Writing Elsewhere

For the moment, this blog is on indefinite hiatus. I intend to return at some point and continue to use this as a place to write more fully formed essays, but right now I’m doing my writing elsewhere.

The two main places you can go to read more of me:

At notebook.drmaciver.com I’m doing daily writing (Well, weekdaily. Weekends have anywhere between zero and ten posts per day), about whatever I happen to be thinking about. Currently a lot of what I’m thinking about is what I think of as “basic humaning skills“, but that tends to vary with my mood and current reading.

In parallel, I have just started a weekly newsletter, which will be about broadly similar subjects, and is also an attempt to create a small community of discussion around them – it’s another place to develop these ideas, and to get feedback and have discussions around them.

It’s also, frankly, a place where I’m going to get tried to get paid for my writing and all the work I do supporting it, so if you’ve enjoyed things I’ve written here, or on the notebook, I’d really appreciate you signing up for a paid subscribe account there!

And, of course, I’m always on Twitter posting about something.

This entry was posted in Uncategorized on by .

There’s no single error rate

This post is going to explain a couple of things that I often see people getting confused about. If you’ve got any background in statistics, decision theory, or similar it may be quite obvious to you – this post is for everyone else.

When people make decisions they often make the wrong ones. Hopefully you are not shocked by this information.

In an ideal world, this would not be the case, and we would always make perfect decisions. Unfortunately that’s fantasy land, and the reason is that in reality we need to make decisions based on imperfect information and a finite budget. It’s no good trying to decide on the perfect dinner if it takes you a year to do so. That way lies Chidi.

So what people often do is try to minimize the rate of errors in the time and resource budget available to them. This seems like a reasonable thing to do. Unfortunately it’s wrong.

The problem is that there is not a single rate of errors, there are in fact multiple different types of error that matter: One for each possible outcome of the decision you made.

I’m going to focus on a particular type of decision: You are given a thing of some sort (a car, a person, an idea), and you have to either accept it (decide to use it in some sense – sell it, hire it, implement it) or reject it (throw it away and no longer care about it). There are many other types of decisions, and much of what I’m going to say will apply to other types, but this is a particularly useful simple model of decision making to consider.

Let’s start with a concrete example: Suppose you are part of the quality assurance team at a car manufacturer. Your job is to certify the finished cars as safe or not (please note that I know nothing about car manufacture and this example is entirely an illustrative just so story).

Each time a car is presented to you, you can make two types of error:

  • You can accept an unsafe car as safe
  • You can reject a safe car as unsafe

Both of these errors are bad, but they are not the same sort of thing. Rejecting a safe car is expensive, but passing an unsafe car is a potentially fatal error (and also very expensive if you care more about that sort of thing).

Although you cannot ensure you never make any errors, there are two strategies you can adapt that will let you reduce one of these error rates to zero:

  • To ensure that you never pass an unsafe care, never accept any cars.
  • To ensure that you never reject a safe care, accept every car.

That is, in order to reduce one error rate to 0% you have to make the other error rate 100%: When you reject every unsafe care, you also reject every safe care. When you accept every safe car, you also accept every unsafe car.

For quality checking cars, obviously both of these are bad strategies that you should not try to adopt. However there are absolutely cases of decision making where these are reasonable strategies for reducing the error rate, they just tend not to quite match the accept/reject model. In some cases if the cost of quality checking is high and the cost or rate of bad candidates is low, it might be worth accepting everything. If you’re rejecting everything, you might as well not run the process in the first place.

One way is to assign a cost to each type of error and optimise for the cost given your resource constraints. Another way is to decide an acceptable error rate (e.g. accept no more than one in ten thousand unsafe cars) and then optimise for the other error rate under that constraint (try to accept as many safe cars as you can while maintaining that other error rate). There are a lot of mathematical considerations here, but I recommend not worrying about them. Most decisions you make you’re not going to have fine enough control over the rates the worry too much about the details of the numbers.

