Monday, January 31, 2022

Obliging Bridge Makes Political Point


If you had to have a bridge collapse, it was just about the best collapse President Biden could ask for.  Nobody was killed, but enough people were injured to garner headlines.  And it happened about four miles away from where the President was scheduled to speak in Pittsburgh, Pennsylvania later the same day, Friday Jan. 28.  I'm sure the bus riders, motorists, and joggers who were involved didn't think it was funny to see the Forbes Avenue bridge over Frick Park resolve itself into several large disjointed chunks.  An articulated bus slid backwards into the rubble, necessitating a rescue of the bus riders by first responders.  But later in the day, President Biden made the collapse site the first item on his tour of the city, reiterating that the $1 trillion federal infrastructure bill passed last November has money to pay for replacing old bridges such as the Forbes Avenue structure, which was reportedly two years beyond its 50-year design life when it collapsed.


For years now, the American Society of Civil Engineers has displayed on their website a report card on the nation's infrastructure, and I don't recall ever seeing a grade higher than C.  They estimate it would take some $125 billion to completely clear the backlog of aging and deficient bridges that the nation has accumulated since its roadbuilding frenzy of the mid-20th century, when most of the interstate highway system was built. 


The Forbes Avenue bridge doesn't carry interstate-highway traffic, but its construction was no doubt made easier by the abundant highway money that prevailed in the 1970s.  No bridge can last forever, and steel bridges in regions where salt is used on the roads have to be monitored with special care because of problems with corrosion.  Not being a civil engineer, I can't tell from the news photos of the collapse just what kind of a bridge it was, but obviously it was the kind that can wear out right after its design life lapses.  Traffic on the bridge was not especially heavy at the time, so my uneducated guess is that a critical structural element simply decided to let go, causing a sequence of events that led to the failure of the whole bridge.  It will be weeks or months before forensic engineers get a chance to analyze the pieces and figure out what gave way.


But in a way, it doesn't matter, if you look at the broad picture of bridge infrastructure in the U. S.  The coincidence that a President who signed a bill that is going to do something about it visited the city where a bridge collapsed the same day simply draws attention to the fact that the built environment is a measure of the society that builds it.  The society that devoted such energy to building the bridges and roads of the 1950s through the 1970s was a different society than the one we have today, and unique in many ways. 


The farther we get from that era, the more unusual it looks in comparison to either the decades before, during the Great Depression when about the only public-works projects were Federal programs, and the decades after, when the rest of the world's industrial capacity recovered after World War II and removed the exceptional economic advantages that favored projects in the U. S. such as the interstate highway system.


But our economy depends on roads and bridges to be there, so it is only doing the responsible thing to maintain them.  And last fall's infrastructure bill was recognized as a sorely-needed act by some Republicans as well as Democrats, which is why it passed despite the huge price tag.  Only about a tenth of the $1 trillion will go to roads and bridges, but $110 billion is a good bit of the $125 billion that the ASCE says the nation's bridges need.  It won't "fix them all," as President Biden said about Pittsburgh's decaying inventory of remaining bridges, but it will go a long and useful way in that direction.


The park that the Forbes Avenue bridge spanned is named after Henry Clay Frick (1849-1919), a nineteenth-century industrialist who chaired the Carnegie Steel Corporation, and is most well-known now for his charitable bequests, such as his former mansion which became the home of the Frick Collection of fine art in New York, as well as the land which became Pittsburgh's Frick Park. 


Frick's name is indirectly connected with a much worse disaster than last week's bridge collapse, incidentally.  As a founding member of the South Fork Fishing and Hunting Club, a private club that named among its members many of the wealthy men of Pennsylvania, he was partly responsible for alterations of what was then the world's largest earthen dam, holding back the waters of Lake Conemaugh, where the club members liked to fish.  Despite warnings by engineers that the dam was defective, the club reportedly lowered the level of the dam by three feet.  After unusually heavy spring rains, on May 31, 1889, the dam broke, sending most of Lake Conemaugh roaring through a valley that led directly to the doomed town of Johnstown, Pennsylvania, killing 2200 people and devastating the town.


The disaster that happened last week in Pittsburgh was nothing like the Johnstown Flood, but there are a few parallels.  Engineers knew that the bridge deck and superstructure were in "poor condition" from a recent inspection.  But as it carried 15,000 people a day, they were reluctant to shut it down to make repairs, even if funds had been available. 


