Gongol.com Archives: August 2025
August 2025 | ||||||
---|---|---|---|---|---|---|
1 | 2 | |||||
3 | 4 | 5 | 6 | 7 | 8 | 9 |
10 | 11 | 12 | 13 | 14 | 15 | 16 |
17 | 18 | 19 | 20 | 21 | 22 | 23 |
24 | 25 | 26 | 27 | 28 | 29 | 30 |
31 |
August 1, 2025
One source: To discover quite by accident something you didn't know you wanted to know. It's a delightful gift to stumble into learning something new without having sought it. But you have to practice being open to surprise and wonder, or else it won't happen.
August 3, 2025
An AI do-and-don't list for teachers - part 1
The school year typically begins in earnest in mid-August, and one of the hot topics for teachers this school year will be the reach of artificial intelligence tools both inside and outside the classroom. As with every other new piece of educational technology, there are good ways and bad ways to put it to use. Some recommendations follow. ■ DON'T tell students that artificial intelligence will replace the jobs they want. Technological change always causes changes to the labor force, but very few jobs are eliminated entirely. Telling young people their hoped-for careers will be replaced is discouraging -- and even labor economists rarely dare to predict the future with that much certainty. ■ DO tell students to look for opportunities to maximize the gap between what they have to give up to have a career and what they get back in reward. Encourage them to think about the path to a "dream job" as a series of opportunities that require trade-offs, like spending time in college, climbing a seniority ladder, or sacrificing other opportunities. On the other side, people are rewarded with more than just money: Social approval, work-life balance, respect, and many other factors are involved. The difference between what you get and what you give up to get it is what matters. ■ DON'T vilify all artificial intelligence tools equally. As with every technology, there are good and bad uses, which depend on the character of the user. (Even a kitchen knife can be used to lovingly prepare a meal or to commit cold-blooded murder.) ■ DO explain the limits of the usefulness of all technological tools, using real terms. Machine learning has the potential to do extraordinary things when large volumes of data are involved, as in medical research. But it also has the capacity to create terrible pain to real people when it's used to do truly ghoulish things like generating spammy obituaries. ■ DON'T promote unquestioning faith in the answers generated by artificial intelligence tools. Just because Google and other high-profile services are nudging people to use their AI tools doesn't make them more trustworthy or credible. It only means they're potentially profitable. ■ DO show students how to incorporate AI-generated content into a careful search process, including how to cross-reference among sources and how to independently verify what is often served with great authority. Show examples of dangerous and stupid errors that can and should be checked by humans, like obvious biographical and historical errors, scientific mistakes, or falsified reporting.
August 4, 2025
If you want a quotation to be remembered, misattribute it to someone famous. Winston Churchill "said" lots of things he didn't really say. Albert Einstein, too. And almost nobody has more misattributions than Abraham Lincoln, whose gift for language and exceptional place in history combine to make him a particularly good "source" for many a memorable bon mot. ■ One of those misattributions is "If I had five minutes to chop down a tree, I'd spend the first three sharpening my axe." It's a terrific proverb, really, even if Lincoln didn't say it. And it's particularly timely at the moment. ■ Conor Sen, an opinion columnist at Bloomberg, ignited a mild online controversy by declaring, "My gut feeling is that parents trying to make their kids elite at reading and writing as a backlash against our screen/video world are like teaching their kids the Dewey decimal system, microfiche, driving a stick shift." He added: "There's not going to be much interesting written content post-2020's." ■ Sometimes people are merely stirring the pot, especially online. But the comments seem to have been taken in earnest by others, and Sen himself has defended the take. ■ Putting aside what Sen intended by "elite at reading and writing" (probably a reference to a recent "sign of the times" article in a high-status publication), reading and writing generally are probably the best examples of "axe-sharpening" life skills that anyone can develop. They prepare the way for practically all other worthwhile endeavors. It's not a matter of elite behavior; in fact, it's quite the opposite. ■ The skills of literacy are valued most by those to whom they are denied. When Booker T. Washington tells in his autobiography how "From the time that I can remember having any thoughts about anything, I recall that I had an intense longing to learn to read", he conveys a desire that wouldn't have been different, even if he had lived in a "screen/video world". ■ Much can be conveyed by the routes of oral transmission (video, music, speech, radio, or tales told around the campfire). But nothing in thousands of years of human civilization has exceeded the capacity of the well-written (and usually carefully-edited) written word to convey knowledge, meaning, and depth. ■ The cultural pendulum has swung far in the direction of the oral formats for now, but it's an episodic event, not a permanent change. Even barbarians ultimately come to regret that something is missing. Either things begin to fall apart at the societal level or, individually, they respond to the very same innate spark that animated Booker T. Washington to know that his early illiteracy deprived him of something he wanted very much. Human nature is curious. Plenty more remains to be both read and written.
