|<<>>|3 of 56 Show listMobile Mode

Links and Notes for November 12th, 2021

Published by marco on

Below are links to articles, highlighted passages[1], and occasional annotations[2] for the week ending on the date in the title, enriching the raw data from Instapaper Likes and Twitter. They are intentionally succinct, else they’d be articles and probably end up in the gigantic backlog of unpublished drafts. YMMV.

[1] Emphases are added, unless otherwise noted.
[2] Annotations are only lightly edited.

Table of Contents


Economy & Finance

Getting High on Inflation by Dean Baker (CounterPunch)

“It’s not easy to determine how quickly supply chain issues will be resolved, but when they are, we are likely to see the price of a wide range of goods, starting with cars and trucks, reverse itself and start falling. This will be true not only for consumer goods but many intermediate goods that have been in short supply in recent months. The end of the backlogs is also likely to mean a reversal in shipping costs, which have risen by 11.2 percent in the last year, adding to the price of a wide range of products.”

Thank goodness! Then we can get back to normal. More stuff! Dean usually drops a comment about how improving shipping will improve the economy, but be bad for the environment, but he’s playing it pretty straight in this article.

“The standard remedy for inflation is to deliberately slow the economy with higher interest rates from the Fed and possibly cuts in government spending and/or tax increases. The idea is that by slowing the economy and throwing people out of work, we can put downward pressure on wages, which will then mean lower prices.

That sounds stupid. I know he’s just citing standard economic cant, but it’s ludicrous on its face. Who would come up with a deliberate strategy like that? Does the whole world do this? Or just the United States? Slowing down the economy would be a great idea right now—it’s pretty overheated on nearly all fronts—but why do you have to throw people out of work? You can only even consider doing this—and hope it works—in a country without unions. How about we reduce pay for wildly overpaid management instead? You could save a lot that way. I’m surprised Baker didn’t suggest it. It’s his idea. Is this the same guy writing this as the usual columns?

“In the latter category, universal pre-K and increased access to childcare will make it easier for many parents, primarily women, to enter the labor force or to work more hours.”

Is that really good? C’mon Dean. You sound like a robot in this column.

“If we’re going to talk about the well-being of these families it is incredibly irresponsible to only talk about the spending side of the ledger and ignore the income side.”

True, but he’s writing as if these gains had already happened, as if people were suddenly magically not poor. A bigger refund in April doesn’t address inflation now. There are more people in poverty than ever. (I seem to recall reading that something like 10M more people were in poverty than two years ago, but don’t hold me to that.) These measures just get us back to the shitty place we were before the pandemic.

“For what it’s worth, it seems that financial markets also agree with this assessment. The interest rate on 10-year Treasury bonds is only 1.56 percent, well below the pre-pandemic level. That is not consistent with a story where markets expect 4 or 5 percent inflation in coming years.”

How does he lend credence to the predictive power of the market? The treasury bill market indicates that the U.S. is doing magnificently. Is that even really true? Or do we just look at the dials and gauges built by the people robbing us blind and nod sagely to ourselves, thinking ‘everything looks fine’. It’s like Andy Garcia watching his vault in the Bellagio, but it’s not his vault, it’s a fake. But he thinks everything’s fine until things are suddenly very much not fine. And they’ve been so for a while, but he was blissfully unaware.

“Also, contrary to gloom and doom predictions, the dollar has been rising in value against the euro and other currencies. That is also not consistent with a belief that the U.S. is facing a wage-price spiral.”

Or that the euro is tanking. Oh, look at that—it’s almost even with the Swiss Franc and the Swiss Franc is kicking the dollar’s ass again.

Smooth Criminals by Nicole Aschoff (Jacobin)

“As Tom Wright and Bradley Hope describe in Billion Dollar Whale: The Man Who Fooled Wall Street, Hollywood, and the World, dozens of A-list celebrities, including Leonardo DiCaprio, Jamie Foxx, and Kim Kardashian, came out, many of whom were paid big money to attend. Swizz Beatz, Kanye West, and Ludacris provided musical entertainment, with Britney Spears bursting from a cake to sing “Happy Birthday.” Jho Low himself was gifted a bright red Lamborghini by nightclub owners Noah Tepperberg and Jason Strauss, as a thank you for the millions of dollars he had spent in their establishments over the years.”

This is what people should remember: billionaires treat millionaire celebrities like millionaires treat everyone else.

“Elites and corporations looking to hide their earnings are usually engaged in unethical but legal tax avoidance rather than illicit activities. But the secrecy afforded by this global architecture makes it difficult to tell the difference.
“[…], there is a deeper challenge in tackling global crime. Crime is historical, integral to capitalism, and fueled anew each day by the logic of our global for-profit system. This is not to say that we should settle on the banality that capitalism is a violent system that begets crime — or the naive belief that, if we can just get rid of capitalism, we’ll eliminate crime. It is simply to say that if we truly want to reduce global crime, it is necessary to comprehend how the internal logic of our for-profit system continually re-creates the conditions for crime to thrive.
“In pondering how it has come to pass that the great majority has “nothing to sell except their own skins” while a select few have wealth that “increases constantly although they have long ceased to work,” Karl Marx warned against the “insipid childishness” of bourgeois origin stories: “In actual history it is a notorious fact that conquest, enslavement, robbery, murder, in short, force, play the greatest part.”
“As Jason W. Moore argues, accumulation in capitalism relies on constantly evolving combinations of exploitation (paying workers less than the value of what they produce) and appropriation (taking, often through violence, the fruits of labor and nature): “Absent massive streams of unpaid work/energy from the rest of nature . . . the costs of production would rise, and accumulation would slow.” Capitalism cannot survive by exploitation alone; profit-making requires appropriation.

“The pitied victims of crime, and the mobsters and thugs who do the work of committing the crimes and often die “like flies,” are visible for our inspection and condemnation. Higher up the chain of appropriation, the elites who buy the coke, the venture capitalists who gamble the laundered money, and the elected officials who spend the bribes are much harder to see.

“Reckless profligacy like that demonstrated by Jho Low can momentarily bring this world into view. But for the most part, the greatest beneficiaries of crime are blurry or invisible. From the penthouse, the violence and lawlessness experienced daily by those at the bottom are nothing more than evidence of a system working according to design.

Elon Musk Sold Some Stock by Matt Levine (Bloomberg)

“Also, if you tell a client not to do a deal that would bring in a big fee, maybe in 10 years you will be spoken of in tones of hushed reverence as The Banker Who Told The Client Not To Do The Deal, but meanwhile you won’t bring in the big fee and your bonus will be lower. Future tones of hushed reverence won’t buy you a beach house.

Having principles costs you wealth in the short run. That’s just how our system works. The incentives are never lined up in the other direction.

“To be clear, you could reasonably disagree. You could have some financial reasons to stay away: Sure TMTG is trading at a $7 billion valuation now, but it could go below $875 million later. That’s entirely possible, given that it is an imaginary company. It has disclosed no financial information, no business plans, no nothing; the market’s belief that it is worth $7 billion is based on political loyalty and memes and some vague bluster from Trump.
“[…] it would be very hard for a professional investor to underwrite a $7 billion valuation for a company that is at this point, I really must emphasize, imaginary. What do its financial statements or projections or business plans look like? Who is building the technology for its social media platform? I will not belabor these questions because no one actually cares, but if you are a professional being asked to underwrite a $7 billion valuation you should probably care?”