Now the choice has been made for them, and residents of Pittsburgh are going to have to find another way to go where they're going if they were used to crossing Frick Park on the Forbes Avenue bridge.  But President Biden pointed out that Pennsylvania is slated to receive $1.6 billion over the five-year span of last fall's infrastructure bill, and surely that will help to repair not only the Forbes Avenue bridge, but others that are time bombs set to collapse at an unknown date in the future. 


Sources:  I referred to articles on the bridge collapse and President Biden's visit at The Guardian's website, and the York, Pa. Daily Record at  I also referred to the ASCE website at,

reporting of the infrastructure bill on CNBC at and the Wikipedia articles on Henry Clay Frick and the Frick Mansion. 

Monday, January 24, 2022

Can 5G Really Knock Planes Out Of the Sky?


Let's hope we never find out.


Last week, a number of U. S. air carriers filed an emergency request with the FCC to delay the turn-on of 5G mobile-phone towers near major airports.  AT&T and Verizon were planning to turn on the towers as part of their 5G rollout, which uses a part of the electromagnetic spectrum called C-band to improve transmission rates while maintaining good geographic coverage. 


The reason the airlines were worried is that for decades, devices called radar altimeters have provided pilots with absolute altitude readings as they land.  Altimeters are used below about 2500 feet (800 meters) altitude in both fair and inclement weather as a landing aid.  In recent years, they have been incorporated into autopilot and terrain-awareness warning systems, and a malfunction of the radar altimeter could cause false warnings or more serious problems such as a crash. 


The altimeters currently in use employ the 4.2-4.4 GHz range of frequencies with a technology called FMCW, standing for Frequency Modulation Continuous Wave.  As the transmitter's frequency slides up and down linearly, the time it takes the signal to go from the plane, bounce back, and return to the altimeter is translated into a low-frequency tone which is proportional to the plane's absolute height.  It's a simple, reliable system, which is why it is used for many safety-related purposes on modern aircraft.


Last year, AT&T and Verizon paid $80 billion for rights to a chunk of spectrum just below the radar altimeter band, from 3.7 to 3.98 GHz.  Their new 5G systems use this band instead of an older lower-frequency band for improved data rates and nearly the same geographic coverage ability.  Other parts of the world have made this switch already, but with certain restrictions around airports for the same reason that has upset U. S. airlines:  the potential that mobile-phone tower transmissions will cause problems to radar altimeters.


Although the two frequency bands are separated by what is known as a "guard band" (a kind of frequency no-man's-land intended to provide adequate separation between allocations), whether a given altimeter will be messed up by a given mobile-phone tower is not that easy to figure out. 


Radio devices of all kinds discriminate against interference with a combination of digital filtering (in software, essentially) and analog filtering, which consists of pieces of hardware called band-pass filters.  Unfortunately for the software people, a strong enough interfering signal outside of a device's intended band can overwhelm the "front end" (the part that turns an analog radio signal into a form usable by digital systems) and disable it.  To prevent this, the system needs to have better analog filtering, which consists of mainly passive frequency-selective circuits that cost money and space.


Before 5G came along, I'm sure the radar-altimeter people put just enough filtering into their systems to prevent the worst existing adjacent-channel offenders from messing up their radars.  That might have been good enough a few years ago, but now that the phone companies are putting transmitters in the 3.7 to 3.98-GHz band, it may or may not be good enough.


These kinds of issues are typically straightened out with the assistance of field tests, but it appears that the Federal Aviation Administration (FAA) has not surveyed radar altimeters with a view toward their interference-rejecting ability until the last few weeks, according to one report. 


As things stand, AT&T and Verizon have agreed to delay for another six months, but not cancel, their plans to roll out 5G near airports.  It looks like a big game of chicken has been going on between the wireless carriers and the Federal Communications Commission (FCC) on one side, and the airlines and the FAA on the other side, with each side pulling rank and blaming the other for causing the problem.  The only thing that seems to have stopped the rollout this time was intervention by the White House, which extracted a promise from the phone companies that the rollout will be delayed another six months.  But AT&T and Verizon say that's the last delay they're going to put up with, as we're falling further behind other countries already with our 5G rollout. 


It's a sad thing when you see not only businesses, but federal agencies, duking it out in public and making grandstanding plays to force resolution over what should be a strictly technical issue.  But that's an engineering-nerd point of view, the viewpoint that ignores money and politics and sees only the technical aspects of the problem. 


