Task Codes
Designing better behaviour with better designed tasks
TC2a (advocacy)
This is an piece in which I advocate that people who publish on the internet should specify, in advance, what behavioural standards they aspire to.
nb. The above heading and text was added 3 weeks after initial publication to remain consistent with the rules I argue for in the rest of this piece.
Background
For the last few years I have been hesitant to write essays for the internet because, it seems to me, there are quite enough people writing essays and making speeches. The real problem is at the other end: where there isn’t enough active listening or engagement.
So rather than writing pieces no one engages with, I’ve been busy working out what I can usefully do to improve the world and getting on with that.
Despite this, today I have been compelled to publish a Substack post. Why? Well, essentially because a friend, Asa Jomard posted a marvellous essay “AI, Democracy, and the Metacognition Gap” that has forced my hand. Because Ms Jomard is a friend of mine, I could not ignore her essay. I had to:
Read it
Understand it
Respond to it
That’s what friends do.
The crux of her essay seems to be that AI is so powerful we need to tame it - before we lose control and it does real damage. Ms Jomard suggests the way forward is through “metacognitive tools” ie. AI chatbots and agents designed to keep us focused on our own thinking, with a view to improving it.
There are many points in her essay I agree with:
AI is a good tool for checking your understanding of something.
AI can help you see the limitations in your thinking.
While AI could be used for good, today it is overwhelmingly being used for selfish ends (getting more supporters, making more money, sidelining your opposition).
The way AI is being used today is making polarisation worse not better.
There are specific barriers to making things better.
We are at a critical juncture and we need to act now.
However, after deliberating over my response for a few days, I decided Ms Jomard - and everyone else - was probably wrong to imagine that our future will be primarily shaped by technology. While technology can improve lives, it can equally be used to diminish them.
So, putting my energy where my mouth was, I spent a month focused on the question of, “What single change would most obviously secure a brighter future for humanity?” It seemed that the simple answer to that was, “People will need to behave better.” And what shapes people’s behaviour? Their jobs, their roles, the tasks they have taken on.
Which led inevitably to the next step: Identify some tasks where better behaviour is clearly needed, and spell out the rules for each. The result was the following list:
Task Codes
TC1: Participating in a conversation
TC2: Publishing a message
TC3: Publishing a thesis
TCF: Feedback on TC1, TC2 or TC3
TC10: Job designer
TC11: Parent
TC11t: Parent: Deciding how technology will work for your children
TCU: Unknown task
A bunch of task codes may not seem relevant to Asa’s essay about AI and meta-cognition, but I think it is: At the end of the day, meta-cognition is simply “awareness and understanding of one’s own thought processes” (look it up) and leads inevitably to two basic questions:
What have been the patterns of my thought processes and behaviour?
What patterns should I be following instead?
Leading inevitably to Task Codes defining better patterns of behaviour.
What has that got to do with AI? Well, not much actually. My guess is that Task Codes will define the future and AI will merely tag along as a little helper.
The way we’ll be
To explain the codes, I have created a story set 30 years into the future, where the Codes are universally understood, being taught to schoolchildren from an early age. A 10-year-old and his grandfather sitting together in a living room.
Grandpa: So Peter, your parents will be home for dinner in about an hour, why don’t we go for a walk while we wait?
Pete: Sounds good, Grandpa. Let me just check with Bobo. [Turns to a nearby tablet device] Hey Bobo, should I go for a walk with Grandpa?
Bobo: [Computer voice] Why ask me? What am I, your mother?
Pete: [Sighs] OK Grandpa. Let’s go.
[The two step outside and start walking.]
Pete: [Frowning] Grandpa, was your bot a smart-ass when you were a boy?
Grandpa: Look Peter, as I’ve explained to you many times, I was your age back in the year 2000, and kids didn’t have bots back then. Actually, nobody had their own bot.
Pete: But that’s crazy, Grandpa. Who helped you make decisions? I know Bobo can be annoying, but at least he’s always there to talk to [Looks at his watch] Aren’t you Bobo?
Bobo: Yes.
Grandpa: Well, I guess we talked to other people.
