Fundamentally, security is about dividing actions into two sets: things you should do, and things you shouldn't do. Who "you" is, what "things" they are, what "do"ing means, and how "should" is defined and enforced are... well, complex enough to require a whole industry. But it all derives from that fundamental partition of should-ness.

If everyone should do everything, you don't need security. This sounds facile, but in practice it's a useful way to think about situations that are naturally limited in who or what you have to consider. I don't need to think about a security model for my toilet because most people don't want to use my toilet, and also it's a toilet. If I had a very popular toilet, or my toilet were made of gold, it would be a different story.

If nobody should do anything, you also don't need security. The most secure system is a bonfire: the irreversibility of entropy is the only fundamentally-unbreakable encryption. More practically, though, end-to-end encrypted systems use this principle. The server is secure because it's not secure: it's just passing noise back and forth. If you gain illicit access to that noise... have fun, I guess?

But usually someone should do something, and a secure system allows them to do it without allowing people to do things they shouldn't. Solving that problem can be very difficult, but nowhere near as difficult as figuring out exactly what problem you're trying to solve in the first place. Where exactly is the line between should and shouldn't?

The web has this problem in a big way. On the one hand, developers should be able to build powerful software that can do things like play sounds, store files, send notifications, run in the background, and give haptic feedback. On the other hand, advertisers and malware authors shouldn't be able to do things like... play sounds, store files, send notifications, run in the background, and give haptic feedback.

So our line between should and shouldn't is obvious: who's doing it. We'll just ask each developer whether they're a nice person building good software, or a heinous evildoer foisting crapware upon the innocent. And, of course, the evildoers will respond "it's not crapware, it's valuable metric-driven high-engagement interactive content that provides users the brand awareness they so desperately crave". Back to the drawing board on that one.

Instead, we figure out what kinds of things good people do, and what kinds of things bad people do. Maybe regular developers mostly store data about their own site, whereas advertisers need to store data in a way that they can harvest across many sites. Maybe regular developers don't mind waiting until you've clicked something before they start vibrating or playing sound, but malware authors want your attention right away .

Now, of course, maybe there are some reasons why a regular developer might want to do malware-looking things, or a malware author might find a way to use regular-looking things to do malware with. So we have to draw that line very elaborately and very carefully. How do we do that? We look at what people are doing and turn that into the definition of what they should be doing. We take the current way the system is being used, and we lock it down.

Unfortunately, this is a ratchet that only tightens. As bad people figure out things they're allowed to do and misuse them, those things become disallowed. But how can good people use something that's disallowed in order to convince the system that it's good? The coastline between good and bad becomes increasingly complex and rigid.

Ah, but light at the end of the tunnel! Because somebody has figured out how to build a new system inside the old system! So we can make our new system without so many complex restrictions, because it's just for this new stuff we're building, and all the old, important stuff is stored in the old, rigid, secure system.

Wait, what's happening? Our new system's flexibility means people are making more and more things in it, using it for things we never imagined, and putting important stuff in there that would be a really good target for the evildoers that are rapidly beginning to examine our new system for weak points? Oh nooo... guess we need to lock it down.

Each stage of the lockdown cycle codifies whatever people should be doing at the time it was locked down. Once upon a time, the thing developers should be doing is distributing physical discs with programs on them that are updated once a year. So we locked that down with anti-virus tools and systems for denying/allowing certain programs.

But when everyone started using applications with internet access, the real problem became vulnerabilities in those applications. You might not be able to install a virus, but you can write an evil Word document or PDF, or maybe you send a specially-crafted email that causes your target to send another specially-crafted email... antivirus software became much less useful, because the the complexity of internet-connected applications made them the new system, and the new thing you shouldn't do is have your fancy and legitimate program accidentally running arbitrary code from the internet.

And then, of course, the web app era came along, built entirely on the basis of deliberately running arbitrary code from the internet. What a revolution, but also what a security and privacy nightmare. Luckily, we're starting to get that whole mess straightened out, but in the process we're getting very specific about what kinds of things we expect people to use the web for.

Eventually, those things may become stifling enough that the next, un-locked-down thing begins to flourish. ]]> https://samgentle.com/posts/2021-01-11-composing-and-improvising https://samgentle.com/posts/2021-01-11-composing-and-improvising Mon, 11 Jan 2021 00:00:00 GMT When organising your behaviour to achieve a goal, there are two equal and opposite skills: composing and improvising.

Composing means making decisions as early as possible. You anticipate the outcome you want, figure out the paths to that outcome, and construct a plan for getting there. You're making decisions at the point of maximum influence: with more time, your decisions can compound for longer, and you have a lot more freedom to choose what to do and when to do it.

Composing requires bringing the future into the present, making decisions there, and then projecting those decisions back out into the future. Both those transformations create uncertainty and risk, and in that sense composing is quite fragile. You start with a mix of incomplete information and conjecture, turn that into a bunch of assumptions, and turn those assumptions into behaviour.

But as difficult as predicting the future is, the real difficulty is enacting it. Every plan is a constraint on your future behaviour: a commitment to do one thing instead of another at a point in time. But when the plan meets reality – or, rather, doesn't meet it, those constraints can become stifling. Obviously, a decision based on incorrect assumptions should be changed, but any replacement decision is being made with less time, less influence, and less freedom. And what if that new decision conflicts with other, more careful decisions you've already made?

Improvising, on the other hand, means making decisions as late as possible. You might still anticipate a general space of outcomes, but not necessarily a specific one. Instead of preparing by deciding in advance, you prepare by putting yourself in the best position to make decisions as they arise. These decisions are made at the point of maximum information: everything is done but the decision itself.

The key to improvising is flexibility. To make a good decision in the moment, you have to keep a range of options open, and be able to choose quickly and freely. You don't need to anticipate the future, you just wait until it's close enough to the present that the right decision is evident. In that sense, improvising is quite robust – to almost everything except planning.

Composing and improvising are, if not exactly opposites, at least mutually incompatible. Composing relies on future behaviour being determined by present decisions, and improvising relies on present behaviour being determined by present circumstances. To do one better, you have to do the other one worse.

There are situations where composing doesn't work, like anything rapidly-changing or unpredictable, and situations where improvising doesn't work, like anything that requires directed action on a timescale longer than a few hours. And it is possible to use one to fill the gaps of the other, but not really possible to use both at the same time.