Compared to places like France, the FCC allocation allows higher power and closer frequency spacing to the altimeter band, and so the airlines may have a valid concern that turning on the 5G towers could mess up some altimeters.  Because the altimeters are so critical to aircraft safety systems, that is a problem that you want to stay away from by a wide margin. 


Evidently, more recently designed altimeters use improved band-pass filters that are more likely to reject 5G interference, but the airlines were unwilling to retrofit their planes with all-new altimeters.  And the FAA seems to be complicit in this, dragging its feet about even doing tests or inspections until recently.


The airlines and the 5G system operators now have six more months in which to work out a solution that is both technically sound, keeping us far away from having radar altimeters jammed by Aunt Suzy's phone call (or movie, more likely), and financially agreeable.  AT&T and Verizon are the big spenders in this picture, laying out billions for the spectrum and more billions for their upgraded systems, which they now want to start profiting from, understandably.  On the other hand, the airlines have not been having a good couple of years lately, in case you hadn't noticed, and I suppose they thought they could avoid yet another expense by jawboning and threats to shut down the entire air-transport system.  It worked once, but it's unlikely to work again.


I'm glad to be writing about a bureaucratic tempest in a teapot rather than about a clear-air crash that is discovered to be caused by a 5G tower interfering with a radar altimeter.  But this whole thing could have been handled better, and now the parties involved have six months to do it right.


Sources:  I referred to an article carried by the Austin American-Statesman on Jan. 19 from AP by David Koenig entitled "AT&T, Verizon to delay new 5G" and a piece by Jake Hertz on the site at  I also referred to the Wikipedia article on radar altimeters. 

Monday, January 17, 2022

Ukraine Gets Cyberattacked Again


First, a little geography lesson.  Ukraine sits north of the Black Sea, bordering Poland, Hungary, and Romania on its west and surrounded by Russia to the north and east.  Like Poland, the Ukraine has been subjugated for much of its history by foreign powers—the old USSR for most of the twentieth century, and even by Lithuania back in the 1400's A. D.  But when the USSR collapsed, the Ukraine gained independence again.  It is the poorest country of Europe, but has rich farmlands, which is one reason why foreigners want to take it over.


If you've been paying any attention to world news, you know that Vladimir Putin has been saber-rattling about a possible invasion of Ukraine recently, massing 100,000 troops on the border between the two countries and ramping up his warlike rhetoric.  Russia has been chipping away at the country since at least 2014, when the pro-Russian President of Ukraine, Viktor Yanukovych, lost an election, and Putin invaded the Crimea, the peninsula that sticks out into the Black Sea and separates it from the Sea of Azov to its northeast.  Having succeeded in that, Putin has since been backing forces that have taken over portions of eastern Ukraine, and it appears that he would like nothing better than to welcome the entire country back to the domination of Russia.  So far, the government of Ukraine has had different ideas.


As part of Putin's campaign, a war that isn't quite a war, most authorities agree that Russian-based hackers mounted a cyberattack called NotPetya back in 2017.  It was aimed primarily at Ukranian institutions, but it also affected thousands of other systems as well.  The White House later estimated that NotPetya caused about $10 billion worth of damage worldwide. 


Now we come down to this week.  On Jan. 15, dozens of Ukrainian government computer systems were infected with malware disguised as ransomware.  An infected computer displayed a demand for a certain ransom to be paid in Bitcoin, but what really happened is that the malware "renders the computer system inoperable," ransom or no ransom. 


Microsoft issued a statement saying that they observed these attacks aimed primarily at Ukrainian government agencies and closely-allied organizations, and that they had issued updates that will address the problems.  But in the meantime, the Ukraine is suffering yet another cyberattack which appears to be instigated by Russia, although no firm evidence of the source has yet been forthcoming.


To my knowledge, nobody has actually died as a result of the most recent cyberattack on the Ukraine.  But to the extent that the public relies on computer-mediated government services, the consequences of a massive shutdown of government computers can range from the inconvenient to the life-threatening, in government-run hospitals, for example. 


In the logic of war, an enemy's assets are always a target, and now that computer networks and systems form so much of the infrastructure of modern life, they have become a uniquely vulnerable target.  Cyberattacks borrow from the fields of espionage, sabotage, and terrorism to create an insidious threat that knows no boundaries.  And defending against such attacks is a responsibility that is widely distributed among both public and private actors. 