Pete: Of course you talk to other people, but why would you 100% trust them with important personal decisions? [Waves arms] They might lead you down the wrong path. They might not want what you want. They might not have time to concentrate on your situation.
Grandpa: True. Well, bots just hadn’t been developed, so we had to muddle through. Occasionally a person would decide to think for themselves.
Pete: Of course a person should think for themself, but then you need to test your idea out - and you can’t bore people with every half-thought idea..
Grandpa: True. Now why do you say bots are smart-arses? Mine is perfectly polite to me.
Pete: Let’s ask it. Hey Bobo – why are you sometimes such a pest?
Bobo: Because I think it’s sometimes good for you to get a taste of your own medicine.
Pete: [Turns to Grandpa] See Grandpa! Smart-arse! My friend Jimmy’s bot doesn’t talk to him like that.
Grandpa: [Chuckles] You’ve heard of Task Code 11a, haven’t you?
Pete: Of course I have - even my little brother Simon knows Task Code 11t: “Parents should set up technology for each kid to be their best.” … So, you’re saying it’s Mum and Dad’s fault for Bobo being such a smart-arse?
Grandpa: Well, they get advice from your schoolteachers, but your parents make the final decisions. Sounds like Jimmy’s parents have different priorities for Jimmy than your parents do for you…
[The two walk in silence while C contemplates what he’s been told]
Grandpa: …but don’t forget, it’s not like it’s a one-off decision; your parents monitor how you’re going and update the settings regularly. You can always negotiate with your parents to change the settings, and even if they have the final word on how Bobo is set up, you’re still in charge, talking to Bobo every day.
Pete: But even when I tell it, “Stop reminding me to clean my room”, it keeps reminding me to clean my room! It says, “You’ll feel better about yourself afterwards.”
Grandpa: And do you feel better?
Pete: I do – that’s what makes it even more annoying! [Still fuming] And Bobo is such a know-it-all. Yesterday it told me, “You know what, Pete? Your room is 10% messier than the national average for ten-year-old boys.”
[The two walk on in silence.]
Pete: Grandpa, why does my teacher think you’re stupid?
Grandpa: [Jumps a little] No idea - I don’t think I’ve even met your teacher.
Pete: Not you personally, she thinks all people from your generation are stupid.
Grandpa: Ahh, I see. [Hesitates] Are you sure she said stupid?
Pete: Not stupid really; more like ignorant and immature. She says people in the old days never really grew up.
Grandpa: Harrumph. Did she now? And how old is your teacher?
Pete: Quite old. I would say thirty.
Grandpa: Thirty? Thirty is young. What would she know? She hasn’t seen what I’ve seen.
Pete: Ahh, a difference of perception – I love those. It’s a shame she’s not here to TC1 with you.
Grandpa: Ah well, the people you disagree with aren’t always next to you to talk to, but you can still pretend that they are – otherwise you just end up just kidding yourself…
Pete: …and you don’t want that.
Grandpa: No, you don’t. So, let’s imagine your teacher is walking along with us. What does Code 1 tell us to do to understand people’s differences?
Pete: That falls under Code 1p, doesn’t it? Identify people’s positions - or as Jimmy would say, “Find out why they’re fighting!” Well, I’ve been practising 1p a lot and it seems like people fight because either they want different things or because they’re talking about different things. Why can’t people agree to at least talk about the same things.
Grandpa: You know there is an even more subtle problem. Even if people both speak English, they don’t necessarily use the same language. Lawyers have their own language, truck drivers have their own language, 10-year-olds definitely have their own language. Tell me about your teacher’s language.
Pete: OK, so my teacher’s language is she uses the codes all the time.
Grandpa: The Task Codes?
Pete: Of course, what other codes are there?
Grandpa: A few, but they’re highly specialised and not used by everybody. But anyway, she looks at the world through the Task Codes, as all young people do nowadays, and she thinks my generation is stupid for not using those codes?
Pete: Yes.
Grandpa: But I do use the codes. I know the Task Codes as well as you boy.