“Now: Think about crypto financial literacy! Schools are gonna invite random crypto firms in to teach “What You Need to Know About the Blockchain,” and the crypto firms are going to come in and help kids set up crypto wallets and convert their lunch money into Dogecoin, and then the crypto firms are going to steal absolutely all of the lunch money.

“And that will be a good educational experience! New York City kids will grow up knowing a lot more about crypto than kids elsewhere! “Crypto is a way for people to bamboozle me with technical-sounding terms and vague rhetoric about empowerment, and then steal my money,” they will think! Correctly!

Doing Fraud on Securities Fraud by Matt Levine (Bloomberg)

Reddit-fuelled meme stock Naked Brand Group has acquired US-based electric vehicle startup Cenntro Auto Group in a reverse-scrip deal giving the former lingerie penny stock exposure to one of the hottest new tech themes. The Nasdaq-listed lingerie company, once behind brands such as Loveable and Pleasure State, shot to fame after Reddit traders piled in following “meme-stock” euphoria in early 2021, ballooning its shareholder base to over 900,000 investors.

“On Tuesday, Naked Brand announced it would swap 70 per cent of its outstanding shares plus $US282 million … for Cenntro Auto, an electric vehicle developer that has built and sold 4,000 commercial electric vehicles.”

Oh c’mon! Are you kidding me? A lingerie company has so much cash on hand now that it just pivots to electric cars? How is that a thing? How is this even useful? Just firehosing money at certain lucky winners in the meme-stock sweepstakes? JFC we have real problems to solve.

“I don’t know! Capitalism is different now! It used to be that, if you were a struggling online lingerie retailer, you could try to sell more lingerie, or you could try to cut your costs of selling lingerie, or you could expand into some related apparel business lines, or you could shut down, I don’t know, I’m sure there were more options, but if you went to an investment conference and said “I’m tired of lingerie let’s do electric cars” investors would be skeptical. But now sometimes you can be a struggling online lingerie retailer and get hit by Reddit lightning. And if that happens, you gotta seize the moment. “People are paying attention to us, we can raise money, only one choice here: electric cars.” (Honestly two choices: electric cars or crypto.) And off you go. Now you run an electric-car company. Congratulations.
“I honestly think that business schools in 20 years will be teaching Aron’s earnings-call transcripts. The guy woke up one day in a bizarre new world and said, well, okay, this is my world now; he sat down and figured it out, and he embraced it fully and rigorously. I feel like a lot of chief executive officers of real physical operating businesses would get tired of people on Twitter pestering them about non-fungible tokens. Aron is like, this is where the money is, I will talk about NFTs for the rest of my life if I have to. His company’s stock is up 1,800% this year.
““That’s just not how it works,” I want to shout; “this is a client-service business and clients need continuity of coverage, and without total commitment you will never learn the skills deeply enough to be useful,” all the usual stuff. But it is the case that (1) everyone who is not indoctrinated into investment banking tends to say “why not hire more people and work them less?” and (2) banks do seem to be having trouble keeping analysts. Perhaps emailing outsiders for advice has some merit.”

People can only work so much. They have to feel that the skill they learn while working their brains out is valuable to them. (And maybe to their value as a person?) Like, Uma was tortured by Pai Mei, but she became a spectacular fighter, tough as nails, and learned the five-finger death punch. These people, on the other hand, learn how to make good slide decks while losing youthful health and fitness and never seeing nature and only having contact with abusive clients whose behavior convinces them that money and escape from their lower class is the only goal in life, if they hadn’t already believed that before. They believe in nothing but Flucht nach vorn, wanting nothing more than to switch places with their torturer.

McKinsey Partner’s Insider Trading Strategy Was Bad by Matt Levine (Bloomberg)

““Insider trading,” I like to say, “is not about fairness; it’s about theft.” It’s not illegal to trade when you know something no one else knows — that’s the whole point of trading! — but it is illegal to trade when you know material nonpublic information that you got illicitly. Generally that means misappropriating material nonpublic information that belongs to the corporation.”
He would need about $143 million to exercise those options, and could owe more than $9 billion in federal income and Medicare taxes upon exercising them. Under California law, Mr. Musk also likely would face a sizable state tax burden because exercised options are treated as compensation partly earned in the state while he lived there.”

The reason he would owe $9B is because he would make so much more.

“A basket of options is worth more than an option on a basket,” is a little bit of derivatives folk wisdom that I sometimes quote around here. You have two businesses, A and B, each of which will produce cash flows of positive $100 if things work out or negative $100 if they don’t. If you buy stock in each of those businesses, each stock will be worth $100 if things work out (and you get the cash flows) or $0 if they don’t (limited liability baby!). Assume those things have equal, independent probabilities; each stock is worth about $50. If you buy both stocks they’re worth $100.

“Now you combine those businesses into one company, Company AB. If you buy stock in Company AB and both businesses work out, your stock will be worth $200. If A works out and B doesn’t, though, your stock will be worth $0: The negative cash flows from B will offset the positive cash flows from A. Same if B works out and A doesn’t. If neither works out, the cash flows will be negative $200 and your stock will still be worth zero.

“There is an old-fashioned view where you run a business, the business produces profits, and you use the profits to build other businesses. In that view, a conglomerate is good, because it produces lots of profits from various businesses and can use them to build more businesses. But this view is out of fashion in modern finance, where the theory is that you run a business, the business produces profits, you return the profits to investors (via dividends or stock buybacks) and the investors fund other businesses. If you want to do a new project that requires funding, you raise money from investors rather than self-funding it. (By starting a startup, for instance, or else by doing a stock or bond offering at an existing company.) Investors like this stuff and consider it good governance, because it gives the investors control over what projects get funded.
“One more reason for the decline of conglomerates is the norm of diversified investors. Forty years ago it was just about reasonable for GE’s executives to say “we want to run a diversified company to give our shareholders the best possible portfolio of businesses,” because it was just about reasonable for GE to conceive of its shareholders as individuals who owned only GE stock. Now the only reasonable way for a big company to think about its shareholders is as index funds. GE’s top five shareholders are, completely unsurprisingly, T. Rowe Price Group and Vanguard Group and BlackRock and Fidelity and State Street, huge diversified institutional asset managers. If they want to own a health-care business and a jet-engine business they can just do that themselves; they don’t need GE to do it for them.
“Arguably the story is not “the conglomerate is dead” but rather “the sort of celebrity imperial CEO who can build a conglomerate is not Jack Welch anymore, it’s Jeff Bezos.” Charismatic tech founders are trusted to run whatever businesses they want; professional managerial types are not. If Elon Musk decided to buy GE’s jet-engine business his shareholders would love it.”

This is an at-best lateral move. Not an improvement for society.

“If you take your company public by merging with a special purpose acquisition company, various IPO rules do not apply. The main one that we often talk about is that SPAC mergers allow you to market your company based on future financial projections, while IPOs are more limited to historical financial results; if you have no revenue but hope to have lots of revenue, the SPAC can be appealing. But there are other differences. For instance: No quiet period.