All these features make cyberwarfare a different kind of thing from conventional warfare, and it is taking time for both military and civilian thinking to catch up to it. 


When this topic has come up in the past, I have taken the position that the U. S. military, in any event, seems to have an overly narrow focus on what cyberwarfare might amount to in the future.  While I am no technical expert in this area, I can see that even cyberattacks on U. S. organizations that have been definitely attributed to government-sponsored hackers in China or Russia do not seem to cause much concern on the part of our government, except to provoke warnings to private interests to do their cybersecurity better. 


That may make sense if you're a Boeing or a Kaiser Permanente, with entire staffs of IT security specialists.  But especially in the U. S., we have a great many small businesses whose functioning is nonetheless critical to our economy.  Many of them can't afford a full-time IT person, so IT maintenance is handled on an as-needed basis:  if something breaks, the owner hires somebody to fix it, but otherwise deals with things on his or her own. 


A supply-chain cyberattack similar to what was used against Ukraine could target a popular piece of software such as, for example, Quicken—something that almost all small businesses use.  With a few keystrokes, such an attack could cause devastation far beyond what we are presently seeing with the Omicron COVID-19 variant, which has done nothing worse than kill thousands of people and cause massive absenteeism, both involuntary due to sickness and voluntary due to vaccine mandates. 


The fact that nothing like that has happened in the U. S., with a few exceptions, may mean that the way we are doing things is just fine and we don't need to worry about a massive cyberattack that would bring the U. S. economy to its knees.  On the other hand, it may mean that whoever is capable of mounting such an attack is simply biding their time, awaiting the proper geopolitical moment when such an attack could be coordinated with more conventional warlike actions for maximum effect.  I hope it's the former, but I suspect it might be the latter.


What am I asking for?  Certainly not for every software app to be government-certified as secure.  At the university where I work, we have experienced a small-scale version of that type of thing, and all it has done so far is to create a lot of confusion and delays in purchasing needed software.  If there are government and military forces out there safeguarding not only their own systems, but those belonging to the public at large, I would at least like to know about it, in a general way.  And because my federal taxes are paying for it, I'd like to know what I'm getting for my money.


In the meantime, we can hope that the Ukrainian government has figured out how to defend itself and its citizens from what has to be the worst spate of cyberwarfare endured by any nation so far.  And maybe we can learn some lessons from them:  either good examples if they succeed, or bad examples if they lose and get absorbed into Russia. 


Sources:  I referred to the article "Microsoft discloses malware attack on Ukraine govt networks" which appeared on the AP News website on Jan. 15 at  I also referred to a Microsoft blog at Wikipedia articles on NotPetya, Ukraine, and the Crimea.

Monday, January 10, 2022

Portable Generators and Carbon Monoxide: What To Do?


In February of 2021, over 200 Texans died when winter storm Uri disabled the power grid for several days.  According to one source, 19 of those deaths were related to carbon-monoxide (CO) poisoning.  If you were one of the fortunate people to either have a portable generator or buy one before the stores sold out and closed, you took the risk that carbon monoxide from it could kill you.  About 70 people every year die in the U. S. when CO from portable generators gets into a living space.  In a recent investigative report, journalists with the Texas Tribune and NBC News looked into the question of what portable-generator makers have done to prevent deaths caused by CO emissions from their products.  I summarize and comment on their findings in what follows.


For people like Craig Curley, the answer is "not enough."  Caught in another weather emergency—Hurricane Ida, that struck the Texas and Louisiana coasts last Aug. 29—Curley bought a portable generator and set it up at his ex-wife's house in case the power went out.  Unfortunately, the generator was close to the open back door, and when he visited the house next day, he found that his ex-wife and their two children were dead from CO poisoning.


For manufacturers, the answer seems to be, "hey, we're trying."  The Portable Generator Manufacturer's Association (PGMA), an Ohio-based industry group, says through its representative Edward Krenik that about 60% of new generators have CO sensors that shut the generator down when the CO level becomes too high.  This leads to the further question of what too high is.


CO is a colorless, odorless gas that in high enough concentrations can kill a person in minutes.  A CO levels chart from a gas-sensor manufacturer says that the maximum safe indoor CO level averaged over 8 hours is 9 PPM (parts per million).  200 PPM will cause noticeable physical symptoms and will kill you in hours, and 800 PPM will kill you in a few minutes.  The ANSI-PGMA G300 specifications dated 2018 and housed at the PGMA's own website says that CO monitors for their generators must stop the machine within 10 minutes if it detects 400 PPM, and within 30 seconds if it detects 800 PPM.  That sounds adequate, but not great. 