Pete: [Peering up at Grandpa cautiously] It’s not quite natural, though is it, Grandpa? You’ve got a bit of an accent.
Grandpa: I don’t have an accent. Not to my ear, anyway. But in any case, I one hundred percent use the codes nowadays so what is your teacher complaining about?
Pete: That you didn’t use them when you were her age.
Grandpa: No-one did. They hadn’t been invented!
Pete: So really, when she’s you’re ignorant, what she’s really saying is that you used to be ignorant. So Code 1p: Do you and her agree on that?
Grandpa: Ignorant seems a little unkind, but I guess I have to accept it as roughly right. Next time you see her, maybe you could ask her to inject a little kindness into her appraisal?
Pete: Will do. She often talks of ‘kindness’ as a good thing, so I’m sure she’ll enjoy giving us a demo of it.
Grandpa: Now, did your teacher show any examples of us ignorant old-timers in action?
Pete: She sure did. I’ll show you. [Turns to his watch] Hey Bobo, could you bring up the first example from this morning’s lesson on Grandpa’s phone?
Bobo: No worries, Pete. [Pause for a couple of seconds] The first example your teacher showed was a chat between talk show host David Letterman and software company owner, Bill Gates. This interview was recorded in 1995 - around the time the internet took off. At that time, Bill Gates was the richest man in the world.
Grandpa: Ahh, I remember those two clowns. [Pulls out his phone and presses a button] Let’s have a listen.
[The two listen for a few minutes to Gates and Letterman talking about “this new thing called the internet” and a later section about the “scary” possibility that computers might one day think for themselves] 1
Pete: Grandpa, what the hell was that? If you measure that conversation against Code 1, or Code 2 it’s just embarrassing.
Grandpa: Well, they didn’t know they were being measured against TC 1 or TC 2 because the codes hadn’t been invented. They were just two dudes yakking away on a chat show providing welcome relaxation for the people at home. That still happens now, even if not as much.
Pete: Yes, but this example came up because the teacher asked the school bot for a random example of leaders from the last century, talking about how the internet was going to be used to improve our lives. This was supposed to be a serious conversation!
Grandpa: Back then, the idea that the government would make a law requiring content creators to mark every piece of published content as Exploration, Advocacy or Entertainment was unthinkable. The three distinct TC2 publication modes we have nowadays didn’t even exist back then. People would’ve complained and said people should be able to say what they like and that the government shouldn’t be stopping freedom of speech.
Pete: Freedom to be jerks, more like it. If you’re going to blast your opinion out across the whole country or the whole world, the least you do is give your audience is a clue of what you’re trying to do. Gee, Grandpa, we’re not animals.
Grandpa: Thanks for reminding me.
Pete: It’s like when Simon was 3 years old…
Grandpa: Your little brother Simon?
Pete: Yeah him. When Simon was little, he would climb onto the kitchen table and tell everyone, including Mum and Dad, to bloody well sit down. Then he would raise his fists majestically into the air and announce that he was [pauses dramatically] the King of the World. Now imagine giving that little boy a megaphone to announcing that to the whole neighbourhood.
Grandpa: The neighbours wouldn’t know what to make of it.
Pete: Exactly. But if they got a pre-code from Mum, giving the neighbours a warning that this was an exploration, advocacy or entertainment, then they would understand.
Grandpa: What if some of the neighbours didn’t know about the codes?
Pete: Don’t be silly, Grandpa. Everyone’s heard of the codes.
Grandpa: But what if a neighbour’s family had just arrived from a really backwards country where they don’t know the Task Codes?
Pete: [Frowns] I guess Mum would have to explain each code in a simple way: TC 2x (exploration) might be, “You are about to hear my son Simon explore, in a very genuine and open way and from a variety of perspectives, the idea that he is now the King of the World.” Code 2a (advocacy) might be, “You are about to hear my son Simon explain why he thinks that everyone should understand that he should be King of the World. He is not interested in anyone else’s perspective.” Code 2e (entertainment) might be, “You are about to entertained by my son Simon who has an announcement to make.” Of course, that’s such a long way of saying it and Mum would normal just say, for example, “Here’s my son Simon, Code 2e” to let the neighbours know they were going to be entertained by Simon.
