Monday, June 28, 2021

Why Did the Champlain Towers South in Surfside Collapse?


Around 1:30 AM on last Thursday, June 24, about two-thirds of the thirteen-story Champlain Towers South condominium in Surfside, Florida crashed to the ground.  At this writing (Sunday morning), five bodies have been recovered, but there are 156 people still unaccounted for.  As recovery efforts have been hampered by hazardous conditions and persistent fires in the rubble, it may take weeks for rescue and recovery workers to pick through the remaining debris.


This tragedy not only took the lives of many unsuspecting people, most of whom were asleep in their beds, but has raised a host of questions that may not be definitively answered for months.  At this point, all we can do is examine the logical possibilities that could explain what happened.


The building was erected in 1981, and was constructed of reinforced concrete.  The reinforcing bars ("rebars") in reinforced concrete must be protected from corrosion, especially in a humid salt-water environment such as the Miami area.  The condominium was sited on the Atlantic Ocean, and some news reports have indicated that subsidence of a few millimeters a year was occurring for some time before the collapse.  While building a secure foundation on a beach is an engineering challenge, the fact that hundreds of similar structures have stood for many decades on beachfront properties shows that it is a challenge that competent engineers can easily overcome.


At least one video exists of the collapse.  It shows the building viewed from the south, on the inside of the roughly L-shaped structure.  The collapse has already commenced at the start of the video, and it is fairly obvious that the entire central tower is moving downward.  This conflicts with some reports that the collapse began near the top and transferred downward in a pancake-type fashion.  There are lights on in a few condo units near the top, and as the electric service connections are sheared the lights go out and the entire central tower plunges to the ground, leaving the east tower to its right standing for a few seconds.  Then it topples to the west as well, leaving only the western tower standing.


An inspection report from more than two years ago indicated that concrete supports in the parking garage underneath one of the towers were showing cracks and spalling (flaking), exposing bare rebar in some cases.  This by itself is not necessarily a sign that serious structural problems are present, because superficial spalling and cracking can occur without affecting the load-bearing capability of a supporting member. 


However, reinforced-concrete structures are unforgiving of failure of even one load-bearing member.  For example, suppose for whatever reason (uneven construction of the foundation, for example), one of the support pillars in the garage underneath the central tower began to receive much more than its fair share of the load of the floors above it.  While concrete will withstand a great deal of force in compression (squeezing), it is intolerant of tensile force (pulling) and will come to pieces with just a slight tensile force.  That is why rebars are installed, to take up the tensile stress that is present in any complicated structure.


Imagine making a squat cylinder of clay with your hands, setting it on its end, and squeezing down on the top.  It's going to get shorter and the middle will bulge out.  The top and bottom will mainly be in compression, but the middle part is in tension.  Rebar is supposed to handle all the tensile forces that arise in reinforced-concrete structures.  But suppose the rebar in this overloaded column was rusty or weak for some other reason, and let go?  The column would then be free to bulge outward and fail.


Once even one column at the bottom of the structure fails, and the floor above it fails, everything above that column is probably doomed.  Floors pull columns out of alignment, those columns fail, and you get the kind of collapse we see on the video. 


This pattern is familiar to demolition experts who use explosives to "implode" reinforced-concrete buildings.  Taking out the central support columns at the bottom with properly-sized explosive charges will typically cause the structure to collapse inward on itself, which superficially looks like an implosion. 


Another possibility is that the roof of the central tower, which was being worked on, was overloaded by construction equipment and failed.  Too much heavy construction equipment played a critical role in the collapse of the I-35W bridge over the Mississippi River in Minneapolis in 2007.  However, this type of failure would have caused a sequential top-to-bottom collapse as the upper floors collided with the lower ones, and that pattern was not evident in the video referred to above. 


A possibility that no one has mentioned, but should be considered at least as a logical possibility, is sabotage.  Many of residents of the condominium were presumably wealthy and included Jews and other ethnic minorities.  Someone familiar enough with explosives and building demolition techniques could have managed to breach whatever security safeguards were in place to keep unauthorized people out of the basement parking garage, installed the explosive charge, vacated, and set it off remotely.  If this was done, the evidence lies beneath tons of fractured concrete and will eventually come to light, but not for quite a while.