However, there’s a decent rule of thumb: Unless you are radically changing your process, any change that makes one of these error rates go down will make the other one go up. You can increase the stringency of your quality assurance and sell fewer unsafe cars, but this will probably cause you to reject more cars that would have been fine. You can be less nitpicky about some aspects of your quality assurance, which will mean you reject fewer safe cars, but this will cause you to sell more unsafe ones.

One specific problem that tends to come up in this model is that often you don’t know what one of the error rates is: If you pass an unsafe car, you’ll find out about it when it breaks down. If you reject a safe car it might just be scrapped. You’ll never know if it would have been fine on the road because nobody is every going to be driving it.

This is where it is particularly important to bear in mind that there is no single error rate, because although there are still two error rates, now you only see one, and it’s tempting to assume that the the other one is the same when in fact the opposite is likely to be true due to our rule of thumb: If the error rate you see is low, the error rate you don’t see is high.

This matters for two reasons. One I’ve written about before is that if the thing you’re accepting or rejecting is a person, the rate at which you reject good candidates is a measure of how unfair you are being to those candidates, and additionally can be systemically biased against some people. I won’t go into that in detail, I recommend the previous article if you want to read more about that.

The other reason is that bad rejections, e.g. rejecting perfectly cars, adds to your cost. If you are rejecting 50% of safe cars then each safe car is costing you twice as much as it needs to produce. This is expensive! Even rejecting 10% of safe cars is an 11% increase in your manufacturing cost. It may still be worth that cost, but if you don’t know what that cost is how can you tell?

So now that I’ve explained the problem, I’d like to finish with two small pieces of advice.

  1. Whenever you have a decision making process, consider what the different types of errors you can make are.
  2. If some of those error rates are currently hidden from you, try to fix that and measure them. You don’t necessarily need to change them, but until you know what they are, you can’t know if you do.

PS. The typical language used for talking about this is false positives and false negatives, but I don’t like that terminology because I find people tend to get them mixed up. Ideally use terms specific to your problems, as I have in this post.

This entry was posted in Uncategorized on by .

A Crash Course in Having Feelings

Disclaimer: I’m not a therapist, or any sort of expert on this. I just read a lot of books and have been working on it myself, and this is a distillation of some of that work. You should probably think about seeing a therapist if you need help with this.

This is a set of basic advice in dealing with feelings, primarily centered around coping with negative ones. It’s also designed to help lay groundwork for improving your experience of positive feelings, but as I am still working on that part myself I am less able to write about it!

It covers some basic skills which will help you feel a bit more emotionally stable and to lay a foundation for getting a better handle on your feelings over time.

It was based on an earlier document that I wrote for a specific friend who is highly intelligent and compassionate, but rather less kind to themself than they are to other people, and who is currently undergoing a bit of a crisis. The techniques I’m going to talk about are general purpose and not specific to that friend’s situation, but the slant is definitely angled in a way that is intended to be useful to them, and this advice will work better for some people than for others.

Philosophical Premises

Feelings arise through a complex interaction between mind and body that I don’t really understand and suspect nobody else does either (although many people understand them a lot better than I do). The experience of feelings happens primarily through the body, but there are feedback loops where the mind influences the body and the body influences the mind, so fully understanding and working with your feelings requires you to understand both (and indeed the split between mind and body is somewhat artificial).

Having feelings is good and party of a healthy, happy, life. Unfortunately it’s also a lot of work, and sometimes when our feelings get too much it can overwhelm us. When that happens the temptation is to try to suppress them, and a certain amount of that can be helpful in the short term, but unfortunately it is not a viable long-term strategy.

Instead it is better to try to develop skills for working with our feelings. The bad news is that these skills won’t magically make everything better overnight, but the good news is that they can be learned incrementally and once you start learning them things will start to move in a better direction.

There is, sadly, no substitute for doing the work, but I’m going to try to give you some pointers as to which work it is useful to do.

Feelings and Feedback Loops

It’s helpful to think of your mind in terms of systems of feedback loops – events that trigger other events, which in turn amplify or damp down the original event. When feelings get overwhelming this is usually because there is some amplifying loop that means that when something causes you to feel bad, it feeds back to make you feel worse.