This means that to practice composing when you're used to improvising, or practice improvising when you're used to composing, you need to give up the skill that works best for you at exactly the time you'd tend to use it. You have to get worse before you get better, because the skill you already know actively sabotages the skill you're trying to learn. ]]> https://samgentle.com/posts/2021-01-06-abstract-ignorance https://samgentle.com/posts/2021-01-06-abstract-ignorance Wed, 06 Jan 2021 00:00:00 GMT There's a kind of argument that starts with one person saying something like "social media companies should ban disinformation", and the other saying "but once you start banning speech who knows what it will include". Then the first person says "well I know what it will include: things that aren't true", the second says "ah, but what is true?" and the prophecy is complete.

This looks like a argument about censorship, but really it's an argument about abstraction. Chances are these people both agree that disinformation is bad, that banning speech is bad, that disinformation is untrue, and that truth is not universally-agreed-upon. The points that are raised aren't in conflict, but the people are talking past each other because they can't agree on which level of abstraction is appropriate.

It's easy to see a more abstract concept as more enlightened, or even more virtuous. But really the best way to understand it is that a more abstract concept is more ignorant. The more abstract your point, the less you need to know for it to be true. An abstract point applies in more situations, because it ignores the details that make it specific to any one situation. Ignorance and generality are two sides of the same coin.

When we make an argument like "banning speech is bad", we ignore what specific speech we're talking about. If we know for certain that the speech in question is loud demonic screaming outside your bedroom window at 4am, banning it is just fine. But if we don't know whether it's demonic screaming or criticism of political parties, it's better to risk the occasional 4am exorcism if it allows us a functioning democracy.

This ignorance is sometimes because we genuinely don't know, like when we're trying to find a rule for situations we haven't encountered. But other times it's a deliberate choice to forget. We know that 4am screaming is bad, but we don't trust the people who determine what constitutes 4am screaming. Better to pretend we don't know and find a more general rule.

But abstraction's deliberate ignorance can also be a disingenuous tactic. Yes, we might not be able to universally define misinformation, but it's not hard to define misinformation for less-universal domains like 5g-induced-diseases, cancer cures made from shark bits, or whether vaccines work. Pretending otherwise is just trying to abstract away the truth.

Despite the towering temptations of metaphysics, we do live in an actual physical universe, with actual things we can test and determine to be true or false. We don't need to find arguments that cover the other universes; they can fend for themselves.

And, I mean, they'll have to if they're going to fight off the hordes of autistic 5g corona-sharks. ]]> https://samgentle.com/posts/2021-01-05-world-memory https://samgentle.com/posts/2021-01-05-world-memory Tue, 05 Jan 2021 00:00:00 GMT You know that old trick for finding your way out of a maze? You keep your left hand touching the wall the whole time, leading you to follow a series of left turns that will, inevitably, bring you to the exit. It's a fascinatingly simple algorithm, and an interesting way to understand it is as a kind of depth-first search.

A depth-first search is one way of exploring a space where, each time you have a choice, you pick the first option. Repeat until you get stuck, at which point you go back to your most recent choice and pick the second option. If you're out of options, you go back further. It's called depth-first because you go down each path as far as you can before exploring any other options.

To use this algorithm, though, you need to keep track of where you've been. After all, when I say "go back to your most recent choice", what was it? And when I say "first option", "second option", etc, how do you know which one you're up to? You need to keep some kind of list so you can know where to go next. But, for some reason, the left-hand-following-the-wall solution doesn't use one – how can that be?

Well, imagine a Y shaped passage, where you enter from one side and the other two are dead ends. If you keep your left hand on the wall, you'll take the left fork first, then when you hit the dead end, you follow the wall around to the right fork, which leads you around the dead end back out the way you came.

Play that again in slow motion: the wall leads you back to your most recent choice when you hit a dead end, and the wall connects each option to the one after it. The reason you don't need to keep a list of decisions is because the wall and the list are equivalent. Or, to put it another way, somebody already stored that list in the form of a wall.

The memory requirement of the algorithm hasn't been overcome, we're just using the physical world as memory. Not in the symbolic way that marks on paper are physical, but profoundly, concretely. The walls create a maze, but in doing so they also store a path that explores the maze. ]]> https://samgentle.com/posts/2021-01-04-the-longest-path https://samgentle.com/posts/2021-01-04-the-longest-path Mon, 04 Jan 2021 00:00:00 GMT One of the strangest things about video games is the way they encourage you to take the longest path. Most games have some system of progression: a storyline, levels, or a map you move through methodically. However, there are also many things to explore in each area, and when you move on you probably won't return to them. To get the most out of the game, you actually want to avoid winning for as long as possible. In that sense, winning is a kind of loss.

I've started noticing areas of my life where I seem to avoid winning. Rather than a desire to lose, I think this reflects a desire to explore, an unwillingness to move on too soon. If I finish this today, I know there will be parts of it still unexamined. Questions gone unanswered. Experiences left unhad. Why would I want to move forward, if it means leaving important things behind?

But, unlike in a video game, progress is not linear, and not even in a consistent direction. There is no helpful-yet-insistent arrow ushering you from one idea to the next. What is finished today can be even more finished tomorrow. Winning doesn't have to mean moving on. And moving in any direction can still be progress, even if it's revisiting old ideas.

This is a difficult lesson to learn, at least for me. My mindset is much more geared towards discovering the answer, solving the problem, slaying the dragon, and moving on to undiscovered/unsolved/unslain pastures. It's a model of progress as completion. But I think there is a lot more to be found in progress as iteration, refinement and steady accumulation.

Ironically, the key to finishing things may be knowing that they are never truly finished, and thus there is no best ending to hold out for. ]]> https://samgentle.com/posts/2021-01-03-the-great-disabler https://samgentle.com/posts/2021-01-03-the-great-disabler Sun, 03 Jan 2021 00:00:00 GMT Technology is meant to be a tool to enable us. To a tool-using species, tools are meant to fade into the background, to disappear and become extensions of our minds. But our technology often seems to be very insistent that it stays in the forefront of our attention. Worse, it often disobeys us, places restrictions upon us, or acts in ways that subtly disrespect or disempower us. Why?