One survivor recalled first feeling the building shake, then hearing a boom and seeing two-thirds of the building coming down.  This is consistent with an explosive event whose vibrations would be transmitted through the solid structure faster than the speed of sound in air.  But abrupt failures of large stressed structural elements often sound like explosions, so this account should not be treated as definitive.


Our sympathy and prayers are with those who lost relatives and loved ones in this tragedy, especially as the long process of recovery continues.  The same process will eventually answer the question of what happened last Thursday, but the wait may be a long one.


Sources:  I referred to the video of the collapse, available at; a USA Today article at;; and the Wikipedia articles "Building implosion" and  "Surfside condominium building collapse."

Monday, June 21, 2021

Foreign Hacking of Utilities: Crimes or Acts of War?


The hacking and week-long shutdown of the Colonial Pipeline last month that spread gasoline shortages throughout the East Coast is only the most visible incident in a series of worrisome cases that may presage something even worse.  A recent Bloomberg News report highlights the fact that U. S. utility systems such as oil and gas pipelines, water systems, and electric grids are extremely vulnerable to the same kind of ransomware attack that shut down Colonial's system.


A friend of mine once summarized engineering ethics with a two-word phrase: "No headlines."  Properly engineered systems that have adequate safety precautions and other mitigations against potential disaster weather all kinds of threats, from internal failures to external hacking attempts, and just keep going.  But as insiders know, beneath the publicly-visible smooth surface there may be many small incidents that engineers manage to catch before they become major problems. 


The Bloomberg report cites one:  earlier this year, the cybersecurity firm Dragos discovered that a hacker had managed to access hundreds of computers involved in operating water systems across the U. S.  Presumably this breach was repaired before any actual damage was done, but gaining access through popular software used by many of the same kinds of enterprises is often the first step in a carefully-planned cyberattack that can take months or years to set up. 


What is even more surprising than the attacks themselves is the wall of indifference that many private firms and cybersecurity companies encounter when they try to gain the attention of government agencies such as the FBI or the Department of Homeland Security.  The pipeline firm ONE Gas of Tulsa, Oklahoma found last January that a foreign hacker was trying to gain access to the computer system that controls natural-gas traffic across the south central U. S.  After fending off the attack, a cybersecurity firm involved in the defense tried to get a response from the FBI, the Department of Defense, and the Department of Homeland Security.  Representatives of these agencies listened to the story in a conference call, but that was the end of their involvement.


From the agencies' point of view, their apparent indifference may be justified.  Because ONE Gas successfully defended itself against the attack, their story was analogous to a homeowner whose attack dogs scared off a burglar trying to break into the garage.  Police might listen politely to such a tale, but because no actual crime was committed, there is not much else they can do.  And thereby hangs a dilemma that needs to be resolved if we hope to avoid successful hacking attempts on utilities in the future: whether to keep regarding ransomware attacks as crimes or acts of war.


The dilemma goes back to some fundamental legal definitions. 


What has been mostly neglected up to this point is the fact that many cybercrime organizations are hosted by, encouraged by, and possibly operated by foreign governments—China, Russia, and North Korea among them.  Readers of this blog are familiar with my position that cyberattacks committed by foreign individuals or entities constitute matters that should be dealt with by the U. S. military.  But centuries of traditional thinking about what war amounts to stand in the way.


The U. S. legal code says that an act of war is "an action by one country against another with an intention to provoke a war or an action that occurs during a declared war or armed conflict between military forces of any origin."  Right there, we run into problems if we try to consider a ransomware attack by, say, agents of the Russian government, as an act of war.


To be an act of war, and thus a concern of the U. S. military, one would have to show that Russia in that case was either intending to provoke a war, or was doing it during a declared war, or was doing it during an "armed conflict" between military forces, no matter where they came from.


The first condition requires that we somehow divine the fact that Russia, in this hypothetical case, intended to provoke a war.  Proving intent in law is notoriously difficult, so let's skip that one.