Titans by Benjamin Braun & Adrienne Buller (Phenomenal World)

“Whether social inequality or the climate crisis, proponents of universal ownership contend that the enormous externalities of corporate capitalism will, eventually, diminish shareholder returns, and therefore universal owners should and will act to minimize them. It’s an elegant theory, but is it true? Ultimately, the answer to this question hinges on how we understand ownership.”
“In the decades since the shareholder value regime took hold of corporate governance, inequality has soared, investment and growth have stagnated, 70 percent of wildlife has vanished, and a steady course has been set for a catastrophic 3 degrees of warming. Indeed, to state that shareholder value has failed—even on its own, efficiency-centered terms—is to state the obvious.
“From this perspective, what follows is a structure of corporate governance centered on protecting comparatively vulnerable minority shareholders against “expropriation” by insiders—namely majority shareholders, managers, and workers. As structurally weak stakeholders with skin in the game, the theory has it, shareholders need strong legal and regulatory protection, as well as extraordinary privilege with respect to the corporation’s governance and profits.”
“Together, these two hallmarks of asset manager capitalism amount to an alluring promise. As strong and universal shareholders, asset managers are structurally incentivized to internalize the negative externalities that were part and parcel of the profit-maximizing calculus of smaller, selectively invested shareholders. Rather than seeking to establish the dominance of a particular firm or industry in which an investor has placed her bets, universal owners strive for consistent and stable long-term growth […]”
“the trouble with the rise of the asset manager is not that these investors are particularly short-termist; rather, it’s that there is an agency problem. While beneficiaries seek a maximized return, asset managers seek higher fee incomes, and their corollary: greater assets under management. This agency problem creates a stumbling block for the alluring promise of universal ownership in various ways.”
“In 2020, a stunning increase of asset valuations contributed a $29 billion gross revenue increase to the asset management industry, compared to a modest $5 billion from net inflows of new money. For asset managers, rising asset prices are the golden goose.”

No wonder the valuations rise and rise. They like bigger numbers. Like Golgafrinchans. They have more than everyone else. What for? They have more. More is better.

“What’s wrong with BlackRock lobbying for expansionary monetary policy? The core problem, in contrast to the Monks and Minow, is that everybody is not a shareholder. To the contrary, as shown in Figure 2, half of all directly held stocks and mutual fund shares are held by the richest 1 percent of US households. The bottom half of households has virtually no equity investments at all, whether direct or via retirement plans.

So asset appreciation is only good for those who own assets? Economics is so mysterious and complex.

“[…] the bill nonetheless offers a wholly inadequate commitment of public funds to climate and other infrastructural investment, instead explicitly leaning on a climate-tailored implementation of Daniela Gabor’s “Wall Street Consensus”—shepherding in private capital, on favorable terms, backstopped by implicit and explicit government support. It’s an archetypal “socialize risk, privatize reward” model for policy, handed to eagerly awaiting asset management giants.
“What we do know is that we no longer live in a Berle-Means-Jensen-Meckling world. Beyond having failed (on its own and other terms), the corporate governance regime of shareholder value has had its justifying assumptions utterly upended. The rise of BlackRock, Vanguard, and their competitors has ushered in a new regime—a combination of concentration, control, diversification and “disinterestedness” that is without historical precedent.

Right-clickers vs. the monkey JPG owners by Ryan Broderick (Garbage Day)

“I imagine the hope with this is that if ConstitutionDAO can actually raise the $20 million needed to buy the Constitution from Sothebys, the buzz around the project will help them sell the Constitution for even more down the road. And because the market has been so crazy this year post-GameStop pump, it’s not a totally crazy bet. Of course, unless the market completely bottoms out, leaving everyone with a bunch of fake money they bought to be fake shareholders of a Discord server and a random historical artifact.

The Trucker Shortage: Why Don’t We Let the Market Work? by Dean Baker (CounterPunch)

Suppose that truckers got $150,000 a year and worked something like regular 40-hour weeks, and weren’t forced to drive unsafe trucks in unsafe conditions? Does anyone think the industry would have a hard time finding enough people to work as truckers? (Actually, if truckers’ pay had kept pace with productivity growth over the last four decades it would be somewhere around $150k a year today.)

“The point here is that the trucker shortage is overwhelmingly a problem of inadequate pay. This is what the market is telling us. But rather than listen to the market, we get a grand tour of other possible solutions. Why does the NYT have such a hard time listening to the market?

“This seems like just another case of prejudice against workers who do not have college degrees. It’s true that higher pay for truckers would get passed on in the prices of a wide range of goods. But the $300,000 plus average pay of physicians gets passed on to us in the cost of our health care insurance.


“In these, and other areas, we have policies that make a relatively small number of people very wealthy, but that is not supposed to concern us. But the idea that we might have to pay truck drivers something like $150,000 a year, and therefore incur higher costs, is somehow intolerable.”

Public Policy & Politics

Noam Chomsky: Ending Climate Change “Has to Come From Mass Popular Action,” Not Politicians by Poyâ Pâkzâd & Benjamin Magnussen (Jacobin)

“Take the sanctions against Cuba, the oldest ones. The entire world opposes them. The vote at the United Nations, the last one, was 184 to 2. Israel goes along with the United States — it’s a client state, so it has to. The rest of the world says no, but they all abide by the sanctions, because US sanctions are third-party sanctions. They tell others, “If you don’t abide by them, we throw you out of the international financial system.” And you have other punishments. The world is basically the mafia, and the Godfather gives the orders, and others obey, whether they like it or not. It’s reality, [it’s] not political science.”

“First of all, why is China a threat? A good recent statement about this [was made] by the distinguished international diplomat, former prime minister of Australia Paul Keating. He said, “What’s the China threat? Well, here’s a country that raised 20 percent of the world’s population from poverty moving on to become a functioning state.” It’s moving forward in the economy, [but] it’s independent of the United States — that’s the China threat. The China threat, he said, is China’s existence. That can’t be tolerated.

“Let’s go back to the mafia. The Godfather does not accept centers of power that don’t follow the rules, so China’s a threat. It’s not a military threat. The military threat is against China. China is ringed with US bases with nuclear armed missiles, right offshore, aimed at China. It’s China that’s under threat, not the United States. The United States is under threat from a potential rival that doesn’t follow orders. That’s the threat.”

“I should mention on the side that the United States is the only maritime power that does not ratify the Law of the Sea. [The] United States doesn’t ratify international conventions — that’s an interference with its sovereignty. Because the Godfather doesn’t accept that.
“France had already made a deal with Australia to send conventional submarines. The United States did not even notify France that it was being abrogated by the US-Australia deal to send advanced nuclear submarines. So, naturally, [Emmanuel] Macron was pretty upset. It’s a blow to French industry, a serious blow. They weren’t even informed. There’s a message there. It tells the European Union, “Here is your role in world affairs. If we need you, we’ll ask you to do something. If we want to do something, we won’t even bother notifying you. You’re vassals.”
“The message basically was, “We have two choices.” We can either start right now cutting back on fossil fuel use, [and] do it systematically every year, until we phase them out by mid-century. That’s one choice. The other choice is cataclysm. The end of organized human life on earth. Not immediately — we’ll just reach irreversible tipping points, and it goes on to disaster. Those are the options.”
The whole neoliberal period was basically class war. It had nothing to do with the markets or anything else. Just class war. This is another form. Do we want to hand the future of our children and grandchildren to elements that want to make as much profit as possible and then don’t care what happens tomorrow? That’s one choice. The other choice is to move onto a livable and better world.”
“[…] the mainstream [definition of anarchy] has been based on a very simple principle: any form of hierarchy and domination is illegitimate, unless it can justify itself. It has a burden of proof. Sometimes that burden of proof can be me — very rarely. If it can’t be met, dismantle it, and turn it into a more free, participatory, cooperative society.”
“It will be a highly organized society. There can be a lot of planning about how we should distribute resources, what our policies ought to be. It could be, or should be, international in scope. So, a rich and complex organization based on popular and democratic control, meeting the condition that any form of hierarchy that can’t justify itself has to be dismantled in favor of more freedom.
“Now it’s supposed to be a wonderful thing if you can subordinate yourself to a master for most of your working life. It’s called “getting a job.” It was considered a horrible attack on human dignity for millennia. Millennia, literally, up through the nineteenth century. It’s taken a lot of work to impose on people the idea that it’s a wonderful thing to spend your waking hours following somebody else’s orders.