Modern technology is less like a tool and more like an agent. An agent acts on your behalf, but it does not exist merely to serve you. It has its own values, its own interests, and there is no guarantee that those interests align with yours. In fact, most modern technology exists primarily to enrich the companies that created it. To some extent, that has always been true, but tools were too dumb to continue enriching their creators once they left the factory and entered your home. Modern technology has no such limitations, and unabashedly carries out its own agenda even while in your possession.

The resources of human attention, human interest, and human action are very valuable, and the your technology requires those resources to achieve its goals. However, you also want those resources for your goals, and this creates a conflict that your technology, smart as it is, would easily lose. To win, it must fight dirty, exploiting quirks in your psyche to subvert your interests. That's why it has to be in the foreground. That's why it has to disobey you. That's why it has to distract you. It must weaken you enough to extract your resources. Technology is still a tool, but it is not your tool; it is someone else's tool being used on you. You are not the ploughman, but the ox, and it is your yoke.

Technology has been called the great enabler. But who is it enabling and what is it enabling them to do? If it is not your agent, it is not enabling you. If it enables others to subvert your intentions, then it is, in fact, disabling you. This is the sorry state of modern technology: the great disabler. Humanity's most sophisticated Trojan Horse. A gift that ostensibly makes us greater while it compromises us from within. ]]> https://samgentle.com/posts/2021-01-02-inefficiency-cascade https://samgentle.com/posts/2021-01-02-inefficiency-cascade Sat, 02 Jan 2021 00:00:00 GMT I've been thinking about global catastrophes lately, for obvious reasons. An interesting thing to notice is the way that the global machinery of production and trade acts, much like a biological organism, at multiple layers of redundancy.

A common misconception is that hearts tend to fail by stopping, as per the classic hospital flatline scene where the defibrillator magically shocks a dead person back to life. In reality, hearts don't often fail that way (and if they do, zapping them doesn't help). What they tend to do instead is flail uselessly in a way that doesn't pump blood very well.

The reason for this is that literally every part of the heart wants to pump. Normally, that's controlled by a bundle of pacemaker cells near the top of the heart. But if that fails? Another bundle further down takes over. And another, and another. Each level of failure just makes things less efficient and more localised.

Similarly, when a pandemic comes along and screws up all our supply chains, the resources needed to manage and treat it end up being produced more domestically and less efficiently. Issues of supply and distribution have, indeed, been a notable feature of the COVID pandemic, from masks to toilet paper to hand sanitiser.

But when considering climate change, a more sobering thought comes to mind. An easy assumption to make would be that climate change is, in a certain sense, self-limiting: the consequences of fossil fuel consumption will disrupt large-scale production, which will also decrease fossil fuel consumption.

However, another possibility is that the large-scale production will be replaced by less-efficient small-scale production. If global-scale solar panel production falters, the slack will be picked up by national-scale fossil fuel power, which will itself be replaced by local-scale fossil fuel power if it fails. Each contraction comes with a decrease in efficiency.

In that case, the disruptions caused by climate change could lead to even higher emissions, as we increasingly trade efficiency for resilience. ]]> https://samgentle.com/posts/2021-01-01-symplany https://samgentle.com/posts/2021-01-01-symplany Fri, 01 Jan 2021 00:00:00 GMT I used to think I was okay at making plans, until I met people who are good at making plans, which led me to conclude that I am, in fact, bad at making plans.

An important difference I've noticed is in how we build plans on top of one another. If a friend knows you are working until 3pm, they might suggest having coffee at 3:30pm. This is really one plan (coffee at 3:30) stacked on another (work until 3). Intricate as it sounds, this is well within the remit of an okay planner. What good planners do isn't just make plans, but support the ways that others stack plans on top of theirs.

One part of this is providing good plan-building information. Things like if you finish work early that day, or you're working from a different location, or you have an appointment at 4:30. Even if none of these things make a difference to you, they might affect your friend's preferences. Maybe, if they knew, they would suggest an earlier time, a different day, or meeting closer to your appointment.

Another part is being aware of what information others might use to plan. Perhaps you once mentioned you like pickles, but you've since gone off them. Could your friend have made plans to go to a pickle cafe? Or perhaps you mentioned that this week has been very busy. Will your friend assume it's a bad week to get coffee when actually you would welcome the break?

And another still is structuring your plans in a way that is easy to plan around. Maybe your finish time is flexible, but 3pm is what you tell people in case they want to spend time with you.

All of these things require what is, at least to me, a higher-level skill. Not planning, but having an inherent sense of how other people are planning around you. Providing others with the information they need to make good plans actually means you need to do less planning to achieve the same outcome. This can make the process seem deceptively easy to an okay planner. ]]> https://samgentle.com/posts/2020-05-31-incidentals https://samgentle.com/posts/2020-05-31-incidentals Sun, 31 May 2020 00:00:00 GMT Lately I've been noticing how easy it is to become attached to incidentals. These are the things in your life that exist to serve some other purpose but aren't, of themselves, important. Incidentals are good! They support and implement a purpose by carrying its consequences forward into all the different parts of your life.

There are many different purposes that could justify the same incidental. Living in the city could serve the purpose of being close to a high density of people, or of being in a more politically liberal area. Owning fancy kitchen equipment could serve the purpose of training to be a chef, or of making your home cooking tasks more pleasant.

If a purpose changes, its incidentals should change too. However, sometimes the link between purpose and incidental becomes unclear, or is even broken entirely. When this happens, you're left with zombie incidentals: rituals and habits that now exist only to perpetuate themselves.

Sometimes, you acquire incidentals that never had a purpose. Maybe you make your bed every morning, not because you care about its aesthetic or psychological benefits, but just because your parents told you to and you've been doing it ever since. These are ghost incidentals: the uninvited spirit of a purpose that just hangs around haunting you.

These spooky-but-useless incidentals are mostly just a nuisance: piles of volitional clutter that merely distract you from their living and intentional counterparts. However, where things really go off the rails is when enough incidentals band together, dig in, and claim to be more important than the purposes they serve.

You can tell if you're afflicted because when some purpose-level opportunity comes along – a dream gig, a career change, an exciting adventure – you find yourself thinking "oh no, what will I do about my potted plants?" If this happens, your incidentals have unionised, and you are in trouble.