For example if you do something embarrassing, and this creates a feeling of shame, which causes your mind to remember and start telling you about all of the other embarrassing things you did, which increases the level of shame you experience, which increases your negative thoughts about your self, etc. The result is that starting from a relatively minor prompting incident you are now feeling overwhelmed by a sense of shame.

This sort of amplification loop is at the core of a lot of negative experiences of feelings, especially ones that seem overwhelming, and most of this article will be about learning to deal with this and other feedback loops, by replacing amplifying feedback loops with more carefully managed ones that don’t explode and, ideally, lead to your understanding your feelings and using them as an opportunity for personal growth.

Making Your Self Safe

One of the big problems with these amplifying feedback loops is that they make your thought processes unsafe. If a thought can trigger this kind of wildly disproportionate explosion of negative feelings, you will be walking on eggshells around your own thoughts, and trying to squeeze yourself into an ever shrinking region of ways to be that your own experiences of your negative feelings are fencing you in to.

This retreat doesn’t actually make you safe. It makes you small, and afraid of growing, because all of the ways you can grow lead into dangerous territory, and you still end up triggering these explosions of negative feelings when circumstances force you to confront something that triggers them.

Instead, you need to start to try to push back and clear a large safe region of thoughts, by defusing amplifying loops that lead to distress, and coming to terms with the things that you are trying to avoid.

This is a slow, and sometimes painful, process, and you can’t do it all at once and shouldn’t try to. Even if you mastered all of the skills overnight, the best you can hope for is to clear these things as you encounter them, and even once you clear space to be safe it will take a while before you really believe you are safe.

Fortunately, feeling safe doesn’t require fixing everything, it just requires room to maneuver and the belief that you can fix problems when they occur. Equally fortunately, fixing things is good even if you can’t yet feel safe – every time you find a problem and confront it, you become slightly safer in your own self even if you don’t yet feel that way. 

How to Retarget a Thought Process

The first thing you need to make safe is the process of working with and influencing your own thoughts. If you’re trying to work on improving your thoughts, punishing yourself every time you fall into old habits is incredibly unproductive and creates a new amplifying feedback loop: You have a negative thought, you respond to it badly, you think “Oh no, I’m not supposed to do that! Why am I such an idiot?!” and the whole thing escalates.

The main technique for avoiding this, which I got from mindfulness practice (I’m otherwise very bad at mindfulness, but this technique is super useful), is to treat noticing you are off course as a good thing rather than treating being off course as a bad thing. Whenever you notice your thoughts straying into a format that you don’t want, the thought process should be:

  1. Oh, that’s the thing I was trying to avoid doing (in mindfulness this is straying from your focus object – e.g. your breathing).
  2. Good job for noticing, me!
  3. I will now do the thing I am supposed to be doing instead (focusing on the breath).

There is no negative judgement, and you don’t treat your thought process as bad or punishable, you just refocus onto something that is more in line with what you want.

With amplifying feedback loops for negative feelings, this might look like:

  1. I see I am punishing myself for doing a bad thing.
  2. I am glad that I have noticed that I was doing this! I will stop now.
  3. Instead I will… (more on this in the next section)

It’s basically inevitable that you will get things wrong, especially in the early process. That’s OK, that doesn’t matter, because in and of itself getting things wrong is no worse than the status quo, but punishing yourself for it will be. On the other side, every time you get something right is life improving, and every time you notice that you need to change something is a positive step towards that which you can be glad for.

How to Talk to Yourself

The way you talk to yourself is one of the key drivers of the amplifying feedback loop around unpleasant feelings, and the most effective thing you can do to start to get a handle on your feelings is to break that part of the loop. This is what cognitive behavioural therapy (CBT) teaches you, and if this is something you struggle with I’d definitely recommend taking a course in it or reading a book about it as well as this blog post.

The basic core observations of cognitive behavioural therapy are:

  1. You can choose how you talk to yourself.
  2. You should be nice to yourself.

Use the retargeting skills to start to control your inner monologue, and instead of punishing yourself for your faults, start to try to have a constructive conversation with yourself.