“Well, not only the anarchists but even the socialist movement took that as a slogan. “Workers’ control of enterprises” was the basis for the traditional socialist movement, [but] it’s long changed from that. But I think all these ideas can be reconstructed quickly — they are [already], to a certain extent. There are worker-controlled enterprises, cooperatives, localist initiatives developing, which are pieces of a more free and just society, which would follow general anarchist principles.

Unfortunately, this cannot happen in time to deal with our immediate crises. The timescales are wrong. The immediate crises [are] crises of survival [and] will have to be dealt with within more or less the existing institutions. They can be modified, but they are not going to be fundamentally changed in time to deal with the crises. That doesn’t mean that you stop working on the long term. You do it, and it changes consciousness, changes understanding, and builds elements of freedom within a highly repressive society. It can be done. But the immediate crises are going to require working with the institutions that exist. We’re stuck with that.”

Is This the End of the Unreformable Democratic Party? by Michael Hudson (CounterPunch)

“Bernie Sanders sounded exasperated on election-day Tuesday when he explained that this $400 billion giveaway to the wealthiest 5 percent was so large, that “the top 1% would pay lower taxes after passage of the Build Back Better plan than they did after the Trump tax cut in 2017. This is beyond unacceptable.””
“President Biden is blaming Progressives for “blocking” the program by trying to preserve the policies that most voters actually want, and which he himself ran on in his presidential campaign a year ago.”

Because Biden is, and always has been, a venal and hypocritical asshole.

“Aristotle[‘s] description of democracy: Many states have constitutions that are democratic in form, he wrote, but actually are oligarchies. The reason, he explained, is that democracies tend to evolve into oligarchies as a result of the increasing concentration and polarization of wealth. That gives the leading families control of the political system. (In his schema, oligarchies aim at making themselves hereditary aristocracies.)”
“In the United States, to be sure, all votes on election day are counted equally, but in practice the One Percent limit the range of policies that can be voted on and then implemented. The first problem is how to be nominated in the first place and vie with rivals in the political primaries. In America, success requires support from the Donor Class.”

The World’s Best Mayor Is a French Communist: An interview with Philippe Rio by David Broder (Jacobin)

“today the working class has changed. But I can assure you, the poor are still here. When people ask me, what’s the difference, Philippe, between when you lived in Grande Borne forty years ago and today, I say it’s that then there was 5 percent unemployment and now the figure is 50 percent. With its renewed hunger to capture wealth, liberal society is also creating poor workers.
France is the fifth-leading economic power but has the thirtieth or fortieth justice system in the world. The youth justice system is a disaster.”
“Here, too, we’ve had to take responsibility. The Grigny 2 housing trust — with five thousand dwellings and seventeen thousand inhabitants — had its heating and hot water cut off because there were unpaid bills. It relied on natural gas, which depends on fluctuations in world prices, just like how it’s going up today. We couldn’t control anything. So we made an alternative, geothermal project, which is 100 percent publicly owned. We got heat from two kilometers under our feet. We cut the bills by 25 percent and saved the planet fifteen thousand tons of CO2 in one year.

As America Falls Apart, Profits Soar by Matt Taibbi (TK News)

“On the day the Rittenhouse trial began, the financial data firm FactSet released an eyebrow-raising report about the Covid-19 economy.

“The firm noted that companies in the S&P 500 were set to post a net 12.9% profit in the third quarter of 2021. They pointed out this was the second-highest result since the firm began tracking the number in 2008.

“The only better result? The previous quarter, i.e. Q2 2021, when net profits sat at 13.1% overall. These results track with the true great story of the pandemic era, which not-so-mysteriously hasn’t made the news much, while Americans have been tearing each other’s faces off over issues like race and vaccination policy: the massive widening of our already-obscene wealth gap.

“The economic news at the top hasn’t just been good, it’s been record-setting good, during a time of severe cultural crisis.”

“The concept wasn’t hard to understand: leaders were promoting unifying myths to keep the population satiated, dumb, and focused on their primary roles as workers and shoppers.

“In the Trump era, all this has been turned upside down. There’s actually more depraved, dishonest propaganda than before, but the new legends are explicitly anti-unifying and anti-patriotic. The people who run this country seem less invested than ever in maintaining anything like social cohesion, maybe because they mostly live in wealth archipelagoes that might as well be separate nations (if they even live in America at all).”

“Keeping the volk at each other’s throats instead of pitchforking the aristocrats is an old game, one that’s now gone digital and works better than ever. That might be worth remembering after the coming verdict, and ahead of whatever other hyper-publicized panic comes down the pipeline next.”

Elizabeth Holmes Swindled Henry Kissinger, and We’re Not Complaining by Ben Burgis (Jacobin)

Elizabeth Holmes seems to know about as much about medical science as I do. When asked how her supposedly miraculous blood testing machine actually worked, her reply was, and I swear I’m not making this up, “A chemistry is performed so that a chemical reaction occurs and generates a signal from the chemical interaction with the sample, which is translated into a result, which is then reviewed by certified laboratory personnel.””
“Still, whatever Holmes’s many failings as a medical researcher, a boss, an explainer of chemistry, and a human being, there was at least one thing she did very well. In the domain of talking rich people out of their money, she may actually be one of the world’s greatest geniuses.

By Backing a Huge Tax Giveaway to the Rich, Democrats Are Giving the GOP a Perfect Midterm Gift by David Sirota (Jacobin)

“In all, raising the SALT cap would result in a BBB bill in which two-thirds of Americans who make over $1 million get a tax cut, and the average reduction for those households would be more than $16,000.”
““Despite what its promoters say, raising the cap to $80,000 would provide almost no benefit for middle-income households,” wrote Howard Gleckman of Forbes, a publication that is not exactly a bastion of anti-capitalist ideology. “It would reduce their 2021 taxes by an average of only $20. Even those making between $175,00 and $250,000 would get a tax cut of just over $400 or about 0.2 percent of after-tax income. By contrast, the higher SALT cap would boost after-tax incomes by 1.2 percent for those making between about $370,000 and $870,000.””


“In a nation where 87 percent of people already make too little to itemize their tax returns and are therefore not eligible for any SALT deductions, Democrats’ whole campaign is designed to confuse and distract from all the data showing that repealing the SALT cap would be a more regressive policy than Donald Trump’s 2017 tax cuts and would exacerbate racial and economic inequality.”
“Indeed, even in the Garden State, 92 percent of a full cap repeal would flow to the richest 15 percent of the population, leaving everyone else in the state with almost no benefit at all.”