August 6, 2025
The trouble with student evaluations
The biggest problem with student evaluations of their teachers isn't the room it opens up for mischief, even though that problem is quite significant, particularly as course evaluations become a prospective tool for government intervention at colleges and universities. That problem is emergent and well worth ongoing attention. But even in a world where every evaluation were submitted (and reviewed) entirely in good faith, the bigger problem is that students are almost by definition inadequately equipped to gauge what they're evaluating. ■ You don't know on the last day of class how much you will retain nor how well this instructor compares to one you didn't have, teaching the same course but in a different way. You can't know these things as a student on the final day of class -- unless, perhaps, you're taking the same class for the second time because you failed the first, in which case there's a decidedly strong conflict of interest. ■ The meaningful outcomes of a course -- from pre-kindergarten all the way through graduate school -- may remain unknown for a decade or more to come. A well-honed ability to compare and contrast teaching quality is unlikely to develop materially along the way. ■ If teacher evaluations have a meaningful impact on things like tenure or pay, then the whole thing sets up a terrible incentive structure, rewarding whatever impresses students in the short term rather than what improves their outcomes for the long term. This is one of the reasons why performance pay for teachers is such a difficult topic: In a truly rational world, teachers should be incentivized to do what optimizes outcomes for their students many years into the future. ■ That doesn't mean students shouldn't be asked for evaluations -- especially open-ended ones. But evaluations constructed with badly-chosen metrics and performed by ill-equipped evaluators can't help but cause bad outcomes. Signs are pointing to their increasing use, so it's prudent to pay attention now.
August 9, 2025
A pseudonymous account on Twitter recently had a viral moment with an observation that for all of the people who claim (groundlessly) that "I wasn't meant for Excel spreadsheets. I was meant to fight in Caesar's legions", the appropriate response is that "the Romans would have gone nuts for Excel". It's clever and undoubtedly true. ■ First, true on the face of it: Imagine how happy we in the modern age were by the transition from slide rules to pocket calculators to spreadsheets. It's been a wild ride, and it would seem all the more incredible to someone who had only known a world in which paper itself was scarce. ■ But it's even more important to see what having Excel would have actually meant: A truly incomparable strategic advantage. As Ned Resnikoff noted, "The Roman Empire was first and foremost a trade and logistics network." And as important as it has always been to have good battle planning in war, it's at least equally important to have logistics figured out. ■ No army ever won without figuring out a supply chain. Some have figured it out by plundering what they encountered along the way, but even the legendarily destructive and genocidal Mongols under Genghis Khan still faced resource constraints like having enough grass for horses to graze. ■ Dwight Eisenhower noted in his reflections on World War II that his Russian counterparts "suggested that of all the spectacular feats of the war, even including their own, the Allied success in the supply of the pursuit across France would go down in history as the most astonishing." ■ Empires and victorious armies are really just vast logistical networks, right beneath the surface. Being able to see that -- to see the underlying systems, rather than just the thing that's evident on the surface -- is one of the most valuable skills in the modern world. It would have been powerful in ancient times, too, and there's no reason to believe that it won't still be an advantage to people living two thousand years from now.
August 10, 2025
Poland and Norway to team up on drone defense systems
To get most directly from Poland to Norway, one would have to cross the territories of Denmark and/or Sweden, and possibly Germany as well. Their geographic separation is being overcome by a plan to cooperatively develop a system for Counter-Unmanned Aerial Systems (C-UAS). One company from each country will enter the planned partnership. ■ Notably, though they are not next-door neighbors, both countries border Russia, which has been aggressively weaponizing drones (UAS) against Ukraine. It doesn't take a whole lot of imagination to see whose behavior may be instigating these kinds of partnerships. ■ The supposed plans for a meeting in Alaska between the presidents of Russia and the United States should have a lot of the world taking notice. Will an unwilling Ukraine be forced into compromises it wasn't invited to discuss? Reasonable minds can hope not, but there's reason to believe that such a worry isn't misplaced. The very thought that outside forces may decide what happens to a country like Ukraine should keep people awake at night. ■ That ought to be alarming for Europe generally. Even if things go entirely Ukraine's way (which is unlikely), the countries that could easily be in the way of future Russian territorial hunger pangs should act as though trouble is somewhere on a spectrum from "not unlikely" to "nearly inevitable". The time has never been more ripe for getting cross-border cooperation underway to make for stronger European-grown defensive technologies.
August 11, 2025
Relgious freedom between secular states
Shared with endorsement: "Authoritarian regimes don't just politicize religion: they reconstruct it to consolidate power, recognizing that spiritual legitimacy shapes identity, loyalty, and meaning [...] Today's 'managed spirituality' is not a relic, but a deliberate, strategic tool of control. Liberal democracies must confront this with imagination."
August 13, 2025
Don't just look for the gas -- check for the brakes
Two grown adults are engaged in the online version of a slap fight, as Elon Musk and Sam Altman exchange barbs and screenshots of their AI chatbots accusing one another of disreputability. Beyond their self-evident personal animosity, it reflects the high intensity of the ongoing "gold rush" phase of artificial intelligence development. Lots of people are heavily invested in making truly extraordinary predictions and promises at a time when capital expenditures in the sector are large enough to meaningfully affect GDP growth rates. ■ Big business has always attracted big personalities, of course. We aren't soon to forget the names of JP Morgan or Henry Ford. But there's a note of caution that applied then, and it applies now, when it comes to promoters -- both the people at the center of the action and their many cheerleaders and partisans. ■ Zealotry is often the red flag waved by the uninformed. Increasing knowledge of any subject area -- be it technology, religion, economics, medicine, or any other -- should tend to increase one's humility about the boundaries of what is possible, prudent, or wise to do. ■ That humility doesn't necessarily mean that one's enthusiasm should diminish, only that one should have an increasing awareness of limitations and potential hazards in at least roughly equal measure with awareness of the possibilities. A beginning driver may see a car and think only of its potential for speed. The experienced driver can appreciate the speed, but should be able to look at the same car and wonder equally about its brakes. ■ The really prominent individuals in this field probably know more about the limitations than they're willing to let on -- though it's also possible they have become so entranced by the vista they've painted that they no longer see the limitations as they should. So it may go with the hype cycle.
August 14, 2025
A recurring theme within Internet discourse is "People getting angry over the minor jokes and small joys of others". The alert viewer can find examples all the time. The anger quite often (perhaps preponderantly) comes from individuals who identify quite vocally to the left of center and who are quick to take an almost Puritanical offense at even small deviations from their sense of orthodoxy. ■ Cooper Lund highlights an example of this condition from someone denigrating another person's rather anodyne observation that it's perfectly fine to go out and enjoy living, shopping, and dining among other people within a community. Lund notes: "There's a lot of people on the left who have substituted their version of communism for puritanical religion and then wonder why more people don't agree with them." ■ The famous line from "Glengarry Glen Ross", "ABC [...] Always Be Closing", gets morphed by some into "ABS: Always Be Scolding". And it's not a good look. The semi-professional scolds need to be reminded periodically that it's quite OK to take "the pursuit of happiness" at face value. ■ It is perfectly sensible to acknowledge that the world is a mix of both good and bad experiences, and that there are many mild and gentle joys worth celebrating (sometimes even in public), even in times that contain big challenges. Happiness itself is a good thing -- so good, it merited an express acknowledgment in the Declaration of Independence. Individually, we should seek happiness, and we should be pleased to see others seeking it for themselves.
The corporate vice president in charge of Windows products at Microsoft has made an appearance on a company video channel speculating enthusiastically about a near-term future in which Windows emphasizes "multimodality" -- stretching users beyond the conventional keyboard and mouse. A considerable portion of this, it appears, drives toward voice-driven computing. ■ Putting aside the many "Star Trek" references the concept conjures up, this kind of promotional enthusiasm tells us something important. It says that we've gotten technological development way out ahead of cognitive science. And that's problematic. ■ Really nobody who knows how human learning and reasoning actually work would say, "Gee whiz, let's fill open-floor-plan offices full of people making constant noise! It'll be great for productivity!" ■ Your brain works differently when it's composing sentences (or sentence fragments) with pen and paper, versus with a keyboard and a screen, versus in regular human conversation, versus being spoken into a machine with live feedback. These are different pathways and the differences affect the outcomes. ■ The same goes for receiving and processing information: What you read on a screen, read on a printed sheet of paper, hear in conversation, hear in a lecture, or listen to a machine read back to you all go through different cognitive mechanisms. (The screen inferiority effect is real!) Attention, comprehension, and recall are all affected by the mode of input. ■ Microsoft has an institutional imperative to deploy new technologies and to make them look like they will magically make office productivity sizzle. But it's important for the rest of us to take a step back and ask whether much-heralded changes really are for the better. ■ The paperless office was a myth, and for good reason: A lot of documents are better off being stored and transmitted electronically, but many (if not most) people still perform better when they read a written page. We need to carry the same awareness that not all modes of interaction are equally good into the imagined workplaces of the future. Just because technology can do something doesn't mean it's better for our brains that way.
August 16, 2025
Left to their own devices, AI bots form terrible social networks
A big computing experiment at the University of Amsterdam attempted to test a fascinating premise: How would a big population of artificial-intelligence bots respond to being unleashed on a social network -- not to create spam, as so many bots do, but left simply to interact with one another? The researchers set up a social network, walled off exclusively for the bots, and let them loose. ■ According to their pre-publication draft, the researchers observed the bots "spontaneously form homogeneous communities, with follower ties heavily skewed toward co-partisanship" and end up with "a highly unequal distribution of visibility and influence" (that is, they develop high-status "influencers"). ■ It sounds familiar in all of the worst ways. So far, those are the same consequences widely seen within online social networks populated by humans. Spontaneous community creation is just fine -- maybe even a sign of healthy interaction -- but the skew towards partisanship isn't. And the surely there's something upside-down about things that aren't even self-aware still seeking social status. ■ There's something else, though, that should stand out: Humans can recognize that these outcomes are suboptimal. Not only that, we can choose to opt out of unhealthy networks. And we can choose to create humane rules for old or new networks to make them more pro-social, if we choose. ■ A network could, for instance, require users to post one compliment a day. Or to submit to a mutual rating system for positivity or sociability. Or to post updates subject to strict rate-limiting in times of flame wars. (That last one isn't anything new -- message board and listserv administrators have been using the technique for decades.) ■ What's interesting is that those solutions are evident to humans (even if our most prominent social networks utterly fail to put them into use), but they seem not to have occurred to these chatbots. We can be faulted for making poor choices about trade-offs...but why should we trust emerging technologies that don't even self-impose thoroughly rational rules for self-betterment?
August 17, 2025
If both death and taxes are inevitable, then it's unsurprising that talk about imposing taxes upon death should also be periodically unavoidable, too. That's currently the case for the UK, where talk of invigorating the country's inheritance tax system is running hot. The Guardian has editorialized, "There is a powerful argument for intergenerational fairness in a society where inheritance, especially of property, dictates life chances, dividing ever younger cohorts into landowner and tenant classes. Taxing inheritance is a modest but necessary levelling mechanism." ■ Such arguments seem to survive in almost copy-and-paste format from every inheritance tax debate cycle to the next. Sure, it's a "powerful" argument, but that doesn't make it persuasive. ■ There will always be some intuitive appeal to the claim that it's unfair for some people to gain a financial advantage over others just because they landed in the lucky end of the gene pool. Yet there's also a widespread intuitive understanding that some people just have better luck than others, and almost everyone is either happy to have it or jealous of those who do. ■ People also tend to realize that the greater the emphasis on activating government power to control choices about what people can pass on after death, the greater the amount of scheming that will result to avoid the penalties of taxation. It's a make-work program for accountants and estate attorneys. ■ Moreover, two of the great asset classes (real estate and private businesses) are also the hardest to piece out so that tax can be paid. It's hard enough to divide a quantity of farmland among individual heirs or to split up a family-run construction business without also accounting for selling off a portion of the operation to satisfy the tax collector. ■ The difficulty inherent to division, especially when combined with a meaningful inheritance tax burden, only serves to subsidize more impersonal structures of ownership, like REITs and publicly traded corporations. Things divided up into millions of small shares are naturally easier to liquidate to pay some taxes than heavy equipment or apartment buildings or privately-held patents. ■ It's important to stop and ask whether a policy is really enhancing desirable social consequences or merely replacing one undesirable outcome with another. It's not always the case that a family-owned farm or business enterprise is better for the community than one that is publicly traded, but fairly often it is. Britain is welcome to make its own choices regarding inheritance taxation, but it would be unwise to ignore the secondary effects of policy choices.
