Free Republic
Browse · Search
General/Chat
Topics · Post Article

To: Drago

Occam’s Razor is a cliche often abused and misused by novices leading to false inference and false pride of thinking a solution has been found which in reality is nothing more than an oversimplification.

In scientific data modeling, the simplest model is usually (99%) false. At the other extreme, an overly complex model will explain data but be unreliable in practice as conditions change.

The R-Squared coefficient of variation metric is used to assess a model’s performance in explaining how much of a dataset’s total variation is accounted for by a model.

Occam’s Razor models have very low R-Squared which means they are useless for the most part.

A high R-Squared value may seem like a good thing but if the model is overparameterized, then the model is unreliable.

The best models are in the middle, where they are not overparameterized and they have a high R-Squared.

Latching onto a simple observation and using it to conclude validity because of “Occam’s Razor” doesn’t cut it.

Taking 7500 and 7600, adding “fat fingers”, then baking it in Occam’s oven doesn’t make an entree worth paying money for.

To gather data and observations worthy of an investigation requires more.

*****************
Begin:

Make a premise: Pilot desired to xmit ‘7600’ but in error xmitted ‘7500’. Leave Occam out of it.

Observational perspective: this is not a subject that hasn’t been ensconced on the brains of competent pilots and ATCs under rigorous review. In other words, the industry knows what they are doing.

Note data point: JetBlue airliner has state of art transponder equipment. Transponder is digital. Keypad can require setting code and then pushing auto mode before squawk is carried outside cockpit. With pilot and copilot, code entry and auto mode set, multiple errors would have to occur before 7500 hijack unlawful intervention code was seen by ATC. This is all factual, nothing speculative.

But there’s more.

Go to FAA regs governing actions for squawk code 7500.

http://fsims.faa.gov/wdocs/orders/ps_orders/a_7110.49d.htm

Go to section (b) PILOT SIGNAL (COVERT) because this satisfies premise.

Read:
1 Controllers shall acknowledge receipt of beacon Code 7500 by transmitting “(aircraft call sign) (name of facility) verify squawking 7500.” An affirmative reply from the pilot indicates confirmation and proper authorities shall be notified by the receiving facility.

This section requires both
* an acknowedgement of code receipt and
* an affirmative reply

Preliminary conclusion:
Either the pilots and the ATCs committed up to 7 or 8 errors in transmitting 7500 or the premise is faulty.


103 posted on 07/03/2018 5:07:45 PM PDT by Hostage (Article V)
[ Post Reply | Private Reply | To 93 | View Replies ]


To: Hostage

Your post just proves my Occam’s Razor point..huge convoluted explanation for a transponder “typo”. Your “rules” explanation at the end doesn’t apply either...they had comms failure. No “affirmative reply from pilot” was forthcoming.


104 posted on 07/03/2018 5:20:03 PM PDT by Drago
[ Post Reply | Private Reply | To 103 | View Replies ]

Free Republic
Browse · Search
General/Chat
Topics · Post Article


FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson