💾 Archived View for dioskouroi.xyz › thread › 29439854 captured on 2021-12-05 at 23:47:19. Gemini links have been rewritten to link to archived content
⬅️ Previous capture (2021-12-04)
-=-=-=-=-=-=-
________________________________________________________________________________
For those who don’t want to click - it’s a proposed new area to search. They haven’t even looked there, yet. The title is very misleading (to me)
Quote: “ No one had the idea before to combine Inmarsat satellite data, with Boeing performance data, with Oceanographic floating debris drift data, with WSPR net data”
This seems sketchy. So many of these stories say things like this. But with all the money that went into the search, I kind of doubt “lets combine data from different sources” never crossed their mind.
Seems like puff piece for now - search the area, and if it’s there, then declare him a genius.
Godfrey presents his proposed location and a summary of his arguments for it here:
https://www.dropbox.com/s/kvu3kpu7kpjvjin/GDTAAA%20WSPRnet%2...
One thing he makes clear here is that several search vessels have already passed through the area he has identified.
Interesting that the diagrams mention confidence values from the previous searches, but the text gives no indication that he applied corresponding Bayesian adjustments to his model's results?
It hasn't been done before because the "GDTAAA WSPR analysis software" he references is his own homebrew solution.
WSPR itself is not Godfrey's invention, and in fact depends on many radio amateurs, globally, sending, receiving and analyzing signals - for the purpose of measuring the atmosphere's radio transmission parameters, I believe. Godfrey may have originated the claim that it could be used to track aircraft at long ranges, and that claim seems to have been met with considerable skepticism on the grounds of poor signal/noise ratio.
Godfrey claims that his WSPR track matches that from the Inmarsat data, where the latter is available. In evaluating that claim, I would first want to know if the WSPR track was developed without any input from Inmarsat or other sources.
UPDATE:
There is this article [1] on a blind test of the ability to track aircraft. It looks good here, but then I read this:
'There was one error of note, the initial turn of the aircraft after departure was to port and not starboard. Mike commented “_I should have told you_ that aircraft departing NSFA on RWY 08 will always turn to port due to terrain clearance considerations. _If you’d known that then no doubt the first part of the route would have been a bit more accurate._”' [my emphasis.]
One should not, of course, be providing additional evidence in a blind test, and the suggestion that this should have been done raises doubts about how the test was conducted: for example, did the person performing the analysis know the general features of the flight, such as its destination? A more convincing blind test, IMHO, would go like this: "In the interval <begin> to <end>, one airplane passed through <some region>. Develop a plot of its flight before and after, as far as possible." - and there should be more than one test, of course, and perhaps they should say, truthfully, _at most_ one airplane...
[1]
https://www.mh370search.com/2021/08/29/gdtaaa-blind-test/
Right, but he states it as if his method is a trivial (or at least non-novel) solution that others have overlooked.
"Why hasn't anyone else attempted using my obscure and unavailable software?"
Others have taken the same approach (ie intercepting-arcs/circles of the various satellite pings, plus 777 range estimates, plus subsequent oceanic current modeling), it's just that none have come up with the same exact solution.
Unless I'm missing something and his software is incorporating an additional new element I'm unaware of.
>Godfrey claims that his WSPR track matches that from the Inmarsat data
Isn't this just semi-independent confirmation of the Inmarsat data?
Given that the Inmarsat analyses do not seem to be controversial, it would be a significant boost to the WSPR method if it independently put the airplane in the same place at the same time as the Inmarsat data - but if, as seems likely, Godfrey himself did the analysis, and given that his intense involvement with the issue suggests he might know the Inmarsat coordinates by memory, that independence is not a given.
See also my update to my previous post regarding a test of the WSPR method.
I cannot say whether Godfrey is claiming the insights of others as his own - ideas often arise independently, and if not, someone has to be first. As far as I know, only Godfrey has attempted to use the method, and while it is, like others, based on radio signals, it uses quite different signals.
I agree, but the reason I said "semi-independent" is that I'm not sure the data/ calculation is independent confirmation, so much as an "independent" re-calculation of the same underlying Inmarsat-derived data.
(Hopefully someone can chime in that's more familiar with the WSPR effort.)
Ie., 1)Is WSPR using additional data sources other than the Inmarsat _sourced_ observations (even if 3rd party observations of the Inmarsat _signal_), and
2) Can WSPR calculations based on those observations confirm the Inmarsat coords both with and without using the Inmarsat data as a constitute component?
3) And the same idea applies to Godfrey's calcs, relative to both the Inmarsat data and the WSPR community's contributions.
RE: Your edit. Yes, the confirmation bias is sort of another way of stating the similar idea. Did he take the same data and play with it until "it looked right"? Vs taking different data, but possibly still having confirmation biases in reaching a similar conclusion?
Versus a completely independent _method_, where various data components can be included/omitted without coming to contradictory solutions.
The airport traffic pattern info you mention at least partially refutes the latter option. If presenting ambiguous results, you've got to at least state both/all possibilities where they exist before using extraneous information to exclude some of them.
Including the blind (wrong) guess just looks bad.
WSPR is completely independent of Inmarsat, and also of any other satellite or any navigation system. It uses signals exchanged between radio amateurs, and its intended purpose is to probe the changing radio propagation parameters of the atmosphere, globally. _If_ Godfrey's method is effective, it provides an independent estimate of an airplane's position.
https://en.wikipedia.org/wiki/WSPR_(amateur_radio_software)
Right- but you've still got to feed signal data (from the plane) into it. I guess my question is:
In the case of flight 370, what radio transmissions (and recievers) are they (WSPR) using using?