Fortunately, once you recognise them for what they are, troublesome incidentals are easily remedied: just stop sustaining them. Anything that demands your space or your efforts, or presents itself as an obstacle in the face of your goals, deserves an immediate ID check. Who are you again? Who did you come here with? I knew it! Outta here, you undead tankie freeloader! ]]> https://samgentle.com/posts/2020-03-09-self-empathy https://samgentle.com/posts/2020-03-09-self-empathy Mon, 09 Mar 2020 00:00:00 GMT It seems like a very strange phrase, self-empathy. Empathy is when you connect with someone's feelings or share their mental state. You, well... are yourself. How could you not experience self-empathy?

The problem is that we aren't only one self. Our future feelings aren't feelings: they're predictions. Our past feelings aren't feelings: they're memories. Who we are changes over time, and our mental state changes with mood, environment and situation. It's a little surprising that this adds up to any kind of self at all.

To connect with feelings that you don't feel right now is a particular skill. It's a skill you exercise when you decide whether to go to a party based on how you'll feel when you get there, or when you decide to relax or push yourself based on how you'll respond. It's also the skill you use to explain yourself to people, not as the you that you experience, but as the you that they experience.

Developing self-empathy takes a certain amount of distance. Understanding your feelings from the inside makes sense in the moment, but won't work later. You have to analyse those feelings while you're experiencing them to keep understanding even as your mental state changes.

On the other hand, it also requires kindness. It's sometimes tempting to believe that your past self was acting incomprehensibly because the emotions that made sense of their behaviour are gone. To develop self-empathy, you have to believe that your feelings were real, even when the only evidence is analysis that no longer feels true. ]]> https://samgentle.com/posts/2020-02-04-distraction-or-diversion https://samgentle.com/posts/2020-02-04-distraction-or-diversion Tue, 04 Feb 2020 00:00:00 GMT I've come to realise recently that a lot of the things I call distractions are really better thought of as diversions. Both are things that you do instead of what you intended. The difference is: why? What makes them more compelling than your original goal?

A distraction is compelling because it's different. You watch TV instead of studying because studying is hard and mentally taxing, while TV is easy and mindless. Or you chat with your friends, because studying is tedious and introverted, while chatting is stimulating and social. Distractions are strong in the ways your intended task is weak.

On the other hand, a diversion is compelling because it's the same. You read people's opinions because you like having opinions. You watch videos of DIY projects because you like DIY projects. You want to be doing those things, but seeing someone else do them is much easier. Diversions are strong in the ways your intended task is strong.

The reason this is a useful distinction to make is that the strategies for dealing with them are very different. For distractions, you need to make the thing you intend to do more compelling and everything else less compelling. If it's human contact you miss, study with a friend. If mindless stuff is too tempting, remove it from your environment when you need to study.

But none of that really works for diversions. Focusing on how much you love DIY projects makes you want to both do and watch DIY projects. The two are hard to tease apart: watching other people do a DIY project is often a crucial part of not screwing it up. The only reason it's even a problem is that watching feels so much like doing – without actually getting anything done.

And it's this last thing that I see as the right strategy for dealing with diversion. At its heart, a diversion is still an attempt to do the thing you really want to be doing, just in a way that doesn't work. It's a kind of motivational judo, and you can use that judo right back: if I'm screwing around because I really like something, wouldn't I rather just go do it? ]]> https://samgentle.com/posts/2020-02-01-broken-promises https://samgentle.com/posts/2020-02-01-broken-promises Sat, 01 Feb 2020 00:00:00 GMT There's a certain mode of being, I think of it as seeking, motivated not by the joy of the moment, or the satisfaction of achievement, but by the thrill of the chase.

It's easy to tell the difference: what happens on that fateful day when you finally get what you want? The hedonist says, "this is just like every day." The achiever says, "this is a wonderful day and I feel proud of what I have accomplished." The seeker says "yeah, good, okay... what now?"

I often feel this when I'm deciding what to buy, researching a broad topic, or surfing one of the internet's infinite waves of mild stimulation. Look at all these choices! What's this— ooh, wait, what's that? Just one more turn? One more click? What if number 7 really does surprise me? It's good, but never good enough. Close, but never quite there. So I keep seeking.

And you can see how the power of the chase was meant to help us stay motivated in pursuit of our dreams. It would be a rare gift to keep your faith in every tiny moment, to laugh at your frustration, to sing through your doubt, to live as if you were already free. Far more reliable to stop staring out at the distance, look down at your feet and just keep going.

Seeking is an abstraction: a concentration of meaning. Abstraction draws a big chalk line through the universe and says "stuff on this side you care about; stuff on that side is just details." So, seekers, what do we care about? "Doing!" And what is just details? "Why we're doing it!"

This makes seekers immensely valuable. You can motivate an achiever with promises of success, but you have to actually deliver or they'll stop doing what you want. The seeker doesn't care whether you deliver; it's the promise, not the success, that motivates them.

And so today we see such a wonderful flourishing of promise factories and promise farms, promise networks and promise feeds, all to cater to the seeker's insatiable appetites. All built on the abstraction they once entrusted with their dreams. All singing that same sweet song:

The second time was just as reality began to set in. The climb was difficult, yes, but it seemed manageable. Now, surrounded by ice, I wasn't so sure. Could I really make it? Startled from my rumination, I saw him walking towards me – down the mountain. I rubbed my eyes. He was still there. I asked what he was doing. He said he forgot something and needed to go back.

The third time, I had settled into a steady rhythm. I saw him at the end of a long valley. I was not surprised this time. He asked how I was going. I said fine. He said he admired my dedication. I said I admired his climbing ability. He laughed and said he wasn't a climber.

The fourth time, I was struggling up the sheer ice, a zombie powered by exhaustion. He was there, of course, resting on a nearby step. He asked if he could climb with me a while. I nodded. I wondered if he already knew as my foot slipped and the mountain fell away from me.

I felt it, then, in the creases of his hand. The mountain, aeons of rock and ice, worn and folded until it was no more substantial than paper. To me, it had seemed a forbidden monument, a tower reaching to the foundations of heaven. To him, it was a frosty staircase. He was looking at me. I stepped back onto the mountain. He smiled.

The fifth time, the last time, I was nearing the summit. It had seemed so far once, and so close now. How could these two feelings describe the same journey? He was standing just ahead of me. I asked if he was going to the summit. "What summit?" He asked.