From an engineering point of view, there is probably a tradeoff between making the CO sensor not sensitive enough, and killing people as usual, or making it too sensitive and having the generator cut itself off without a good reason to do so.  The levels chosen in the standard seem to be a reasonable compromise between the two extremes. 


There is another alternative, which was explored by a research team at the University of Alabama. They outfitted a portable generator with fuel injection and other measures that reduced the CO emissions by 90%.  Despite industry criticism that such measures would cost too much, at least one manufacturer has adopted them and sells a generator that emits much less than the typical levels of CO.


Why hasn't the government required all portable generators sold in the U. S. to be either virtually CO-free or to have CO safety interlock devices, or both?  It has to do with a kink in the Consumer Product Safety Commission laws that allow manufacturers to try developing their own standards first.  And that's what they've done, but for some people like Craig Corley, it's too little and too late.


The field of engineering ethics has seen this type of process work out in other fields before.  A new product or system is introduced, whether it's portable generators cheap and reliable enough for residential use, or steam railways, or steamboats.  Fatal accidents occur that could have been forestalled if the operators took certain precautions (don't put the generator close to a window, don't ignore track safety rules, don't let your boiler water run too low). 


At that point, the designers of the system face a choice:  do we double down on telling people how to run things safely, or do we design safety into the device itself so that even idiots who never read instructions can't hurt themselves?  That's putting it a little harshly, but that's what the choice amounts to—jawboning or foolproofing. 


If persuasion and consumer education don't work, the deaths continue and eventually, some particularly gruesome tragedy pushes the regulators and politicians over the tipping point, and a new regulation is passed.  Voluntary industry standards have the flaw that there is always somebody out there wanting to sell things a little cheaper and a little more dangerous.  So if society wants to be sure that such evasions don't happen, third-party enforcement—typically by a government agency—is the only choice.


The PGMA is now facing that choice, and so far they are dragging their feet.  They point out that even if all new generators from this moment on were sold with both CO detectors and emission controls to reduce CO by 90%, there are a lot of old dangerous units out there which will be around for decades, considering that emergency generators spend most of their time not being used.  And nobody wants a witch hunt to ferret out old dangerous generators from private hands. 


While even one accidental death in this fashion is too many, you could make the economic argument that if CO regulations increase the price of portable generators so much that some people don't buy them, and an emergency comes along like the Big Texas Freeze, and those folks who otherwise would have bought a cheap generator die instead, well, CO prevention could actually lead to more deaths.  Not being an economist, I'm not going to make that argument, but I trot it out just to show that the problem is more than a simple one of greedy generator manufacturers heedlessly endangering the public.  


My guess is that sooner or later, all portable generators for consumer use will be required, by the government, to have CO shutoff sensors, and the higher-end models will make low CO emissions a bonus selling point.  It's impossible to make the units perfectly safe, and if users follow instructions, the units are no more dangerous than running your car in your driveway—depending on how old your car is.  I've heard that with catalytic converters and all, it's no longer possible to commit suicide by closing your garage and sitting in the car with the motor running.  But I'm not about to try it to find out.


Sources:  The article "Generators can cause deadly carbon monoxide poisoning.  But the industry resists rules to make them safer." by Lexi Churchill, Perla Trevizo, and Ren Larson of the Texas Tribune and Mike Hixenbaugh and Suzy Khimm of NBC News, appeared originally on the Texas Tribune website at, and was republished in the Jan. 2, 2022 Austin American-Statesman, where I saw it.  I also referred to the PGMA's website at, obtained the figure of CO deaths during the Big Freeze from, and referred to for the effects of CO poisoning.

Monday, January 03, 2022

A Philosopher Defines Engineering


Most engineers give little if any thought to professional philosophers, but the reverse is not the case.  One philosopher who has given a great deal of thought to engineers and engineering is Michael Davis, a professor at the Illinois Institute of Technology.  In his recent book Engineering As a Global Profession, Davis takes on the task of defining engineering, and finds that it's not as easy as you might think.


One of my own favorite short definitions of engineering was penned almost incidentally by the English essayist G. K. Chesterton (1874-1936).  In discussing the difference between cultures given to contemplation (such as India) and cultures given to engineering (such as America), he says the latter are "people engaged in the application of physical science to practical commerce."  Davis would say that while some engineers do exactly that, other engineers who we legitimately call by that name work as teachers or government inspectors, neither of whom apply physical science to practical commerce.