It may help you to think of yourself as split into two parts talking to each other: A thinking and feeling self (this split isn’t real and thoughts and feelings are highly intertwined, but it is a useful guide). Your thinking self is quick and highly verbal, your feeling self is slow and not very good at words. When you engage in self talk, your thinking self is mostly speaking to your feeling self, which can listen but cannot easily talk back.

Your feeling self is not stupid, but it is child like, and you should talk to it as if it were a child in your care, who you love and who trusts you.

As with a real child, your feeling self is not necessarily right, and indeed is often wrong or behaving badly, butO shouting at it or lambasting it for its bad behaviour is not going to help and will just make the problem worse. You can’t even explain to it why it is wrong, because it’s mostly preverbal. Instead you need to be careful and patient with it, gently correct it by showing it the right way, while taking its needs seriously even though it cannot currently express them in a way that you understand.

How to Listen to Yourself

Unlike a child, your feeling self will never grow up and become verbal and able to tell you what it wants. You have to learn to listen, and pay attention to it, and the way to do that is mostly to pay attention to the physical sensations in your body that manifest as you experience feelings. The main tool I use for this is something called Focusing (and if you want to learn more about it I recommend the Focusing book, though it’s a bit culty sounding).

Focusing is a way of paying attention to your body in order to understand your feelings. It works as follows:

  1. Think about the problem you want to understand. Ideally describe it out loud. Pay attention to how your body feels while you are doing this. This is the “felt sense” of the problem.
  2. Try to come up with a short description of that bodily sensation – e.g. tense, exhausted, excited… 
  3. Ask what it is about the problem that feels this way. Try to describe it, again, ideally out loud. Pay attention to how you feel in your body as you do, and use the changes in how it feels to guide you towards a description that “feels right”.
  4. Your goal is to try to trigger a body shift – a change in the felt sense of the problem that feels like becoming unstuck – by examining the sense of the problem until your understanding of it changes.

If you struggle with this it may be worth practicing some more general body scanning techniques, paying attention to how you feel without worrying about the emotional content.

In general Focusing teaches you to use attention to your body as a diagnostic tool – it helps you learn to pay attention to what you are actually feeling in your body, and use that to start to unpack some of the emotional content.

Once you have this skill you can use it to influence your self-talk, and your planning: Focusing lets you find out why you feel a certain way, which allows you to help guide your feeling self into healthier patterns and habits in the circumstances that lead to that feeling, or to reassure it and help it feel differently.

Putting It All Together

The key to change and recovery is to break feedback loops that hurt you and introduce feedback loops that help you heal. The mindfulness skill of retargeting thought processes is key to this because it lets you safely intervene in your feedback loops.

The CBT skills are then the foundation because they essentially stop you picking at the open wound: They let you reframe how you talk to yourself, moving from punishing yourself for feeling badly to creating a neutral state which gives you space for recovery.

This creates a safe space within yourself to be able to look at your feelings. As long as looking at negative feelings can cause you to spiral, your feelings are unsafe for you, and so you will be tempted to ignore them and repress them. Once you can look at your feelings safely you can begin to actually heal, by using Focusing to understand those feelings you can start to understand and unpack them, and this can feed in to your self-talk and your self-talk can use Focusing to shift your feelings.

All of this takes time and care, but progress tends to come in fits and starts, and the basics are all relatively accessible practices that will let you reach a lot of low hanging fruit.

This entry was posted in Uncategorized on by .

Open to Interpretation

Let me tell you some stories.

You are walking through the countryside, and meet a shepherd tending to his sheep. You strike up an idle conversation, in the course of which you ask him how many sheep he has. “Oh, quite a few” he replies. Asking further, you learn that he has more sheep than his brother, but they both have fewer than their father did, and that the largest flock in the area belongs to their cousin who has more than them both combined. What you don’t learn is how many sheep that is.

After further discussion it emerges that there is a simple reason for this: The shepherd is innumerate. He has never learned to count. So the reason he is not answering your question is not that he doesn’t want to, it’s that he does not have the concepts which would allow him to understand the question as you mean it, let alone to answer it.