I walked on, alone, the mountain shrinking beneath my feet. ]]> https://samgentle.com/posts/2019-12-12-finishing-school https://samgentle.com/posts/2019-12-12-finishing-school Thu, 12 Dec 2019 00:00:00 GMT I've noticed that it's easier for me to start things than to keep working on existing things, and much easier to do either of those than to finish things. In a sense, this is so obvious as to be boring: of course starting things is easier than finishing them! But, actually, why is that?

In a sense, it's actually pretty counterintuitive. If your life's passion is to fill jars with marbles, the first 50% of a jar is no harder than the last 50%. If you stop halfway through to start a new jar... well, first you'll need to find a new jar, which could be a hassle if the jar shop is closed. And even if you have a new jar all prepared, to fill it you'll need twice as many marbles as to fill the old jar. Even for a marble enthusiast, that's an embarrassingly poor cost/benefit ratio.

I think the first thing we can conclude is that most projects aren't like filling jars with marbles. So, how are they different?

One reason is complexity: for most projects, earlier decisions constrain later ones. Instead of jars of marbles, they're more like crossword puzzles. If you're not sure whether you're getting the words right, things get increasingly untenable as your errors stack up, and eventually it starts to seem like a really good idea to do something that isn't a crossword puzzle.

Another reason is attrition: a fixed chance of abandoning the project that compounds over time. Imagine that, with each marble you put in, you have a 1% chance of breaking the whole glass. In non-marble projects, that might be finding a problem with your approach that makes it unworkable, discovering a better solution, or that the situation that made the project worth doing has changed.

Or something a bit more self-sabotaging, like avoidance: you actually don't want to finish it. Maybe you're worried the thing won't be good enough, or you're afraid of being judged or criticised. It can be more comfortable to toil away on something in obscurity than summon the ego-courage to declare it finished. Who are you to bother the world with this new thing?

But my favourite is entanglement: not finishing something until you finish something else. I want to write something, and I have a project I'm working on, so why not kill two birds with one stone and write about the project! Genius. Except then the project takes longer than I expect and I don't write anything either. Also, have you ever tried to kill two birds with one stone? The physics alone is enough to justify treating that entire analogical lineage with suspicion.

So far we've been mostly looking at root causes of projects being easier to start than finish. Also worth examining are the chronic effects: habit and underdevelopment.

If you tend to not finish things, over time you'll build habits to enable not finishing things. Maybe you'll find coping strategies and workarounds to make it less costly, or develop an almost-sustainable pattern of anxious avoidance leading to last-minute panic excellence. Hypothetically speaking.

And the flip-side is that, even when you try to finish things, you probably don't really know how. Like anything else, finishing things is an ability that has to be developed. It has its own unique rhythm and skillset. I mean, it's not that difficult, but neither is chewing and you made a mess of that the first time you tried it.

It may be better to turn the problem on its head. Perhaps there are a thousand reasons why finishing things is hard, but those difficulties all thrive in the space created by a lack of ability. If you have a lot of practice finishing things; if you are used to the rhythm of it, and have built habits around it; then any factor that makes finishing difficult will have a lot more entrenched structure to push back against it.

So my conclusion is, perhaps unsurprisingly, to practice the art of finishing. Perhaps for me that looks like dusting off old abandoned projects and un-abandoning them, or splitting up large projects into a series of smaller projects that can be individually finished to boost my finish-per-start ratio.

Analysis is often an interesting exercise, but sometimes it's best to fall back on the simple principle that you get better at what you practice. ]]> https://samgentle.com/posts/2019-07-09-sincerity https://samgentle.com/posts/2019-07-09-sincerity Tue, 09 Jul 2019 00:00:00 GMT I want to be a poet. There, I said it.
— A friend from high school, circa 2001

The human is a curious being. Possessed of all knowledge, instantly present everywhere at all times, able to observe everything and communicate with everyone. It is not delighted by this; it is tired. Its voice, once rich and melodic, draws down to a few notes: wry humour, detached observation, and sarcastic rebuke. Its tongue and its cheek, planted together for so long, have finally fused, forming a single tongue-cheek proboscis that it uses to snuffle through its feed for tiny morsels of novelty.

There's a profound coolth to this kind of existence. An invulnerability. To every question, an answer. To every answer, a retort. A game, but not fun. Just something to kill some time. Cow clicker, but for life. Life clicker. Cow lifer? Life cower? Ooh, I like that one. Anyway, it's all good as long as you keep it self-aware. Keep it ｍｅｔａ. The game is forever, and the only way to lose is to die without a wink to the camera.

Jaded

It feels easy to be jaded sometimes. The difficulty of doing great things is vastly dominated by the difficulty of believing in great things, that you are the person to do them, or that greatness even exists at all. There is such an aggressive innocence in the phrase "I'm doing something great" that only the profoundly egotistical or profoundly naive dare utter it without the appropriate hint of self-aware ambition-shaming.

So what does it mean to be jaded? For a simple word it manages an exceptional density of meaning. It's pessimistic, of course. When someone says they've figured out a solution to the Arab-Israeli conflict and you bust out a killer here-we-go-again eye-roll, that's pessimism: "bet you havent". Being jaded doesn't mean you have to believe in bad outcomes, but it does mean you can't believe in good ones. In fact, it's probably closest to not believing in anything at all.

This leads into the next ingredient: a sense of weariness, of being tired out. Not because you've worked too hard, but because you've seen too much. Not sleepy so much as used up, worn down and ground out. You can complement this with a kind of opulent boredom: "seen too much" becomes "seen it all", and "worn down" becomes "full up". 57 channels and nothin' on. One way or another, the core idea is that you don't care because the resource that permits caring is exhausted.

But the real spice in the semantic sauce is that, despite being generally negative, jadedness somehow carries a positive connotation of wisdom and experience. The jaded aren't old and bitter, they're grizzled veterans of life! Hear me, youngblood. Let my worldly condescension fall upon ye like the blankie I assume ye still keep for comfort. For I once stood where ye stand, and once thought what ye think. A thousand times a thousand stars have spun and leapt between, and though they aught but advance, O heavens, permit me this moment to escape the great escapement, to wend back upon the wheel that these hands might however fleetingly touch, and in that concurrence release this charge of knowledge: your Arab-Israeli peace idea is bad and dumb.