And then which transmissions (and recievers) are Godfrey using, if different?
And do either incorporate the raw data as collected by the Inmarsat receiver(s)?
Godfrey's claim is that he can detect the plane from reflections of these radio waves and the effect they have on signal strength, and to determine its position in three dimensions. Skeptics suspect he is seeing patterns in noise.
WSPR has nothing to do with any transmissions from the plane. He's effectively trying to use data from WSPR as a crude radar system.
You're incorrect:
- The debris was found after most of the searches, by one man
- it appears various militaries are hiding what their capabilites (or lack thereof) are
> The title is very misleading [...] Seems like puff piece for now
This has become typical for the BBC:
Phrase a headline as a question to which the implied answer may be "Yes", but in all likelihood is actually "No" or "Not yet".
It’s not that new:
https://en.m.wikipedia.org/wiki/Betteridge%27s_law_of_headli...
I was about to post the same.
He’s speaking with a great deal of certainty for a situation with so many confounding unknowns. I wonder whether such proposals give false hope to families looking for answers.
Even if he’s correctly narrowed down the area of the crash, we don’t know anything about the way the aircraft broke up (maybe it hit the water at full speed and disintegrated into small fragments that would not show up as aircraft-shaped on any seabed scan). Maybe the only pieces still identifiable as aircraft debris are the ones that floated off and have already been recovered. Maybe wreckage has already drifted elsewhere, or been buried beneath sea floor silt. Or even if they find the FDR, and even if it contains more details of what happened, it might not contain the true explanation of why it happened.
>> it might not contain the true explanation of why it happened.
I think this is important. It's one thing to say where it could be based on known locations, course changes, fuel, and range data. But to make sense I think an intended destination - and hence motive - is important.
One of the more plausible suggestions to me was that they were trying to get to... I think Christmas Island and didnt quite make it. That still begs for a reason but at least offered an intended destination.
Short version: A British aeronautical engineer, working with a team, has come up with a 40 nautical mile radius area where they think MH370's wreckage will be found, and is looking for somebody to finance a search effort. Nobody has searched this area or committed to doing so.
He and his team claim to have performed a novel combination of a variety of data to come to this conclusion. From the article:
Mr Godfrey said it was a "complicated exercise", but previously there was simply a lack of lateral thinking, across multiple disciplines, to bring this together. "No one had the idea before to combine Inmarsat satellite data, with Boeing performance data, with Oceanographic floating debris drift data, with WSPR net data," he said.
The Rolls Royce engines phone home periodically so I dont know if its actually Boeing performance data.
https://en.wikipedia.org/wiki/Malaysia_Airlines_Flight_370#:...
.
Just saying because when this disappearance happened, no one was even looking at the Rolls Royce engine data at the time, until someone mentioned it then it was looked at.
The data/knowledge on the RR engines phoning home was given out on this BBC program.
https://www.bbc.co.uk/programmes/b00t0yx9
I agree on the puff piece sentiment.
MH370 was an event that was well publicised globally, both on the day and for an extended period of time afterwards.
The argument that not one engineer / big-data guru / whatever in the whole world suggested combining data sources is simply not viable.
I would put it that it is more a case of:
(a) The priority (naturally) was speed in the hope of finding survivors/bodies/debris before the ocean took care of them.
(b) The need to sketch out agreements with all these third-party data people, alongside the question of "who's gonna pay for it".
(c) Time. Its taken the guy in the BBC article, what, 7 years to come up with something. Let's say "the world" put a team on it, you could, what, halve the time to 3 years or so ? Meanwhile you've got the world's media, governments and relatives of the deceased begging you take some visible action "now".
Other than an approximate position, the article contains no links to any information about where this proposed location is, let alone to the evidence which supposedly leads to this conclusion and to the assumptions which have been made. The MH370 Independent Group site has more information, but it is not well-organized - it comes across as a dump of everything anyone in the group has had to say, rather than an argument leading to a conclusion.
I would guess that this location is a result of Godfrey's recently-published WSPR analysis [2].
Update: At
, the latest post (2021-11-30) says "The GDTAAA WSPRnet MH370 Analysis Preliminary Findings place the crash location at 33.177°S 95.300°E just 6 nm East of the 7th Arc", but the link to the paper that supposedly justifies that conclusion does not work.
I feel Mr. Godfrey should put a bit more effort into presenting the argument for his conclusion before he goes around making statements that will cause additional emotional turmoil for many of the victims' relatives.
Update 2: That link exists in another version of the same post: [3]. The non-functional link on the landing page appears to have been a mistake in copy-pasting, and on the basis of [3] I see I was mistaken in saying that Godfrey is not doing enough to make his case.
[1]
https://www.mh370wiki.net/wiki/Independent_Group
[2]
https://www.dropbox.com/s/wif8oqzgm74sdqv/GDTAAA%20V4%20MH37...
[3]
https://www.dropbox.com/s/kvu3kpu7kpjvjin/GDTAAA%20WSPRnet%2...
From the article:
_The exact point determined by data calculations is around 33 degrees south and 95 degrees east in the Indian Ocean._
From a 2016 study by oceanographer David Griffin:
_the plane could be located at 35.6 degrees latitude, 92.8 degrees longitude, consistent with their previous suggested area._
That's a difference of about 150 nautical miles, which isn't chump change, but not exactly a new revelation here?
Previously on HN
https://news.ycombinator.com/item?id=27044995
, sentiment was that this analysis was garbage.