You offer to teach him, and he accepts, and you spend some time teaching him a Yan Tan Tethera based system. It gets late, and he offers to let you stay the night, and as the sheep are put away in their enclosure for the night you help him count them, and in the morning he counts them as they leave. The two numbers are the same, a fact that seems natural to you but comes as something of a surprise to him.

After this, the farmer’s world has changed. Very little has changed about objective reality – “just” some new knowledge in one man’s head – but to him the world has a fundamentally different character. Things in the world have a property they lacked before: quantity.

This does not just let the farmer answer questions like “How many sheep do you have?”, it lets him answer questions like “Have you lost any sheep?” more easily, and it lets him answer questions like “Do you have enough food for your sheep over winter?”. Questions that were perfectly legible but difficult to answer now become possible to solve with his new concepts.

A boy was raised in a strict religious environment. As he comes of age, he starts to feel different from the other boys his age: They are becoming interested in girls and he is… not. The girls he knows seem fine, but they hold no special attraction to him. He is, however, much more interested in watching the other boys, and does not understand why this doesn’t seem to be a widely shared interest…

At this point the story diverges as one of three different things happen.

In the first branch of the boy’s history he confides in a close friend at school, someone from a different background. The friend in question shrugs as if it’s no big deal and says “Maybe you’re gay?” and goes on to explain that some people are straight, and thus mostly attracted to people of the opposite gender, some people are gay, and thus mostly attracted to people of the same gender as them, and some people are bi, and are regularly attracted to people of a variety of genders and it’s perfectly normal to be one of these.

At this point the boy’s world is changed. He has a word that describes what he is, and this lets him find out more, and connect to other gay people. He keeps his sexuality a secret from his family, knowing from talking to others that they will not approve, but eventually moves away from them, and after several relationships of varying lengths finds a man who he marries. They live happily ever after.

In another life he had no friend he could talk to about this, and so talked to his priest. This seemed to him a reasonable choice – the priest was a kind and good man to those who fit within his worldview. This kind and good man explained to the boy in his compassionate, fatherly, voice that his desires were a perversion, sent by Satan to tempt him, and that god would forever despise him if he gave into them and became a true pervert. As such, it was his duty to resist temptation and to pray to god for strength.

The boy nodded solemnly, and went home to pray. Eventually through his increased devotion, and his desire to resist temptation, he joins the priesthood. His life is not an unhappy one, but he always feels there is something missing.

In the third version of his life, the boy tells no-one. He is clear that he is different, and does not feel that his difference is wrong, but he knows how his community will react and hides it from them. Eventually he marries a woman, who he does not love, and they have children together, who he does, but his marriage is made miserable by the sense of wrongness he has of it, and while his life is not without happiness it contains significantly more misery.

A student came to me with a maths problem that they were stuck on. I showed them how to look at it, and then they solved it. I taught them nothing new in doing so, but I guided them through applying the tools they already knew until the problem made sense.

A friend comes to you. They have had a fight with another friend, and want to talk about it. You listen thoughtfully, validate their feelings, and ask them questions about their experience.

You have not taught them anything new, but at the end they feel like they understand the situation better than they did before, and are happier for it.

There was a woman. She was a looter, and will be charged as such.

That is to say, when the flood came and she was trapped in the city, she broke into a vacant store in order to steal food for her children.

The man who murdered her husband when they were running to escape the rising waters was of course an upright citizen defending his property.

There was a woman. She was very good at her job. She was also black. The white men who she worked with did not believe she was good at her job. They never made any overt reference to her gender, or her race, but they were constantly asking questions about her work that they would only rarely have asked eachother and minimizing her achievements.

She complains to HR, and they explain to her that everyone gets these sorts of questions and that is her responsibility as an individual employee to showcase the quality of her work. She attempts to explain about systemic racism and the aggregate effect of the different levels of demands placed on her as a black woman, but the person from HR holds to a theory of individual action and refuses to acknowledge this complaint as valid. Accusations of racism are dismissed: Race has never been brought up by the coworkers, and they’re not bad people, therefore they can’t be racist.

Later, stories circulate the office about how she is paranoid and making up conspiracies against her in order to be the center of attention.

There was a child. They were lazy, and never did their chores, or their homework.