Aluminium

And that's really the crux of the problem with jadedness. Like so many other heinous beliefs, it manages to graft an undeserved sense of superiority onto a fundamentally limiting and self-sabotaging pattern of behaviour. I mean, obviously, the ideal would be believing helpful things, but if you're going to believe harmful things, at least have the courtesy to feel embarrassed about them.

Aluminium is a good example here. As far as metals go, it's easily one of the most deceptive. It doesn't rust like iron (that over-reactive wimp). In fact, it purports to be basically inert. But if you look closely, it's the complete opposite. Aluminium reacts alright: it reacts so strongly with oxygen that it instantly forms an impenetrable barrier of rust. In essence, it's immune to corrosion because its corrosion is so powerful it doesn't let any more oxygen in.

What a phony! Aluminium pretends to be a super chill metal like gold, which doesn't react with much because it's at peace with the universe. But aluminium's not at peace at all; it's in hiding. The only reason it doesn't react is because it's already reacted so fiercely that it ended up isolating itself from the outside world and, sure, preventing further damage, but also preventing any other interactions, desirable or not. This makes aluminium really difficult to work with and one of the most energy-intensive metals to produce, despite being abundant in nature. That last bit isn't a metaphor, just an aluminium fact I wanted to share.

As for us non-aluminium lifeforms, we are not quite so strongly bound to our corroded outer layers, and I think it's important that we seek to shed them. Jadedness isn't grunge wisdom, it's just comfortable failure: surrender dressed as superiority. If you start every prisoner's dilemma with "just so you know, I'm defecting", all you're doing is trading being a potential sucker for being a guaranteed loser. What about the possibility that things could be better? That seems like a lot to give up.

Better

So what should we be, if not jaded? I spent a good while thinking about what the opposite would look like. Start by leaving behind the drab confines of certain pessimism for the wild frontiers of uncertain optimism. Then accept that caring makes you vulnerable, but care all the same. Finally, abandon the false idol of "wisdom is experience". Experiences only make you better if they make you better. Otherwise they're just stuff that happened.

And what does better mean, then? What's this alternative vision of wisdom that isn't just experience? I think it's the way that those experiences improve or strengthen your self: your sense of who you are, what you want and what you believe.

Imagine you get some shiny new steel girders and you want to use them to strengthen your favourite bridge. One approach would be to hit it with the girders until it toughens up, but this seems unlikely to help. Another option is to attach all the girders to the bridge, but this isn't going to strengthen the bridge, just make it heavier and probably uglier. Instead, you have to improve the structure of the bridge. If the existing structure isn't made of steel, you probably want to replace it. If it's already steel, maybe you can redesign it with these extra girders in mind. If the design is already good, maybe you can use the extra girders to reinforce it slightly. Or maybe not. I mean, nobody said you had to use them all.

Over time, the bridge doesn't just keep getting bigger and louder at parties, or more and more dinged up from repeated girder impacts. Rather, it converges on a structure that works well, changing often at first but more slowly as the right approach becomes more clear. After a while, additional girders become less and less useful, until eventually you reach a kind of equilibrium where you're not, like, avoiding more girders, but it's not clear exactly what you'd do with them because the bridge you have is working great anyway. And that's wisdom! At least, I hope so or I wasted a killer bridge metaphor.

So to roll these things together: optimism, caring, and the open pursuit of the best version of yourself. That's what I think the opposite of jaded is. And the best word I can find for that is sincere.

Having drawn this distinction, it feels quite important to be sincere rather than jaded. Like the me who isn't sincere isn't someone it makes much sense to be. It sometimes feels like we live in a world that is not well-suited to sincerity. But then, perhaps that is exactly what you say when you're still a bit jaded. After all, it's not such a bad world. There are some pretty neat people in it, and some great things to do. ]]> https://samgentle.com/posts/2019-01-29-the-leverage-instinct https://samgentle.com/posts/2019-01-29-the-leverage-instinct Tue, 29 Jan 2019 00:00:00 GMT A few years ago, I wrote a post called What changes?, which included this pretty revealing observation: "Let's say you have an enormous mountain of things to do. Now you finish one thing. What changes? You still have a mountain."

It was part of an argument that meaningful transitions are good for motivation. I still basically agree with that, but I now see it as trying to make the best of some fairly limiting assumptions. To explain, I want to start with a bit of a procrastination syllogism:

• Premise: we do things because they work for us in some way, even if the mechanism isn't obvious or rational
• Premise: some people procrastinate
• Conclusion: there are desirable properties of a task that increase the longer you leave it
• Corollary: not everyone desires these properties

So when I described my quote as revealing, what it revealed is one of those desirable properties: the longer you leave something, the more important it becomes. As the deadline approaches, what changes? gets larger and more motivating. Why is Superman always racing against imminent danger, rather than working proactively to prevent it in the first place? You almost start to wonder if he likes it that way.

Let's call this the consequence instinct. You want to do big things! Impactful things! Things that make a difference! In this case, difference is quite literal: good outcomes from acting minus bad outcomes from not acting. The further apart these outcomes, the bigger the consequences, and the more important it is to act, right?

Unfortunately, this instinct provides little motivation when the alternatives are also pretty good, which also means little incentive to have good alternatives. Imagine Superman catches the bus full of orphans just before it plunges... into a giant safety net that he built months earlier? Boring.

The problem is that every option is a safety net. Do it now? Do it later? If either would still get it done, then there's no difference – no consequence – to that decision. Every hero knows that true heroism is forged in the crucible of those last days or hours, when you've blown through every safety net and the only remaining options are total victory or total disaster.

If not everyone desires this, what do they desire instead? What causes someone to seek safety nets rather than destroy them? It must be a kind of opposite to the consequence instinct: a desirable quality that decreases as the task gets closer. Amusingly, I found a hint in The Tree Planter's Paradox, a post I wrote around the same time: "'The best time to plant a tree was 20 years ago' means best time as in the time when it would have the most beneficial impact on you. On the other hand, 'The second-best time is now' uses best time to mean the time you have the most impact on it."

