A full critique would be longer than the article.
I don't know what IEEE publishing standards were used in publishing this, but I note the disclaimer at the top of the article...
Quote:
The views expressed here are solely those of the author and do not represent positions of IEEE Spectrum or the IEEE.
That's not a damnation by itself... But nevertheless, IEEE has distanced itself from this article for some reason. Maybe IEEE always puts that disclaimer up....but, if so... Exactly what is the prestige of published in IEEE?
Much of the problems with this article are in the presentation of the problems (and solutions) with the 737 max as being unique / novel. Dynamic instability is not new. Further, it presupposes that the flight characteristics of the 737 max are, in fact, danamically unstable. But, the description of the flight characteristic is not correct, nor does it describe dynamically unstable flight.
Fly by wire is not new---to the degree that the 737max could be considered fly-by-wire. MCAS is not new, it was outfitted on the kc-46 for example. MCAS is NOT and anti-stall control system---and was NEVER intended as such. It was intended to mimic the flight control characteristics of the 737NG, so that 737NG certified pilots could directly transfer over the the max. Anti-stall systems work ENTIRELY in a different manner. Again, this was not intended as a safety system, rather just to make it fly like another aircraft. It is clearly imperfect---catastrophically.
Flight critical software development is not new. Flight critical software development is not under control of a software development manager. Review, evaluation, test and certification are roles that are performed by entirely independant entities... Each role is in dependant of the others in LEVEL A. I've written flight critical software, personally... And I've created entire flight critical development procedures for companies that I've worked for. The auther got more wrong than he got right. For reference, the standard that generally governs such development is DO-178. For direct control of flight surfaces that would be Level A--the most stringent. Such development is NOT performed by junior engineers....those kids get to cut their teeth on Level E---"does not apply". Level A is done by the most expert of software engineers. Engineers who DO inherently understand flight dynamics, and control laws. Often they have advanced degrees in flight sciences or aeronautics.
The article presupposes that that "right" aircraft design changes would have been too expensive for Boeing to contemplate. That is to wholly not comprehend the spending power of a giant such as Boeing---or where Boeing makes its real money. It's not just in the metal. As a sometimes partner, and other times rival with Boeing...trust me, they can spend money like very few can.
Finally, DERs and DARs would not be used by most aircraft OEMs including Boeing for a program of this magnitude. The DER/DAR system is typically used by smaller companies that only sparsely do airworthiness related work... Or for smaller projects that the OEM doesn't want to waste its own more precious resources on. Rather, an organization like Boeing or the company I work for has an independent arm called the ODA (organizational design authority). This is not a materially different process, but the fact that the author doesn't know the difference suggests he's mostly talking out of the wrong end.
The anonymous quote towards the end of the article is humorous, in that it is clearly by someone who CLAIMS to have written avionics software. However, they says this was "ten years ago" as if things were wholly different back then. They were not. I wrote my first DO-178, level B software in 1995, 25 years ago. At that time, the standard was on Rev B. I haven't looked up the original publication date of Rev NC. But, it certainly predates my entry into the engineering field.
Note... This is not a point by point exhaustive critique of the article. It's just enough to point out that there are numerous factual errors... And the article is riddled with bias. There are some truths in it, but... It's hard for those outside the industry to tell fact from fiction.