August 20, 2025
A well-seasoned expression says, "If you want to travel fast, go alone. If you want to travel far, go together." There are times for going fast, but there are many more times when it's worth going far. One of those times was when Ukraine's president traveled with seven other European leaders to meet with (and attempt to persuade) the President of the United States that his country is worth a modest investment of solidarity in its fight to repel a Russian invasion. ■ The whole of human civilization lost out in a big way when Russia, emerging from the wreckage of the Soviet Union, failed to achieve the necessary escape velocity from decades of Communist oppression to emerge as a true, liberty-based democracy. There was a window of opportunity -- and other countries with the same Communist baggage, fewer natural resources, and less technological progress at their disposal managed to make it through. ■ Russia did not, and though the Russian people suffer most directly from that failure, others have suffered, too: Prominently, the people of Georgia, Belarus, and of course, Ukraine. In a better alternate history, a reconstructed Russia is contributing peacefully and mightily to the world's scientific, cultural, and economic progress, much as post-war West Germany and Japan or post-occupation South Korea or France do in our real world. ■ Societal openness and individual freedom work like that: They generate desirable progress domestically, but that progress spills over into good things for other nations, too. It's a matter of traveling far by traveling together. ■ Ukraine deserves every reasonable chance to win its present military conflict. Europe seems alert to the fact such a victory would be important to its security, too, which is why France, Germany, the UK, Italy, and Finland sent their leaders along on the urgent mission. Going together matters. ■ And in the long run, the world needs to remember to promptly offer every encouragement whenever a country shows signs of turning away from a sinister system of government and giving liberal democracy a try. It won't always result in success. Reforms don't always stick and culture doesn't always grow. But as we have seen, the costs of failure are enormous.
August 22, 2025
Politico has published a story with a peculiar headline: "US to take part in Russia's answer to Eurovision". Eurovision, of course, is the enormously popular musical contest based in Europe, and the Russian contest in question is Intervision, a Eurovision knockoff put on by the Russian government. ■ Russia was kicked out of Eurovision in 2022 over its invasion of Ukraine. The rival state-run contest is Russia's attempt to legitimize its own cultural standing in the world. ■ Any decent American should be ashamed of taking part in a knockoff contest under such a terrible cloud. It's not like the Russian contest existed prior to the war of aggression against Ukraine; it's a consequence of the war, and anybody taking part should be ashamed to bring it any reputational standing. ■ It's also important to ask, "Under whose authority is an American being sent?" When Politico says, "The United States is taking part", do they mean that the State Department is sending the singer, or that it is more like a self-appointed committee of five people exercising terrible judgment? ■ It's a bad look either way, whether any official activity is involved or not. But it's doubtful that there's anything official going on. What agency or department would even pick a representative? And the reporting should be clearer about that before saying "The United States" is doing anything. ■ Don't lend credibility to the non-credible. If the Russian government wants to carry out a second-rate singing contest, that's one thing. Even NBC can do that, and it doesn't have any state power behind it. ■ But if an activity is plainly not worth legitimizing -- like an attempt to rehabilitate the cultural image of a country bent on conquest through means like 80 attacks on maternity units in hospitals -- then reporting should be clear about the remainder of the world contributes to that credibility. ■ Are there official acts involved? If so, how high do they go? Has a low-level bureaucratic functionary somewhere rubber-stamped a bad idea, or is the Secretary of State trying to send a message? These distinctions matter. In a case like this, no engagement is a good idea, but the private acts of weak citizens are very different from decrees from the White House.