Are we talking about a “full cap repeal” or are we talking about raising the cap to $80,000? I’m wondering if Sirota is being sneaky with his numbers and formulation here. If there is a difference, then it’s misleading to discuss figures related to a “full cap repeal” when that’s not what’s on the table. Don’t fight misinformation with misinformation.

“First and foremost, a SALT cap repeal is a precision-targeted enrichment scheme for the corporate attorneys, hedge fund managers, business consultants, real estate investors, and other affluent caricatures who host and attend Democratic fundraisers in wealthy enclaves like Easthampton, New York, Short Hills, New Jersey, and Laguna Beach, California.”

As expected, the Democrats swept into office with promises on their lips, none of which materialized and,

“The year is now ending with Democrats’ poll numbers plummeting as they make headlines reneging on those promises and instead demanding SALT tax cuts for affluent locales.

This is a dream scenario for Republicans. Even though they have nothing better to offer, they are getting another political bailout from a Democratic Party that is still captured by its affluent donors.”

Is The Jury Next To Go? by Scott H. Greenfield (Simple Justice)

“There is a legitimate argument that many potential black jurors are stricken, whether for cause or by peremptory strike, based upon their experience with police. Do they harbor animus toward cops? Will they consider the testimony of a police officer fairly and not presume they’re lying scum? This is the “black experience” problem, where the treatment of black people by police comes back to bite the prosecution in the butt. Why, the argument goes, should black people be stricken from the jury because their life experience with police is the product of being treated like garbage? Why should cops get to cause black people to be legitimately stricken because they’re racist toward them? It’s a damn strong argument.

“But here, the argument takes a blind leap over the head of the defendant, or in the Arbery case, the three white defendants accused of his murder, because the jury of 11 white people and one black person denies Arbery a jury composed of members of his race. Whether that’s circumstance or deliberate, it’s irrelevant. There is no right to have a jury that looks like Arbery, and to do so would ignore that the Sixth Amendment secures that right only for the defendants, not the victim.

Rittenhouse Verdict: Who Sent What Message? by Scott H. Greenfield (Simple Justice)

“There is nothing to celebrate when people die or were harmed when it didn’t need to happen. Rittenhouse didn’t need to go to Kenosha, didn’t need to bring a long gun. Protesters didn’t need to go out to protest, and rioters didn’t need to burn down buildings in Kenosha. But these things happened, and they were resolved with the mechanism a society uses to determine whether a crime has been committed, as it should be. There is absolutely nothing about the trial of Kyle Rittenhouse that suggests the same outcome should someone else do something similar. There is no other message to be sent than the jury has spoken in one case.”

From a comment below the article:

“The people who use the word “precedent” to describe a jury verdict in a self-defense case just don’t get it.”

Journalism & Media

Star Trek: Discovery is tearing the streaming world apart by Chris Stokel-Walker (Ars Technica)

The average American household accesses eight streaming and video-on-demand services in a given week, according to data gathered by technology research company Omdia—though that includes free catch-up services and websites like YouTube. In the UK, the average is nearer six to seven, and in mainland Europe, five to six.”

Op-Ed: We Don’t Need Due Process For People We Know Are Guilty by Garth Strudelfudd (The Babylon Bee)

“Due process. It seems like a great idea. Everyone gets their day in court, and the rules apply equally to all. But now with the Kyle Rittenhouse trial, we’ve seen what a terrible idea due process is when you know someone is guilty and just want him to be thrown straight into prison.

“Kyle Rittenhouse is a murder-crazed shooty person who hates people of color so much that he shot white people (since white is the combination of all colors). No one in my bubble disputes this. Yet, there he is getting a day in court, where the same rules of fairness we want to be applied to oppressed people get applied to him.”

“It is a mockery of justice to see due process used for someone who we all know is guilty. I mean, we saw pictures of Rittenhouse holding an AR-15; that by itself should be enough to send anyone to prison forever. In addition, he crossed state lines. Let me repeat that: STATE LINES. Who would brazenly do such a thing, except to cause murder and chaos?”
“In the future, if blue checks on Twitter declare you guilty of murder, you can still have a trial, but no more due process. The judge has to hate you and yell at you the whole time […]”

When the traffic firehose is pointed at you by Ryan Broderick (Garbage Day)

“[…] the kinds of collaborative posts that “Amish bitch” exemplifies have limited means of creation and circulation. Basically, a Tumblr user writes a dumb thing, hundreds of other users pile on, creating a crowdsourced piece of internet content with hundreds of new entry points. It’s kind of like the text-equivalent of a multi-layer TikTok remix. Compare that to the contextual paucity of a retweet, a retweeted QRT, or a set of screenshots of a thread. Both a Twitter thread and a reblog chain both work off of the basic digital affordance of “I am responding to you publicly,” but diverge wildly after that. The fact that a line of responses can congeal organically and then be shared as a memetic unit in its totality is the vital factor here. Yes, this is all very complex. I know that I’m talking about a Tumblr post. But stick with me here.”

On the Constitution DAO (Decentralized Autonomous Organization),

“As goofy as this whole project was, I do think it’s fitting that 2021 started with the GameStop pump and end with a crypto Discord actively bidding on Sotheby’s.”

Citing Anil Dash on DAOs,

““Many are just doing this for the meme value. They’ve gotten crypto rich and don’t mind burning some of their riches on a stunt. But a substantial cohort are very sincere about wanting to learn how self-organizing could work; the rhetoric closely matches that of the early web,” he wrote. “I’m curious to see how it plays out. I also hope people realize they can still just do an old-fashioned crowdfunding for things they care about. We’ll know how it goes soon.””

Yeah, it’s fucking designed to because it’s almost certainly a scam. If there are people who earnestly believe in this as a way forward for … anything … then they are the ones being scammed. The cynical fucks who just see it as a money-making opportunity in which they relieve credulous fools of their money in an unregulated market where doing that is actually legal—they’re going to clean up, as usual.

The Rittenhouse Verdict is Only Shocking if You Followed the Last Year of Terrible Reporting by Matt Taibbi (TK News)

“Because of all of these simple factual misconceptions — that Rittenhouse was a militia member and a white supremacist who’d traveled a great distance to a town to which he had no connection, then fired first and indiscriminately — analysts not only pre-judged Rittenhouse’s guilt, but offered advance explanations for any possible acquittal.
“Now Rittenhouse has been found innocent, and surprise, surprise, the immediate reaction is that it can only be explained by white supremacy. To a degree, I don’t even blame people who’ve come to this conclusion, because it’s all they’ve heard for a year: Rittenhouse is a racist murderer who went way out of his way to shoot innocent people, and was given a pass by an evil system.”

Science & Nature

Die Zerstörung von maiLab by Sarah Bosetti (ZDF)

This is a wonderful, sarcastic, extremely literate and loquacious commentary. In German. Six minutes.