The second condition requires that a declared war is going on between Russia and the U. S.  To say the least, that would lead to other problems, so let's skip that one too.


The third condition requires only that the action happens during an "armed conflict between military forces of any origin."  There are two big roadblocks in this phrase.  The first is the word "armed" and the second is "between military forces."  I'm no lawyer, but it seems that if the armed forces of Russia attacked not our military, but only civilian U. S. targets, it wouldn't be an act of war, at least not by this definition.  And the other problem is even worse.


"Armed conflict" means fighting with arms:  guns, bombs, and similar armaments, all of which are designed explicitly to kill people.  The notion that some guy sitting in his basement in Minsk could hit a few keys and shut off power or water to millions of people in a country on the other side of the world was clearly not anticipated by whoever wrote this definition. 


If computers are used to kill or injure people, however indirectly, does that make them "arms"?  In our Big Freeze here in Texas last February, power failures led more or less directly to the deaths of over a hundred people.  I can imagine that fatalities would also result from a focused effective attack on power or water utilities, and so a lot more than just inconvenience could result.


While I appreciate the reservations that the U. S. military feels about getting involved in what looks to them like simply criminal activity, the day may come when we are faced with a situation that goes beyond anything we have seen so far:  a crippling blow to widespread and vital infrastructure, accompanied by a not-so-subtle message from a foreign government that if we will just do thus-and-such, we'll get our water and power back.  Legalese aside, that would be an act of war as far as I'm concerned.  And the best way to prepare for such an attack is to put resources, including military resources, into making sure that it can't happen here. 


Sources:  The Tribune News Service article carried by Bloomberg News was picked up in some form by numerous news outlets, including the Arkansas Online website at  I also referred to Cornell University's law school website for the U. S. code's definition of an act of war at  My column on the Colonial Pipeline hack appeared on May 17, 2021.

Monday, June 14, 2021

Should We Worry About the Chip Shortage?


A recent article in Wired goes into the reasons for the current worldwide integrated-circuit shortage, which is affecting everybody from videogamers to drivers who want to buy new cars, but can't because the automakers can't get enough chips.  Reporter Eric Ravenscraft says there are several interrelated reasons, starting with COVID-19.  But underlying them is a more structural problem:  the fact that over 60% of global semiconductor manufacturing revenue comes from factories in Taiwan, and a good portion of the rest comes from mainland China.


The economists will tell you that shortages happen when demand outpaces supply so fast that producers can't keep up with consumers.  As millions around the world transitioned to working from home, the demand for more laptops, webcams, and associated equipment soared.  At the same time, manufacturers were dealing with COVID-19 shutdowns and absent workers, although Taiwan managed to keep their total number of reported coronavirus cases under 1000—until about a month ago, when their numbers shot up to the current figure of 12,000, according to the Worldometer coronavirus website.  This does not bode well for the near-term future of chipmaking.


So even if semiconductor manufacturing had continued at its normal pace, we would have seen prices rise and supplies shrink simply because of increased demand. 


Ravenscraft then turns to the fact that the Trump administration in particular enacted policies that are consistent with trying to start a trade war with China.  The details are complicated, but the net effect on availability of chips was negative.


And because semiconductors are some of the most complexly-engineered things on the planet, you can't just stop making one kind of processor one day and start making a different kind the next day.  Millions of coordinated process steps, masks, and other details have to change for yield on a new product to rise into the profitable range, and that can take weeks or months.  Here in Texas where our Big Freeze in February knocked out power to a Samsung semiconductor plant in Austin for over a week, it took them on the order of two months just to get things going again.  So chipmaking is not a turn-on-a-dime industry.  Sudden increases in demand cause big problems with supplies that can last months or longer, as these seem to have done.


Looking at the big picture, the semiconductor industry has become as essential to the world economy as oil or even food.  People can't eat chips (not the silicon kind, anyway), but they are nonetheless vital to the way we live now.  I have dabbled in the history of technology from time to time, and now and then the occasion arises to determine the relative importance of a given technology to a given culture at a given time.  The following thought experiment is useful:  Suppose everyone woke up one morning and the technology in question had vanished off the face of the earth.  How disruptive would that be to normal life?