August 23, 2025
All's fair in love, war, and late night
In an interview with Marc Summers (the same person who once hosted "Double Dare"), Jay Leno claimed that his secret weapon in succeeding Johnny Carson as host of the "Tonight Show" was his willingness to act as a guest host for a fraction of the price that other guest hosts wanted to charge. ■ There are plenty of legitimate criticisms to offer about Leno: His handling of the transition to Conan O'Brien's time in the seat is almost impossible to defend with a straight face. And his treatment of Monica Lewinsky as a punch line (rather than as a young person who was involved in an office affair that had a pretty clear aggressor at fault) probably crossed the line into public bullying. Yet other criticisms are entirely fair play. ■ But he probably doesn't deserve to be skewered for his approach to landing the "Tonight Show" gig in the first place. Some weird things happen when an attractive opportunity emerges that has a monopsony structure. In a monopsony, there's only one buyer (in this case, NBC). There was no "also-ran" opportunity chasing after the "Tonight Show": Carson made it not just an 800-lb. gorilla, but really the only game in town. ■ And when there's only one buyer, prospective sellers -- in this case, Leno and his fellow guest hosts -- may be entirely rational if they choose to do just about anything to land the gig, including working for absurdly low rates. Land the contract, prove your value, and raise your prices later. That wouldn't excuse doing something illegal or immoral to anyone else to get the job, but within the bounds of law and decency, almost anything else is fair play.
August 24, 2025
Musicians have long copied, mimicked, and reinterpreted the work of other performers. Someone develops a new style, and someone else tries to woodshed it until they land the technique. Someone writes a phenomenal song, and others cover it in their own ways. The Beatles land in America and a TV producer comes up with The Monkees. ■ But it's a long way from human imitation -- inevitably imperfect and often a mark of admiration -- to open impersonation. The latter is a problem online, where fake albums are surprising the real artists credited with making them. ■ There's probably something a little flattering about the knowledge that anyone would go to the effort to train AI on your work, but the flattery is probably outweighed by the insult of the scam. The word "authenticity" pops up quite a bit when people talk about computer-generated music: It can be adequate, in the same way that unremarkable instrumental music can be passable in a hotel lobby. ■ It seems that many of the fakes are after low-stakes ripoffs; they are not (at least yet) after the big artists with much success. The cost of protecting one's image might be reasonable only for those who already have lawyers on staff.
August 25, 2025
Why you care about an independent central bank
The biggest obstacle to a real public understanding of economics is not that the subject itself is intuitively hard. The real problem is that so many of its conclusions are relatively straightforward, but the most important ones tend to be either non-obvious or downright counterintuitive. That's because at heart, economics is a study of human behavior and human beings just don't always make sense. ■ One of those important, but non-obvious conclusions is that the best possible situation is for prices to rise (but only buy a little bit) all of the time. It's easy enough to see why prices rising too fast (high inflation) are bad things. But people tend not to realize that it can be equally bad for prices to fall for a prolonged period of time (deflation). Deflationary periods, when prices fall across-the-board for too long, cause people to decide to hold off on spending so that they can get better deals later. This causes economic contraction that can be just as painful as when prices rise too fast. ■ Individually, it's hard not to like the feeling of falling prices. But spread across the scale of a population, the phenomenon becomes a huge problem. It isn't obvious that low and predictable inflation is the ideal condition, but that's one of the key assignments given to the Federal Reserve Board. ■ The Federal Reserve is charged with aiming for prices to rise gently and predictably. This leaves people best off across the board because it preserves most of the savings that they keep from year to year while very gently encouraging people to spend rather than hoard their cash. ■ Finding the Goldilocks number for inflation (not too much, not too little) and sticking to it is an incredibly difficult task. It's difficult not just in mathematical terms, but also in social science terms. Not everyone has the same incentives to cheer for the same amount of inflation at the same time. Retirees, for instance, typically want inflation to remain very low, while companies with heavy debt burdens might like a lot more. ■ That's what makes central bank independence utterly vital to an advanced economy. It's not that a central bank should be completely ignorant of politics (after all, it is acting on the most consequential of the social sciences), but there needs to be sufficient independence that the people in charge of managing the money supply are free to act without fear that they will be persecuted or punished for doing the right thing when it requires going against prevailing public opinion. ■ Again, it is not intuitively obvious that the smartest thing a government can do about inflation is to stay out of the question and leave it to professionals who are outside of their reach. But that is the right conclusion. The more credible the case that a government will leave the central bankers alone, the more confidence people can have not only in that government, but also in the money supply of that country. ■ Confidence is hugely valuable to those who have earned it -- it's expensive to procure and requires self-discipline to maintain. Once you have it, the stupidest possible decision is to fritter it away.