I wasn't deliberately dropping oblique hints to my future self; at the time, this just seemed like a good argument for making long-term consequences feel more salient in the present. But notice the consequence instinct shining through: what amazing thing is going to happen in the future if I do this? What horrific calamity will happen if I don't? But it's an unwinnable strategy: however consequential it is today, it will be more consequential tomorrow.

No, the real answer is hidden in the "second-best time": not when the consequences of the decision are highest, but when you are most able to bring about those consequences. When do you have the most resources? When can you most efficiently apply them? When are your efforts most likely to succeed? Our hero spent gigajoules of energy catching a bus in mid-air rather than 10 minutes arguing for better guardrails at a council meeting.

This is the leverage instinct. It focuses on your relationship to your actions rather than the relationship of your actions to everything else. Unlike the consequence instinct, it has a robust answer to the question "if later will work, why not later?" – because later is costlier, riskier, and less efficient. In other words, with less time left you will have less leverage over the situation.

I call these instincts because, at least in my experience, they fall below the level of explicit goals or strategies. Rather, they're heuristics for the importance of a decision or action in a given moment. In the earlier pieces I wrote, I didn't even consider how I was thinking about importance, because it was so obviously the same as consequence. It was too instinctive to seem like a factor.

These instincts connect deeply with our sense of what virtuous action even looks like. Why are we so captivated by high-velocity heroism? By the ticking clock; the red or blue wire; the one moment that changes everything?

Perhaps simply because life isn't like that. The power of our decisions is so small, and the power of time so great, that to pit them against each other is to watch an ant fight the Sun. Yet what could be more heroic than to overcome those odds? To kick back the tide and scream into the winds of entropy: "I beat you!"

"For now", it will say, "but I have all the time in the world." ]]> https://samgentle.com/posts/2019-01-12-technical-creativity https://samgentle.com/posts/2019-01-12-technical-creativity Sat, 12 Jan 2019 00:00:00 GMT I've been watching a bunch of musicians on Youtube lately. I've found it surprisingly familiar, despite my lack of any serious musical background. I think it's because of the way music so explicitly embraces the idea of technical creativity.

As an art, music is particularly technical. Unlike visual art, it never really had a representational phase; nobody comes up to a guitarist and says "hey, can you play a river?" But unlike other abstract arts, it has some pretty inflexible constraints; consider the experience of seeing a random noise painting vs hearing a random noise band.

All the depth of formalism and structure in music theory is an attempt to wrangle with the fact that making music – not good music, just anything recognisable as music – is surprisingly difficult. You can't just take what you think of and play it; you have to know the rules before you even start getting creative.

This, I think, is an essential truth of technical disciplines: you have to start with the rules. There's no point designing a bridge if you don't know civil engineering, even if you can get a civil engineer to look at it afterwards. The problem is that most shapes of bridges won't work, so the only feedback you'll get is "nope, not even close".

In this kind of low-density idea space, you need an approach that eliminates most of the options that won't work right at the start. Then, once you've narrowed things down, the remaining space is much denser and you have the freedom to explore a little.

For many technical people, just getting to that point is difficult enough that creativity, as in the voluntary search for extra complexity, is the last thing on the cards. I mean, I got the damn thing to work, isn't that enough? What? What colour should it be? I dunno, what colour does it need to be?

To be creative in such a constrained system, you have to internalise its rules to the point where they no longer seem like rules. That's how you get to just do what you think of: you stop thinking of so much stuff that won't work. In a sense, you mind moves from the low-density space to the high-density one when you internalise the filter.

And so this is what I think music gets particularly right: cursed with a low-density idea space, musicians learn theory sooner and more aggressively than most other artists. Their goal isn't to become amazing theoreticians, but to internalise enough theory that what they think of will usually work.

At that point, what's left is which workable option to pick: not a technical question, but a creative one. ]]> https://samgentle.com/posts/2019-01-02-nothing-personal https://samgentle.com/posts/2019-01-02-nothing-personal Wed, 02 Jan 2019 00:00:00 GMT Never believe that anti‐Semites are completely unaware of the absurdity of their replies. They know that their remarks are frivolous, open to challenge. But they are amusing themselves, for it is their adversary who is obliged to use words responsibly, since he believes in words. The anti‐Semites have the right to play.
  — Jean-Paul Sartre, Anti-Semite and Jew

I think part of my problem is comedy has suspended me in a perpetual state of adolescence. The way I’ve been telling that story is through jokes. And stories, unlike jokes, need three parts: a beginning, a middle, and an end. Jokes only need two parts: a beginning and a middle.
  — Hannah Gadsby, Nanette

The left can't meme.
  — Old gypsy saying

The alt-right is kind of weird. Not the people in it; the thing itself. What could possibly bring together internet trolls, Christian fundamentalists, renegade intellectuals, conspiracy theorists, anti-feminists, neo-nazis, gamers, comedians, authoritarians, libertarians, techno-utopians and archconservatives?

When you try to analyse it in terms of shared goals or values, the whole thing seems completely nonsensical. But I don't think it's those things at all. The alt-right isn't a movement so much as it is a shared pattern of behaviour.

Comedy is tragedy plus time, so the saying goes, but I would say it's tragedy plus distance. Distance applied to tragedy. Tragedy distanced. Have you ever laughed out loud at a painful memory? Ever made a joke to smooth over an uncomfortable situation? Ever fallen over and laughed on your way back up? Tragedy distanced.

Why do bullies laugh? There's no wit in what they do. No surprise. No dissonance. Just the strong inflicting suffering on the weak. If your theory of comedy can't account for how funny it is to hurt people, then it's only half a theory. Consider slapstick. You laugh because the hurt isn't real. Or the hurt isn't real because you laugh. Tragedy distanced.

The experience of playing a video game where you kill people by the thousands isn't abject horror, it's... well, nothing in particular. Fun, hopefully, if the killing is well-designed. Annoyance if it isn't. Curiosity about how well you can kill and whether there are ways to do so more effectively. But this display of rarefied psychopathy is fine, because they're not real people and they're not really hurt.

Concerned Parents of a certain era were concerned that these games would unlock violent tendencies in their children like some kind of psychopath DLC, but they had it totally backwards: we're all psychopaths when the people aren't real and the harm doesn't count. The difference is where we draw that line. Virtual people? Animals? Long-dead real people? People on the other side of the world? People you don't like?