Scientists extend and straighten iconic climate “hockey stick” by Howard Lee (Ars Technica)

“For the new study, scientists carefully vetted over 500 proxy records from oceans around the world; the data shows the fossilized remains of plankton and microbes in sediments where the age is known from radiocarbon dating. Researchers then used statistical methods to calculate sea surface temperatures from the chemical properties of those remains. “We spent seven years developing the models for the different kinds of marine temperature proxies, incorporating knowledge from biology and geochemistry and using the best statistical practice,” explained coauthor Dr. Jessica Tierney of the University of Arizona and leader of the lab in which this research was conducted.”
“Osman and colleagues concluded that most of the conundrum’s resolution is due to the sophisticated handling of geographic unevenness in their data compared to prior work. “Older reconstructions are just binned latitudinal averages,” said Tierney. “[That] method has a downside in that it doesn’t account for temperature changes in regions that are unsampled.”

Curves and Surfaces by Bartosz Ciechanowski

This is an excellent soup-to-nuts, interactive lesson for curves and surfaces, splines, control points, Bezier curves (cubic, etc.), NURBS, cubic B‑spline curves, cubic Chaikin subdivision surfaces, Catmull-Clark subdivision schemes. It takes quite a while to get through it—and I didn’t dwell on all of the math—but it’s very interesting and probably well-worth revisiting. It’s good to know what underlies the tools we use. There are dozens upon dozens of interactive exhibits for every step of the way, from 2D to 3D.

“When the surface curves, its normal directions spread out. In simple terms, a curved mirror “sees” more of its environment but since the mirror’s area stays the same, the reflection of the sphere has to shrink to fit in.

“When we bend the surface we’re actually changing the curvature of the profile of the surface. Roughly speaking, curvature defines how quickly a curve changes its direction.”

“After just a few steps, the subdivision curve matches the B‑spline curve in the background. This is a fantastic news, as it gives us yet another way to think about simple quadratic B‑spline curves – they’re obtained by repeatedly cutting off corners of their control polygon.
“The crucial invention here is that nothing about the rules for creating the new subdivided control mesh relied on how the mesh itself is connected! If you recall, NURBS surfaces were limited to rectangular grids. Catmull-Clark subdivision scheme also works when the vertices of the initial mesh have different number of neighbors”

These Mighty Shorebirds Keep Breaking Flight Records—And You Can Follow Along by Lauren Leffer (Audubon)

“On September 28, one small bird completed a very long flight. An adult, male Bar-tailed Godwit, known by its tag number 4BBRW, touched down in New South Wales, Australia, after more than 8,100 miles in transit from Alaska —flapping its wings for 239 hours without rest, and setting the world record for the longest continual flight by any land bird by distance. And 4BBRW isn’t even done yet. In the next few days, the Godwit is expected to end its southbound migration in New Zealand after its well-earned island stopover […]”

Art & Literature

David Graeber’s Possible Worlds by Molly Fischer (New York Intelligencer)

“Graeber had been working on a short essay about COVID that was published after his death. The pandemic was “a confrontation with the actual reality of human life,” he wrote. “Which is that we are a collection of fragile beings taking care of one another, and that those who do the lion’s share of this care work that keeps us alive are overtaxed, underpaid, and daily humiliated.” Surely it was the moment to stop taking such a state of affairs for granted, he wrote. “Why don’t we stop treating it as entirely normal that the more obviously one’s work benefits others, the less one is likely to be paid for it; or insisting that financial markets are the best way to direct long-term investment even as they are propelling us to destroy most life on Earth?””

Philosophy & Sociology

Early Civilizations Had It All Figured Out by Gideon Lewis-Kraus (The New Yorker)

“For Graeber and Wengrow, this basic story, whether relayed in a triumphal or a defeatist register, is itself a trap. If we accept that the rise of agriculture meant the rise of the state—of political élites and intricate structures of power—then all we can do is tinker around the edges. Even if we regard the Paleolithic era as a garden paradise, we know that our reëntry is forever barred. For one thing, the requirements of hunting and gathering could support only some trivial fraction of the earth’s current population. A life under government control now seems inescapable.”
“[…] disavows the intellectual trappings of a knowable arc, a linear structure, and internal necessity. As a stab at grandeur stripped of grandiosity, the book rejects the logic of technological or ecological determinism, structuring its narrative around our ancestors’ improvisatory responses to the challenges of happenstance. The result is an almost hallucinatory vision of the human epic as a series of idiosyncratic digressions. It is the story of how we made it up as we went along—of how things could have been different and, perhaps, still might be.
“In the Pacific Northwest, men of rank among the Kwakiutl held lavish, greasy potlatches and took war captives as slaves; their neighbors to the south of the Klamath River, the Yurok, prized restraint and self-denial, and committed themselves to modes of subsistence that rendered slavery, which they found morally repugnant, unnecessary.
“At a certain point, however, the people of Teotihuacan decided against investing in more fancy villas. Instead, Graeber and Wengrow write, “the citizens embarked on a remarkable project of urban renewal, supplying high-quality apartments for nearly all the city’s population, regardless of wealth or status.” They accomplished all of this without wheeled vehicles, sailing ships, animal-powered traction, or advanced metallurgy. Perhaps most important was that, although they were in contact with the monarchical Mayan societies nearby, the people of Teotihuacan flourished for some three centuries without submitting to the rule of anything like a king.
“Though Graeber and Wengrow have marshalled a vast amount of archeological evidence, they acknowledge that much of what anyone has to say about ancient societies is speculative. Their hope is that, even if some of their examples remain dubious, the accumulated weight of recent findings—and the more inventive assortment of political organization they imply—establishes the glib tendentiousness of Big History. As they put it, “We are at least trying to see what happens when we drop the teleological habit of thought.”
“Critiques of grand narratives have been important to the modern self-image of these fields—in part as penance for having once been happy to serve the priorities of empire, peddling “civilization” as a gift to the “primitives.” One consequence, however, is that wholesale synthetic accounts of human history tend to be written in the extravagantly roughshod mode of Harari’s “Sapiens” or Jared Diamond’s “Guns, Germs, and Steel.””
“[…] they worry that if people aren’t offered an alternative framework they will still default to some version of the pernicious cultural-evolution myth—and accept that the familiar hierarchies of governance are simply the price of sophistication.
Graeber and Wengrow hope that, once we grasp how ancient mega-sites (in Ukraine or in Jomon-era Japan) could grow large and manifold without a literate bureaucracy, or the way early literate societies (Uruk, in Mesopotamia) might have managed the trick of participatory self-governance, we might renew and expand our own cramped notions of what’s politically tenable.
“When we speak of the onset of social inequality, we’re accepting the idea that real freedom is the plaything of children. The species grew up, and grew out of it. Peter Thiel wonders why we don’t yet live in the future of our dreams. Graeber and Wengrow think the first step forward is a reminder of the past we deserve.

Why Work Need Not Be Your “Passion” by Scott H. Greenfield (Simple Justice)

“It looks like work/life balance might be coming back into fashion. Of course, Cech includes a heavy dose of society subsidizing the unpleasant parts of life in order to allow people to work as little as possible to free them up for the things that bring them joy. It’s unclear how all those people doing the work to provide society with the funds to pay for your joy are going to feel about it, but then, isn’t your passion really all about you?”

This is really such a poisonous attitude, especially from a lawyer who is clearly doing a job he enjoys and who gets to spend his free time writing on a blog he truly enjoys. I think he is—as so many do—focusing on the vocal minority who maybe shouldn’t be the first in line to complain about their jobs, but they do anyway. Boring jobs are enervating.