They felt immsense shame over this. They knew they should do these things, but they couldn’t. They started, and were distracted by something else, or they intended to and they forgot, or they tried to but felt such overwhelming anxiety at the idea that they couldn’t even start.

Which is to say, they were lazy and then made excuses for that laziness.

When they grew up some more, they learned about ADHD, and considered that they might have it. They got a diagnosis, and medication, and a lot of their “laziness” improved. It didn’t make like perfect, but it made it manageable.

They still feel a great deal of shame about how lazy they are, even though they know that it’s ADHD.

You go to a friendly coworker, and explain how you are being sexually harrassed: Another colleague is visiting on you all sorts of unwanted sexual attention, and does not stop despite your asking him to.

Your friend asks you why you cannot just take it as a compliment, and the conversation takes a very different turn than you were hoping. You eventually manage to explain the problem to him and teach him enough feminism 101, and in the end he seems to understand and acknowledges that your experience is bad, but by then it is clear that he needs to learn a lot more before it is possible for him to help you out.

I travelled to America recently. When I did, my sleep was disrupted. I couldn’t understand why. They told me that it was because of jet lag – the time was different here because of the rotation of the earth and when the sun rises at different points of it.

This explanation was obviously nonsense. The world is flat after all, so how could the sun rise at different times in different places?

All of these are stories about interpretation. Many of them have other significant ethical features, but they all have a core that is about how we interpret the world.

The world exists, and we inhabit it, but we also interpret it, turning our rich experience of reality into concepts that we can relate to and understand. When we see a sheep, we experience it as a sheep. When we see a flock of ten sheep, the concept “ten” is something we experience as a property of the sheep, even though it exists only in our heads.

Interpretations can be tied to theories, and those theories can be right or wrong: The earth is not flat, so any interpretation of facts that I base on it being so will likely result in wrong conclusions. Thus interpretation is constrained by reality, but it is not determined by it: A situation can be interpreted in many different and mutually incompatible ways, but not all ways are compatible with the world we live in.

Which of these interpretations we choose is not value neutral. The same person may be lazy or have ADHD, may be gay or a pervert. The choice between these two is not solely one of verifiable facts (although there are factual differences), but one of the values and subjective experience: What is considered right and wrong, how the world seems to those involved.

These interpretations have consequences. An interpretation transforms your world, which may uplift you or may cause you shame. It might cause you to exonerate someone you shouldn’t, or condemn someone you should. It might let you solve problems you otherwise could not, or it might prevent you from solving problems by hiding the truth from you.

Interpretation is not just about concepts, it’s a social act. I can help you interpret a situation even if I never tell you anything new. You can share your interpretation of a situation with me, which I can then either accept or reject.

By accepting and building on each other’s interpretations, and by working together to interpret more richly and deeply than we can on our own, we create a shared understanding of the world.

A theory-driven, value-laden, shared understanding of the world, with significant consequences for those we share the world with.

When building anything social, the question we must always ask is this: Who gets to participate, and are they all treated as equals?

Although I talked about “our” shared interpretation of the world, there is no single “us”: There are many groups of people, some disjoint, some overlapping, each sharing interpretations, and different groups may interpret the same events very differently.

Even within a group, not all of a person’s interpretation will be shared: Trying to explain racism or sexism, or your needs as a person with ADHD, to your colleagues will ofen quickly put the lie to the the idea that your company has a single shared way of looking at the world.

How do we ensure that we interpret the world in a way that works for everyone?

References and Further Reading

Although I have not explicitly used any of their terminology, this is primarily a post about epistemic injustice, a concept coined by Miranda Fricker in her book of the same name. My favourite piece on the work is Kristie Dotson’s paper “A Cautionary Tale: On Limiting Epistemic Oppression”, which is a response to Fricker’s work but also stands well on its own as an introductory piece. The types of epistemic injustice talked about in this post are mostly “hermeneutical” and “contributory”, but also contain other aspects of the ethics of interpretation that I don’t think fit cleanly into either of these (some of which are not even obviously epistemic injustice issues: The looting example is not about the injustice done to them in their capacity as knowers, it is about the injustice done to them by other people’s ways of knowing).