The alt-right shares an empathy problem: not that they lack empathy, but that their empathy is incompatible with their beliefs. This requires careful management. Daryl Davis famously met with hundreds of KKK members to ask them "how can you hate me when you don't even know me?" Of course, that's exactly how. You can only hate them because they aren't real. They're caricatures. Painted faces on a stage who sing and dance and fall but never truly hurt.

And so too the whole sorry cast: the Welfare Queens, the Feminazis, the Fake Transgender Attention Whores, the Crybaby Minorities Who Act Hurt For Sympathy, the Hooknosed Conspirators Whose Dark Rituals Control Global Finance, the Kids These Days And Oh My God Karen Did You Hear What They Do At Parties Now, the Ones Who Kick Up A Fuss About Things That Aren't Serious.

Because that's the core of it in the end: not serious. The levity inherent in any ideology that demands indifference to suffering. Tragedy distanced, be it through the indirection of the internet, the depersonalisation of crowds, or the jovial denial of consequence.

So lighten up, buddy. We're all having fun here. Well, I'm having fun. But you have to admit it's pretty funny that you're not. ]]> https://samgentle.com/posts/2019-01-01-striving-is-a-sometimes-food https://samgentle.com/posts/2019-01-01-striving-is-a-sometimes-food Tue, 01 Jan 2019 00:00:00 GMT One of the silliest terms in all of software development has to be the "sprint". It's an Agile concept intended to evoke a steady, sustainable release cycle by analogy to an all-out push to get over a line. Of course, if you take it literally you're Doing Agile Wrong, but perhaps it would be easier to do right if the words meant what they said.

But forget Agile, let's talk about sprinting. The kind of sprinters who sprint with their legs rather than a keyboard tend to do so at competitive events. Everyone shows up on the day, runs as fast as they can a few times, then goes home. The fastest get to take decorative pieces of metal with them. Most competitive sports work similarly, despite it being an obviously unsustainable way to work, even by Silicon Valley standards. Why is this?

One answer is that our idea of athleticism is generally more about peak output than sustained output. A single incredible feat looks much more impressive than a lifetime of steady achievement. Even marathons, purportedly more concerned with sustained performance, only last a day or so – the blink of an eye in any other field. A true marathoner would have no regard for such trivialities; their goal is nothing less than the highest total distance on foot since birth, as measured at their death.

Extreme, perhaps, but not so far from how we evaluate excellence in other areas. A great writer isn't someone who writes one really great sentence, or even a really great work, but someone who has written many great works. This takes a lot of time, and unsurprisingly age is much kinder to writers than sprinters. That said, you also don't get many writers retiring at 30. And they don't receive anywhere near as much decorative metal.

But perhaps the entire opposition between these extremes is a fiction invented to sell conference tickets. In reality, athletes spend almost all their time in steady and sustained training, and almost none in competition. Most writers don't write at a single metronomic pace (except for Stephen King, who is a humanoid puppet controlled by a cabal of sentient typewriters). Rather, they work to deadlines and vary their output accordingly.

And a sprinter's career is not just evaluated by a single peak performance, but by many such performances over time (tragically, the rules limit you to one piece of metal per peak performance, and it's the same size even if you win by a lot). Conversely, a lifetime of okay books will never add up to one really good one; even the most prolific writers will have peaks and troughs in their career, and it's the ones whose peaks go highest we remember as great.

In fact, I'd say it's more illustrative to consider peak and sustained output as complements. Why don't sprinters always sprint? Because the foremost goal of anyone trying to be great is to improve, and improvement requires both marathoning and sprinting; both steady progress and occasional breakthroughs; both words-per-lifetime and words-per-whatever-you-can-manage-in-the-next-hour.

A reliable baseline gives you the safety to take the occasional shot at excellence, and each attempt shows you the limits of your current ability. A new peak, higher than the last. Could it someday become your baseline? ]]> https://samgentle.com/posts/2018-12-20-expertise https://samgentle.com/posts/2018-12-20-expertise Thu, 20 Dec 2018 00:00:00 GMT What does it mean to be an expert at something? Is it different from just being good at it? How can you tell if you are or aren't? And is it just a bourgeois construction designed to maintain the iron grip of privilege over the values of society?

These extremely specific questions have been on my mind, in part because the process of gaining expertise has been mostly invisible to me. Sure, I used to be bad at programming, but I started when I was 10 and back then I was bad at everything; it's hard to distinguish my development as a programmer from my development as a human.

Recently, though, I've been using more skills like electronics, that I picked up later in life, or writing, that I've done a lot but never formally studied. In some sense, I am good at these things, but the experience of doing them compared to programming is like night and day: I am not an expert.

What does that look like in practice? Well, with writing I'm at this kind of high-amateur plateau. I've written a bunch of words, and I write 'em good, but my process is messy and slow, my theoretical knowledge isn't there, and my structure can be a bit shambolic. This all adds up to a lack of reliability: I can't write to a deadline, be certain in my grammatical choices, or have a high confidence that my writing achieves what I intended.

The high-amateur plateau is common in self-taught programmers: they're often quite skilled, but the skill is fragile. You can write code, but can you write code in a way that ensures some other outcome? Can you write code fast? Can you write code while people are watching? Can you write code and be confident that it works and it's good without someone else telling you?

I think expertise is more than just being highly skilled; it's being skilled to the point where your skill becomes invisible. In that sense, it's a qualitative shift: from spelling to vocabulary, vocabulary to phrasing, phrasing to voice. To learn the higher-level skill, you need the lower-level skills to support it. To do this, they have to be so reliable that they disappear.

So the reason I am not an expert in writing is because my writing skills are not invisible. I can't really think about what I'm writing, because I'm still thinking about grammar, structure, or editing. By contrast, when I'm programming, I think about what I want my program to do, not about programming itself.

This suggests an interesting strategy for pursuing expertise. Of course, you still need to develop your skill through practice and training, but at a certain point it may be better to start learning something else with that skill so that it is forced to become invisible. The real benefit of a computer science education may be that it teaches you programming in order to teach you computer science.

And finally, it's worth wondering if we could use this to approach even higher levels of expertise. Even if I think only about what I want my program to do, that can still be very complex. What would it mean to reach a point where I no longer have to think about that, and the skill of making a program that does something becomes, itself, invisible? What would I think about instead? ]]>