He goes on,

“No longer are people concerned with such banal matters as eating, or feeding their children, or paying their rent.”

I think this attitude is so poisonous because it suggests that anyone who doesn’t want to spend 8-10 hours of their day doing something stultifying is a shirker, a mooch. That’s not counting the commute, for most people. Greenfield fails to imagine a world where we’re not forcing 80-90% of the population into misery so that the top 10% has a great life.

“Let those other poor suckers work hard at their drudgery so the money they earn can be used to free up your life for your passion. Sure, the dirty jobs that keep society functioning need to be done, but not by you”

And not by you, either—at least not for the money they’re offering. I think what we should consider is a realignment of compensation with level of satisfaction, no?


Notes on Web3 by Robin Sloan

“Many Web3 boost­ers see them­selves as disruptors, but “tokenize all the things” is noth­ing if not an obe­di­ent con­tin­u­a­tion of “market-ize all the things”, the cam­paign started in the 1970s, hugely suc­cessful, ongoing. In a way, the World Wide Web was the real rupture — “Where … is the money?”—which Web 2.0 smoothed over and Web3 now attempts to seal totally.
“I feel like this sim­ple premise is often lost in the haze: the Ethereum Vir­tual Machine, hum­ming heart of Web3, is a com­puter that charges you many dol­lars to exe­cute a very small pro­gram very slowly. It does so in an envi­ron­ment with spe­cial properties, and in some cases, those prop­er­ties are worth the expense. In others … it’s like run­ning your web­site on a TRS-80 with a coin slot. A”

Class 1 / Class 2 Problems by Kevin Kelly (The Technium)

“Crypto is hard to use, easy to trip up, biased to early adopters, an energy hog, and of marginal utility except to make money. But all these problems will be overcome by entrepreneurs. Someday blockchain will be ubiquitous and boring. It will be perfected and its wide-spread adoption will enable many thousands of new types of organizations and relationships that we can’t even imagine today. Blockchain tech could unleash collaborations of several million members working on one project in real time, or orgs that are far more leaderless than today.

What the actual fuck are you talking about? How is blockchain going to do that? I mean, I’d understand if you said CRDTs (Wikipedia), but blockchain? What does a simple read-only database have to do with collaborative software?

Honestly, crypto-enthusiasts are so frustrating. They’re so focused on establishing a non-fiat currency that they think they have to sell it as perfect—even though it’s pretty much the first idea anyone thought of. So, as usual: something must be done; this is something; we must do this.

It’s the same when you try to raise concerns about the energy use. It’s enormous and wasteful relative to what we get out of it. Everything else in the world uses less and less power and Bitcoin uses as much power as a medium-sized country. There must be a better way to establish a decentralized non-fiat currency, if that’s what you want. But, instead of acknowledging the problem and trying address it with improved technology, crypto-enthusiasts simply double down on the original idea and try to disprove that it uses a lot of power. Most of them, like the guy cited above, literally have no idea what the blockchain is or can do—or how it could be improved. It just is. It’s like the Bible at this point, for most of these people. You can’t doubt it. You can barely even look at it. And you certainly couldn’t improve on it. Shut your mouth. The same people who think that technology can solve anything—e.g. climate change—are so afraid of trying to improve crypto technology that they deny any problems exist rather than having to revisit it.


Rust Is The Future of JavaScript Infrastructure by Lee Robinson

Deno, created in 2018, is a simple, modern, and secure runtime for JavaScript and TypeScript that uses V8 and is built with Rust. It’s an attempt to replace Node.js, written by the original creators of Node.js. While it was created in 2018, it didn’t hit v1.0 until May 2020. Deno’s linter, code formatter, and docs generator are built using SWC.”
Rome, created in August 2020, is a linter, compiler, bundler, test runner, and more, for JavaScript, TypeScript, HTML, JSON, Markdown, and CSS. They aim to replace and unify the entire frontend development toolchain. It’s created by Sebastian, who also created Babel.
“dprint, built on SWC, is a 30x faster code formatting replacement for Prettier.”

RavenDB 5.3 Features: Incremental time series, throttling and rate limits by Oren Eini (RavenDB Blog)

“It is trivial in RavenDB to implement a counter that would track the overall views on a post. It will also handle concurrency, distributing data between nodes, everything that needs to be handled. So why can’t I use that?

“It turns out that I typically want to know more than just the total number of views on the post, I want to know when they happened. Counters are only a partial answer for that.

“That is why incremental time series were created. They are here to marry the ability of time series to track a value over time and the distributed counters ability to aggregate information concurrently and in a safe distributed manner.

The inside story of the outside investigation of SoftRAM 95 by Raymond Chen (Microsoft Blogs)

The hope is that even though you took away a bunch of fast pages, you replaced them with a larger number of medium-speed pages, thereby lowering the number of accesses to slow pages. In the above example, we’re assuming a compression ratio of 2x, so we took away 8 fast pages and turned them into 16 medium-speed pages.”
The whole compression architecture was implemented, with a stub compression function that did no compression, presumably with the idea that “Okay, and then we’ll put an awesome compression function here, but for now, we’ll just use this stub function so we can validate our design.” But they ran out of time and shipped the stub.
“When they added code to implement the compression buffer, they didn’t use critical sections or any other synchronization primitives to protect the data structures. If two threads started paging at the same time, the driver corrupted its data structures due to concurrency. The next time the driver went to uncompress the data for a page, it got confused and produced the wrong memory, and that’s why Windows 95 was crashing. They were inadvertently simulating a broken hard drive.

Lesser Known PostgreSQL Features by Haki Benita

This article is chock full of really interesting bits:

The synonyms “feature” is an elegant way of implementing zero-downtime migration.

“I used to think that synonyms are a code smell that should be avoided, but over time I found a few valid use cases for when they are useful. One of those use cases are zero downtime migrations.

“When you are making changes to a table on a live system, you often need to support both the new and the old version of the application at the same time. This poses a challenge, because each version of the application expects the table to have a different structure.

“Take for example a migration to remove a column from a table. While the migration is running, the old version of the application is active, and it expects the column to exist in the table, so you can’t simply remove it. One way to deal with this is to release the new version in two stages − the first ignores the field, and the second removes it.

If however, you need to make the change in a single release, you can provide the old version with a view of the table that includes the column, and only then remove it. For that, you can use a “synonym” […]

“The application is connected to database db with the user app. You want to remove the column active, but the application is using this column. To safely apply the migration you need to “fool” the user app into thinking the column is still there while the old version is active […]

Local-first software by Martin Kleppmann (Ink & Switch)

“In the pursuit of better tools we moved many applications to the cloud. Cloud software is in many regards superior to “old-fashioned” software: it offers collaborative, always-up-to-date applications, accessible from anywhere in the world. We no longer worry about what software version we are running, or what machine a file lives on. However, in the cloud, ownership of data is vested in the servers, not the users, and so we became borrowers of our own data. The documents created in cloud apps are destined to disappear when the creators of those services cease to maintain them. Cloud services defy long-term preservation. No Wayback Machine can restore a sunsetted web application. The Internet Archive cannot preserve your Google Docs.
“Today it is easy to create a web application in which the server takes ownership of all the data. But it is too hard to build collaborative software that respects users’ ownership and agency. In order to shift the balance, we need to improve the tools for developing local-first software.