Some technologies would not be especially missed:  dental floss, for example.  But as we found out in Texas last February, electric power is pretty necessary to normal life for most of us, and if all the semiconductor chips vanished out of our devices one fine morning, we would be arguably worse off than we were without electric power.  Cars newer than 1960s models wouldn't run, nobody's phones would work, businesses from mom-and-pop stores to Amazon would freeze up instantly and lose most of their records (how's them apples?), and, well, it's starting to sound like some descriptions of the consequences of an electromagnetic pulse that would result from detonating a nuclear weapon above the atmosphere, only worldwide.


Fortunately, short of divine intervention, we won't have to suffer theglobal disappearance of semiconductor technology.  But this little thought experiment shows how crucial it is to modern life, and how even a mild thing like a chip shortage can have extensive and surprising ripple effects.


The other factor that makes a chip shortage serious is the rapid pace at which new products overtake old ones.  This is largely a circumstance of the industry's own making, because anybody with billions of dollars tied up in a semiconductor manufacturing plant has to keep it running at a profit or lose their investment.  So markets have to be found for all those chips, and that is why you have to turn loose of your old electronics even if you like it, because the rest of the world moves beyond it and you have to keep up or else just quit.  And sometimes "old" means only a year or two. 


Of course, the manufacturers would say that they're just trying to make their customers happy, and there is some truth to that.  But let me bring in a different voice from a different tradition.


Paul Kingsnorth is an environmental activist and writer who, after thirty years of activism and dabbling in Buddhism and Wicca, recently joined the Romanian Orthodox Church.  As an activist, he saw problems with the way the world was going, as he puts it in the June/July issue of First Things:  "We would remake Earth, down to the last nanoparticle, to suit our desires, which we now called 'needs.'"  Needs for semiconductors, for instance.


But after his conversion, he views Christianity as the history of man's rebellion against God.  The global issues he was so focused on formerly—climate change, economic inequality, and so on—he no longer views as problems to be solved, but as spiritual issues, consequences of sin.  Toward the end of his article, he writes "In the Kingdom of Man, the seas are ribboned with plastic, the forests are burning, the cities bulge with billionaires and tented camps, and still we kneel before the idol of the great god Economy as it grows and grows like a cancer cell.  And what if this ancient faith is not an obstacle after all, but a way through?"


The chip shortage looks like at most an annoying interruption of what we have come to regard as business as usual.  But what if COVID-19 and its consequences form an opportunity for us, individually and corporately, to ask some questions about the whole set of assumptions that underlie that business?  Maybe business as usual isn't where we should be trying to go.  But that is a discussion for another day.


Sources:  Eric Ravenscraft's article "Here's Why Gadgets Are So Hard to Get Right Now" was published on the Wired website on June 12, 2021 at  I also quoted from Paul Kingsnorth's article "The Cross and the Machine" on pp. 35-41 of the June/July 2021 issue of First Things.

Monday, June 07, 2021

Censoring the Censors? Florida's Anti-Censorship Law


On May 24, Florida Gov. Ron DeSantis signed a bill designed to stop social media firms from censoring free speech.  At least that's what the governor's website claims it does.  Two big-tech industry groups, Netchoice and the Computer and Communications Industry Association (CCIA), sued the state of Florida in early June over the legislation, which is scheduled to take effect on July 1.  What exactly does the law do, and why are organizations such as Facebook, Twitter, Amazon, and Google sponsoring a lawsuit to halt it?


People of certain political persuasions need not look far for motivations to pass such a law.  Following the assault on the U. S. Capitol on Jan. 6 of this year, the alternative social networking service Parler, which attracted users that had been suspended from mainstream media such as Twitter, found itself without Amazon's hosting service and cut off from Apple's and Google's facilities as well.  And former president Donald Trump is still banned from mainstream social media.  These are just the tips of an iceberg of censorship that some people view as politically motivated.