To a lesser extent, my presentation also draws on Rebecca Solnit’s “A Paradise Built in Hell” (especially the example of why “looting” is a problematic category), and Will Buckingham’s “Finding our Sea-Legs”, for making me think more about the impact of the stories we tell – both an instance of interpreting the world in its own right, and the basis of the structure of this blog post.

This entry was posted in Uncategorized on by .

Advice to new speakers

Public speaking is a great skill, and one that I strongly recommend people develop. The ability to get up in front of a crowd and speak to them is surprisingly useful in its own right, and besides teaches you a lot about communication and confidence.

Public speaking is also terrifying for new speakers, and as a result they often try to play it safe. This is entirely reasonable – when you are nervous, making your environment safer is the right thing to do. Unfortunately, new speakers are often wrong about what makes a talk safer.

Here is what you as a new speaker should do to actually make your talk safe:

  1. Under no circumstances read your talk from a written version of it. This includes reading your slides.
  2. Pick as short a time slot as you can find. Lightning talks are ideal. Twenty minute slots are fine. Under no circumstances give your first talk in an hour long slot.
  3. Practice talking about that topic repeatedly, refining your talk and slides until you are 100% confident that you can fit it in the time slot.
  4. Wear a watch on stage so you can keep track of time.
  5. Pick a niche topic that you find very interesting and that it’s unlikely many people know much about.
  6. Don’t take questions.

In contrast many new speakers pick an easy topic, read from their notes, and overrun their time slots. These are about the only things you can do as a new speaker to lose audience sympathy1. They can tell if your speaking skills aren’t polished, but they won’t care much – everyone either has been a new speaker or hasn’t yet given a talk and is very impressed that you are. If you keep them interested and don’t inconvenience them, the talk is a success.

In my post on satisficing I listed the following as the success criteria for giving a talk:

  1. It has to make sense.
  2. It has to fit its time slot.
  3. It has to convey something interesting.

I should add a fourth one for new speakers: It shouldn’t stress you out too much to give it. The above advice is designed to make it as easy as possible for you to satisfy all of these requirements.

I’ll now elaborate a bit more on how it does this.

No Reading Your Talk

The reason you should not read from a prepared version is that it will make your talk extremely hard to follow. It’s not impossible to give a good reading, but it is a legitimately very hard task, and you almost certainly don’t have the skill set to do it.

Doing this well requires two skills:

  1. Having a natural voice while reading out loud.
  2. Writing speech that sounds natural.

Most people are not very good at either of these.

Reading out a prepared speech will usually cause you to speak in a flat monotone, with minimal variation in pitch and speed, no pauses, and extremely restricted body language. This is almost impossible for a listener to understand, because all of those things are major cues that we use to understood spoken words.

Avoiding this is harder still if your prewritten talk is not actually well written for being spoken, which it probably isn’t. Writing is more measured, and tends to be more formal and stilted when read out loud. Most writing would be improved by making it suitable to be read out loud, but you don’t want to make your public speaking dependent on improving your writing skills.

In contrast, if you design your talk to be spoken from the start, you will avoid both of these pitfalls automatically: You already know how to speak, and you can lean hard on that and refine it further.

This doesn’t mean you speak without planning – rehearsing your talk is very important – but it does mean that your talk will be partly spontaneous and will not be the same each time you give it. That’s fine, that’s how talks work.

You can of course prepare cue cards if you’re worried about losing track of your talk, and you can read out anything you need to quote. I don’t personally do this, but if it’s something you find helpful then it’s a very different proposition to reading out your whole talk.

You can also write out a version of your talk if you would find it helpful for getting your thoughts together and planning it out. I do this occasionally. The important thing about doing this though is that you write it out and then don’t use it – the process of writing it is what was important. You are not giving the talk by reading it, or even attempting to memorise the wording you used in it.

Practicing Time Management

Time management for talks is a hard skill and I would like to give you an easy way out of it as a new speaker, but I can’t.