Data Laced with History: Causal Trees & Operational CRDTs by Alexei Baboulevitch (Archagon)

“By the laws of physics, remote machines are incapable of sending and receiving their changes as soon as they happen, forcing them to participate in busy conversations with their peers to establish a sensible local take on the global data. But a CRDT is always able to merge with its revisions, past or future. This means that all the computation can be pushed to the edges of the network, transforming the tangle of devices, protocols, and connections at the center of it all into a dumb, hot-swappable “transport cloud”. As long as a document finds its way from device A to device B, sync will succeed. In this world, data has primacy and everything else has to orbit around it!
“In contrast to all the other CRDTs I’d been looking into, the design presented in Victor Grishchenko’s brilliant paper was simultaneously clean, performant, and consequential. Instead of dense layers of theory and labyrinthine data structures, everything was centered around the idea of atomic, immutable, metadata-tagged, and causally-linked operations, stored in low-level data structures and directly usable as the data they represented. From these attributes, entire classes of features followed.”
“Much like Causal Trees, ORDTs are assembled out of atomic, immutable, uniquely-identified and timestamped “operations” which are arranged in a basic container structure. (For clarity, I’m going to be referring to this container as the structured log of the ORDT.) Each operation represents an atomic change to the data while simultaneously functioning as the unit of data resultant from that action. This crucial event–data duality means that an ORDT can be understood as either a conventional data structure in which each unit of data has been augmented with event metadata; or alternatively, as an event log of atomic actions ordered to resemble its output data structure for ease of execution […]”
“The decomposition of data structures into tagged units of atomic change feels like one of those rare foundational abstractions that could clarify an entire field of study.”
“Even gapless causal order, which is critical for convergence in most CRDTs, becomes only a minor concern, since events missing their causal ancestors could simply be ignored on evaluation. (RON treats any causally-isolated segments of an ORDT as “patches” that can be applied to the main ORDT as soon as their requisite ancestors arrive.) Serializing and deserializing these structures is as simple as reading and writing streams of operations, making the format perfectly suited for persisting documents to disk. Since operations are simply data, it’s easy to selectively filter or divide them using version vectors. (Uses for this include removing changes from certain users, creating delta patches, viewing past revisions, and setting garbage collection baselines.)
“[…] therefore posit that at this point in the pipeline, there ought to be a simpler arranger step. This function would perform the same sort of merge and integration as the reducer/effect functions, but it wouldn’t actually remove or modify any of the operations. Instead of happening implicitly, the cleanup step would be explicitly invoked as part of the garbage collection routine, when space actually needs to be reclaimed.”
“The arranger/reducer/effect and the mapper/eval functions together form the two halves of the ORDT: one dealing with the memory layout of data, the other with its user-facing interpretation. The data half, as manifest in the structured log, needs to be ordered such that queries from the interface half remain performant.
“That’s all there is to the ORDT approach! Operations are piped in from local and remote sources, arranged in some sort of container, and then executed or queried directly to produce the output data. At a high level, ORDTs are delightfully simple to work with and reason about.
“ORDTs are meant to fill in for ordinary data structures, and sticking operations into a homogeneous container might lead to poor performance depending on the use case. For instance, many text editors now prefer to use the rope data type instead of simple arrays. With a RON-style frame, this transition would be uncomfortable: you’re stuck with the container you’re given, so you’d have to create an extraneous syncing step between the frame and secondary rope structure. But with an object-based ORDT, you could almost trivially switch out the internal data structure for a rope and be on your merry way.
“In ORDTs, garbage collection isn’t just a matter of removing “tombstone” operations or their equivalent. It’s also an opportunity to drop redundant operations, coalesce operations of the same kind, reduce the amount of excess metadata, and perform other kinds of cleanup that just wouldn’t be possible in a strictly immutable and homogeneous data structure. Although baseline selection has to be done very carefully to prevent remote sites from losing data, the compaction process itself is quite mechanical once the baseline is known. We can therefore work on these two problems in isolation.”
“In an available and partition-tolerant system system, is it possible to devise a selection scheme that always garbage collects without orphaning any operations? Logically speaking, no: if some site copies the ORDT from storage and then works on it in isolation, there’s no way the other sites will be able to take it into account when picking their baseline. However, if we require our system to only permit forks via request to an existing site, and also that all forked sites ACKs back to their origin site on successful initialization, then we would have enough constraints to make non-orphaning selection work.”
“But questions still remain. For instance: what do we do if a site simply stops editing and never returns to the network? It would at that point be impossible to set the baseline anywhere in the network past the last seen version vector from that site. Now some sort of timeout scheme has to be introduced, and I’m not sure this is possible to do in a truly partitioned system.
“Then, any site receiving new baselines in the sequence would be required to apply them in order. Upon receiving and executing a baseline, a site that had operations causally dependent on removed operations but not themselves included in the baseline would be obliged to either drop them or to add them to some sort of secondary “orphanage” ORDT.”
“In summary: while baseline operations are not commutative for every possible value, they can be made commutative with just a sprinkle of coordination. Either you ensure that a baseline does not leave orphaned operations (which requires some degree of knowledge about every site on the network), or you ensure that each new baseline is monotonically higher than the last (which requires a common point of synchronization).”
“Finally, remember that in many cases, “don’t worry about garbage collection” is also a viable option. Most collaborative documents aren’t meant to be edited in perpetuity, and assuming good faith on the part of all collaborators, it would be surprising if the amount of deleted content in a typical document ended up being more than 2 or 3 times its visible length.”
“[…] most of the work goes into figuring out how to atomize the data structure, define causal relationships, arrange operations inside the structured log, and optimize performance for critical methods. In other words: perfect engineering work, and not something that requires a PhD to manage!”
“[…] but many will be. Be sure to define your operations in absolute terms, with no implicit context. Causal links to other operations should provide all the context you need. Recall the issues caused by poorly-specified string index operations!”
“Instead of storing just the UUID in the site map, I also store the wall clock time at which the UUID was added. In the site map, these tuples are sorted first by time, then by UUID. Assuming that modern connected devices tend to have relatively accurate clocks (but not relying on this fact for correctness), we can ensure that new sites almost always get appended to the end of the ordered array and thus avoid shifting any of the existing UUIDs out of their previous spots. The only exception is when multiple sites happen to be added concurrently or when the wall clock on a site is significantly off.
“If an atom has priority, that atom and all its descendants get sorted ahead of any sibling subtrees in the parent’s causal block, even if it has a lower Lamport timestamp. (Put another way, a priority flag is simply another variable to be used in the sorting comparator, i.e. priority+timestamp+UUID.) This property gives us a lot of structural control, ensuring that, for instance, delete atoms hug their target atoms and never find themselves lost in the weave if concurrent insert operations vie for the same spot.
“Conventional event sourcing tends to define events at a fairly high level of abstraction: buy shirt instead of add 1 to shirt counter and subtract $10 from wallet. But as we saw in the CT section, this falls apart when the events are fine-grained and frequent, since performance can approach an abysmal O(n2) whenever history needs to be replayed. Zooming in and implementing the event sourcing pattern on the data structure level suddenly makes the whole thing come together. Most of the performance hotspots are fixed without affecting convergence, history-tracking, or expressiveness. It was fascinating to discover how a design pattern could arc so elegantly across architectural layers!