The bill signed by Gov. DeSantis would provide for fines of up to $250,000 for "willfully deplatforming" political candidates, and also allows private citizens to sue firms committing such indiscretions as well.  Social media firms cannot de-platform any statewide or local political candidate, or else they face penalties of up to $250,000 a day. 


The Netchoice-CCIA suit alleges that this law constitutes a "smorgasbord of constitutional violations," according to an article in National Review.  If allowed to take effect, the suit says it will impede the social media firms' ability to "protect their services, users, advertisers, and the public at large from a variety of harmful, offensive, or unlawful material." 


Just to take a hypothetical extreme case, you have to admit that the CCIA suit has a point.  Suppose for the sake of argument that a Florida candidate for statewide office chose to post some child pornography in a political blog on Facebook.  Leaving aside the bad judgment on the part of the candidate, Facebook would be caught between a rock and a hard place.  If they deplatform the candidate, they would face fines from the State of Florida.  But if they leave the post alone, they will be an accomplice to numerous other violations of federal and state laws. 


On the other hand, the law is part of a growing trend for legislatures and governments in general to put the screws to Big Tech in various ways.  Now that social media carry the bulk of public discourse (as opposed to the legacy media of radio, television, and newspapers), they find themselves walking a fine line between suppression of free speech and toleration of harmful content.


The Florida law is only the latest in the history of attempts on the part of states to suppress or control mass media.  Lousiana saw a more blatant effort when Huey Long, who formally left the governor's office in 1932 to be a U. S. senator but effectively kept running the state anyway, passed a steep tax on large-circulation Lousiana newspapers, calling it a "tax on lying."  In the present case, ironically, Florida is not penalizing the media for what they're saying, but for what they're not saying, or rather for what they're not allowing certain parties to do or say on their platforms. 


This issue can be viewed as a contest between rival goods:  the good of free speech for political discourse and democratic government, and the good of avoiding what, in the phrases of the CCIA suit, is "harmful, offensive, or unlawful."  And the unfortunate thing is that the parties to the conflict have little or no agreement as to exactly what these rival goods are.


Take the concept of free speech.  Back at a time, say sixty or eighty years ago, when there was a general public consensus that airing political views of almost any stripe, from theocratic conservatism to anarchic Communism, was to be tolerated for the good of the republic, there was also a general public consensus that hard-core pornography should be banned from publicly accessible media, including the U. S. Postal Service.  Gradually, however, the concept of free speech was expanded to encompass what earlier generations would regard as pornography, and today there is a multibillion-dollar worldwide online trade in porn, which not only effectively enslaves thousands who are involved in producing it but corrupts those who watch it.


Unfortunately, there is today no general consensus on what sort of content is harmful, offensive, or even unlawful.  If everything I think is harmful or offensive was banned from the Internet, the Internet would be a much smaller enterprise.  Yet when social-media companies hire hundreds of young people and give them whatever training they receive to identify undesirable content that violates the firm's "standards," and remove the content if necessary, they  exercise personal judgment which inevitably turns out to be inconsistent, intermittent, and subjective. 


Of course, important decisions such as banning Donald Trump are undertaken with the knowledge and approval of the firm's leadership, but my point is that the diffuse and multifaceted nature of social media means that whatever restrictions they try to apply will also be diffuse, multifaceted, and guaranteed to make somebody unhappy.


I don't have a good solution to this problem.  Section 230 of the Communications Decency Act protects social-media firms from being sued as though they themselves originated the stuff that their users post.  If we began treating social-media companies as common carriers like the telecommunications firms, and insisted that they let anybody post anything, we would need a more effective legal means to go after the individuals who would exploit this new freedom than what we now have.  But if we simply shrug our shoulders and let Mark Zuckerberg and his friends suppress anything they don't like, we have handed the keys of democracy to a bunch of billionaires, and that is likely to turn out badly too.  Stay tuned.


Sources:  The National Review article "Groups Representing Tech Companies Sue Florida over DeSantis Anti-Censorship Bill" appeared on June 2 at  I referred to Gov. DeSantis's website description of the bill at, and Elizabeth Kolbert's 2006 article on Huey Long in The New Yorker at, from which I took the quotation about a tax on lying.