Chesterton's definition is an example of trying to define engineering by function.  This sort of definition says that engineering is characterized by what engineers do.  The functional definition of engineering says that any material artifact that takes some thought and planning to do is an engineered product.  By this definition, engineering goes back at least to the time of Egypt's Great Pyramid, which was built around 2600 B. C. 


Davis thinks that however smart or impressive the people were that built the Great Pyramid, they were not engineers.  Neither was Benjamin Wright, one of the people who supervised the construction of the Erie Canal in upstate New York in the early 1800s, despite the fact that he was titled "Chief Engineer."  Davis prefers to call Wright a surveyor who supervised a project that we would now call an engineering project, and points out that in those days, anyone in charge of an "engine" (which could be any machine from a steam locomotive to a crane) was called an "engineer."  To this day, we call people who drive railroad locomotives engineers, but that is obviously a different use of the word than the one we are trying to define.


Well, if Davis doesn't want to use a functional definition of engineering, what does he want to use?  One type of definition he favors is a disciplinary one.


A discipline, according to him, is "an easily recognizable body of knowledge, skill, and judgment useful for a certain activity."  And he traces the discipline of engineering back to the École Polytechnique of Paris, which around 1800 evolved a curriculum of mathematics, physics, chemistry, and mechanical drawing to enable its military-officer graduates to build structures of interest to the army, such as bridges, roads, and fortifications. 


With the advent of steam railways in the 1820s, such people proved to be useful for designing both rolling stock and the railways themselves, which were largely non-military.  Over the next decades, the military-trained individuals who applied their military-engineering discipline to such civilian projects started calling themselves civil engineers, to distinguish themselves from military engineers.  In the U. S., Rensselaer Polytechnic Institute granted the first degree in civil engineering in 1835, and used basically the same curriculum that the military academies used for their engineers.  Since then, thousands of private and public universities have offered engineering programs of many kinds, but Davis would say they all share the same basic disciplinary structure.


One more aspect of engineering rounds out Davis's definition:  engineering is a profession.  And what is a profession?  Over the years, Davis tried many definitions drawn from various approaches to the problem.


Sociologists look at economic or political signs of a profession.  Economically, professions tend to control the market for their services so as to increase their own prosperity.  Politically, they tend to favor laws that uphold professional standards and discourage non-professionals from pursuing the professional activity in question.  Davis eventually rejected these ways of defining a profession as either not adequately capturing what engineers do, or as including things that most people would not consider a profession.  For example (mine, not his):  drug dealers tend to make a great deal of money as long as they stay out of jail.  By the economic definition alone, one might be tempted to call drug dealing a profession.  But no reasonable person would.


After many years of thought and interviewing dozens of engineers, Davis came up with a definition of profession that he is reasonably happy with:  "A profession is a number of individuals in the same occupation voluntarily organized to earn a living by openly serving a moral ideal in a morally permissible way beyond what law, market, morality, and public opinion would otherwise require."  Of course, Davis thinks engineering is a profession, and he also includes law and medicine in the list of professions. 


But interestingly, he does not think business management in general—the kind of things MBAs do—is a profession.  Why not?  Davis puts it this way:  "In the 1920s, management ('business administration') seemed destined to join architecture, engineering, law, medicine, nursing, social work, teaching, and the like occupations as a profession. . . . But, by the 1960s, it was already clear that business management was not going to be a profession (in our preferred sense).  Business managers were happy to declare that their primary loyalty was to their employer; their primary goal, to 'maximize' their employer's profit."  In other words, business managers did not as a group profess to serve a moral ideal beyond maximizing profit for their companies. 


So at last, what is Davis's definition of engineering?  To get an adequate answer to that question, you will have to read Davis's book.  But in the space remaining, I will say that engineering is not merely a function, although designing appears in most lists of what engineers do.  Engineering is most certainly a discipline:  an organized body of knowledge and judgment that requires maturity and experience on the part of its practitioners.  And engineering is a profession:  a group who pledge themselves to a moral ideal that, however imperfectly realized in some cases, serves to unite and guide the group to improve the material aspects of human life. 


Sources:  Michael Davis's Engineering as a Global Profession:  Technical and Ethical Standards was published in 2021 by Rowman & Littlefield, and all quotes above are from the book.