Like I said, NIST should have ignored him and not bothered with is meaningless detail. It added nothing to the report.

You suspect wrong. There is a full thread on that topic, which has nothing to do with femr2 trace data or analysis.

Where, unimportant details are harped on to the nth degree.


They replicated Chandlers erronious results. Dumb.

I agree with you here. They had no place chasing after a "truther" meme. ;)

I have no intention of getting in a pissing contest with you over what you see as important. You obviously don't believe in the saying that too much information is not always better. ;)
 
Let me show you.
Great. Will be interedting to see what acceleration trend information you are able to extract...

The east penthouse was seen falling into the building.
Oh, not acceleration trend behaviour.

There was a kink forming
Oh, not acceleration trend behaviour.

And in what direction did this "kink" form ?

and windows were seen breaking along the face of the building
Oh, not acceleration trend behaviour.

This trend leads to a belief that what ever was holding up the penthouse failed. The exterior of the building falling all at one leads us to believe that the exterior columns were left un-braced by the collapse of the core.
Oh, not acceleration trend behaviour.

I don't know where you got this. I said good enough for their objectives. That's not defending them.
So you have no critique of Chandlers methods then ? :)

You can "oh dear" all you like. Unless its put to a practical use, it's just data.
I can indeed, and with pretty good reason. Shame you don't seem capable of identifying the practical use of information already supplied many times. Ho hum.

Who cares.
If you don't, why bother commenting ?

Like I said, NIST should have ignored him and not bothered with is meaningless detail.
Yet they did, and made just as urine poor a job of it that Chandler did.

Again, I suggest you discuss matters related to "NIST blew" in the appropriate thread.
 
No, Peter Hayden mentioned a transit was placed near a bulge in the southwest corner between floors 10 and 13. There is no mention of change...no motion recorded.

Interesting. When you refer to "motion" you mean only "motion" of the entire building then? A visible bulge running up three floors isn't "motion" by your definition?

As to the transit, given the chaos of the day, I'm not surprised that every FDNY verbal interchange that led to evacuating the Salomon Bros. building area weren't recorded. It seems unlikely that a lack of motion caused this evacuation. Maybe someone else knows better than I.
 
Interesting. When you refer to "motion" you mean only "motion" of the entire building then? A visible bulge running up three floors isn't "motion" by your definition?
If it moves, it's in motion. I understand where you're coming from but there's no record of change in the distortion of the building.

As to the transit, given the chaos of the day, I'm not surprised that every FDNY verbal interchange that led to evacuating the Salomon Bros. building area weren't recorded. It seems unlikely that a lack of motion caused this evacuation. Maybe someone else knows better than I.
It may well be one of the "chaos of the day" events, similar to occupants of a helicopter stating that the top of WTC1 was leaning, whilst taking pretty high resolution photos of it which do not indicate any lean at all. Easy to be mistaken in such chaotic and rapidly evolving situations.

Regardless, trace data indicates a significant change in building motion behaviour roughly 2 minutes prior to release.

If you're aware of any other available data which confirms early motion, great.
 
If it moves, it's in motion. I understand where you're coming from but there's no record of change in the distortion of the building.
For the record I nearly made that error. :o
...till I read the second time and recognised what you actually said.

Easy to forget femr that your logic is usually very tight. And I'm supposed to be the lawyer. :o:o

....the "usually" is only my CYA because off hand I cannot recall any exceptions.
 
You can "oh dear" all you like. Unless its put to a practical use, it's just data.
I can indeed, and with pretty good reason. Shame you don't seem capable of identifying the practical use of information already supplied many times. Ho hum.
Hey fellows;
Here is a bit of "third party spectator perspective"
You have a lot of common ground where you are in agreement.
...and some areas of disagreement.
WHERE
A lot of the disagreement arises from using differing definitions of issues - not clarifying what you mean when you use words for example the word "purpose".

I would start by clarifying "purpose" but I already tried that several times and failed. :o

So you have a couple of main choices:
1) Work towards getting rid of the fog of terminology so you can see clearly what if any real areas of difference exist between you; OR
2) You can keep maximising the differences by continuing the sparring.

(and, yes, I read clearly both of your moves/hints/drifts towards "disengagement if only this last....."

Some of my recent posts try to address the main issues of difficulty as I see them. Why not humour me , back off and read what I posted. Then either ask for more clarity or tell me to b**** off. I won't be offended. :)
 

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