Unfortunately, it’s also a vital skill set: You are obliged to fit your time slot, and you will make things worse for everyone if you do not. As long as you fit in your time slot, everything else is forgiveable, but overrunning at best inconveniences everyone and at worst significantly disrupts the schedule, and you will make the organisers mad at you.

As you get more experienced as a speaker you will (if you pay attention) get better at judging time of talks and it will become more automatic, but in the beginning there is no substitute for practicing your talk.

The specific advice I gave for time management was:

  1. Pick a short time slot.
  2. Practice your talk until you are 100% confident that it fits the time slot.
  3. Wear a watch on stage so you can keep track of time.

All of this is designed to make your life as easy as possible while getting good at this vital skill.

Short time slots are much easier to get right, because it’s easy to pad or cut a minute or so, but if your timing is off you will tend to drift in a systematic way over the course of your talk. If you’re 5% off on a 20 minute talk then you have a minute to correct for. If you’re 5% off on an hour long talk, you have three minutes to correct for. This may not sound like much, but rushing through the conclusion of your talk in the last thirty seconds when you thought you had three minutes left is incredibly stressful, and the conclusion of the talk is the bit people will remember most.

Wearing a watch on stage also helps you keep track of timing so you can adjust on the fly – if you notice halfway through your time slot that you’re not quite yet halfway through your prepared material, you can adjust by speeding up a bit, cutting digressions, etc.

Pick A Niche Subject

It is possible to do an interesting talk about something fairly commonplace and widespread, and it’s possible to do interesting and useful tutorial talks, but it’s quite hard, and my recommendation is to wait on it until you’re more comfortable speaking. Even today I tend towards niche subjects, because it guarantees an easy win.

If people come out of your talk having learned something interesting and new, you won at speaking. If your subject is something interesting and niche, this is practically guaranteed because by definition a niche subject is something that not many people know about.

There are a bunch of ways to find good niches:

  • Do a talk about something you’ve personally experienced – e.g. interesting debugging war stories.
  • Dig into the details of something commonplace to a much greater degree than most people know about. e.g. talk about an aspect of language internals, how some sort of software works.
  • Pick a weird interest of yours, or something that you’ve learned in the course of a specialised job.
  • Talk about something from another field you’re in and how it relates to the conference subject (or don’t bother to relate it if the conference has a relaxed policy on talk subjects!)
  • Talk about a skill you wish were more widespread – e.g. communication and writing skills.

It’s a little hard to give specific advice about this because everyone is different in what niches they know and like, and what they’d feel comfortable talking about, but my recommendation would be to write down a long list of things you could maybe give talks about (aim for quantity not quality: Don’t worry too much about whether it’s a good idea until you’re actually ready to submit it to a CfP!) and then shop some around friends and coworkers to see what they think: The ideal reaction is a faintly puzzled “Huh. Yeah, I guess that would be a good subject for a talk.”

Don’t Take Questions

At some point in your talk (probably the end), say something along the lines of “I will not be taking questions at this time, but if you want to talk to me about this I’ll be around for the rest of the conference”. If you don’t feel confident saying it, that’s fine, ask your session chair to say it for you.

It’s a bit of a power move, so it makes sense to get help, but it’s one that is absolutely the right thing to do for most talks, and by doing it for your talk you encourage others to do the same for theirs.

The reason for doing this is twofold: Firstly, answering questions on the spot is hard and stressful, so why do it if you don’t need to? Secondly, most people are bad at asking speakers good questions, so the Q&A session is generally low value and is safe to ditch.

Note that you will probably not be able to pull this one off at an academic conference. It’s genuinely fine at most industry ones.


Public speaking is hard, and it’s OK to be stressed about it – it’s practically expected. Do whatever you need to do to make your life easier. Try to stay calm on stage if you can, but don’t worry about feeling nervous if you can’t.

Ultimately, the audience is on your side, and nothing too bad will happen even if things go wrong.

Your goal as a first time speaker is to give a pretty good talk, and to become a second time speaker, and a third, and a fourth, and so on. Take any affordances you find helpful to get to that. Those can be the ones in this post, or they can be anything else that works for you.

This entry was posted in Uncategorized on by .