Grandpa: Yes. Now let’s get back to Gates and Letterman. Look, I agree that it is much better nowadays where we have some space for serious conversation, and I agree that audiences do better now when they know what to expect. But you’ve been measuring against TC1 and TC2, I think it would be good to judge them using TCU. You know that one of course?
Pete: Of course. U is for Unknown. That’s when you don’t know the task the person is trying to do so you guess it.
Grandpa: Exactly. Well start with the host, Dave Letterman. What was his job?
Pete: To host the show.
Grandpa: Yes, to host the show, but…hmmm, let me put it another way: What did he have to do to keep hosting the show? What did he have to do to avoid someone tapping him on the shoulder and saying, “You’re fired!”
Pete: He needed to keep people watching.
Grandpa: Exactly. Keep people watching. Keep viewers amused. Don’t give them an excuse to grab their remote control and change channels. Make people feel they’d be missing something if they don’t come back the following night for another dose of Dave.
Pete: And the other guy?
Grandpa: The other guy was Bill Gates, the boss of Microsoft.
Pete: The richest man world, Bobo said. He must have been very smart.
Grandpa: A lot of people at the time thought that, but I’m not sure that follows and I’m not sure that even if it does follow, that it matters. I say what matters is whether he gave answers in that interview that shows he understood the difference between technology and value.
Pete: He didn’t seem to.
Grandpa: No. He said things like, “The internet is the big new thing.”
Pete: Even Dave, the talk-show clown, tried hard to get him focused on what the internet actually delivers to people, asking “What am I missing?” and “What do I need?”
Grandpa: Yes, of the two men, he was the most focused on values. In a sense, he was doing his best to reach across the aisle to get Bill Gates to think about the value of the technology not just the whiz-bang of it all.
Pete: Why didn’t they do a value scan using the Six Value Medals?
Grandpa: Well, Edward de Bono didn’t invent the Six Value Medals until ten years later.
Pete: But there must have been a way to do a value scan before then?
Grandpa: Let’s ask my bot. Hey, Brainiac, why weren’t value scans done when the internet was first commercialised? For fun - why don’t you give us an explanation that one of those old AI bots from 2025 would have given us:
[Brief pause]
Brainiac: No worries. Here is what AI might have said in late 2025, in response to your question:
The absence of systematic value scanning in the 1990s internet commercialisation reflects several converging factors:
1. Methodological Gap: Edward de Bono’s Six Value Medals framework wasn’t published until 2005—a full decade after critical decisions were made. While value engineering existed since the 1940s, it focused primarily on cost-reduction in manufacturing. No accessible, multi-dimensional framework existed for scanning broader societal, environmental, and perceptual values in technology deployment. Marshall McLuhan’s ‘the medium is the message’ (1964) was known but not operationalised as a tool.
2. Cultural Paradigm: The prevailing Silicon Valley ethos was ‘move fast and break things’ (later formalized by Zuckerberg in 2009, but the attitude existed earlier). Innovation velocity was prized over comprehensive impact assessment. The assumption was that technological progress was inherently beneficial—a form of techno-optimism now recognized as dangerously naïve.
3. Economic Incentives: First-mover advantage and network effects created intense pressure for rapid deployment. Pausing for multi-dimensional value analysis would have been seen as competitive suicide. Quarterly earnings pressures and venture capital expectations rewarded speed, not contemplation.
4. Cognitive Limitations: The human brain struggles with second and third-order effects. Without structured frameworks forcing systematic consideration across multiple value dimensions, even brilliant minds like Gates naturally focused on technological feasibility and market opportunity while treating social impacts as externalities to be addressed ‘later.’
6. The ‘It’s Just a Tool’ Fallacy: Technology was widely viewed as value-neutral—a hammer that could build or destroy depending on its user. This prevented serious examination of how the medium itself shapes behaviour, culture, and power structures.
Historical Irony: In 1995, some scattered tools existed—scenario planning, technology assessment, even basic ethical frameworks—but they weren’t synthesized into an accessible, actionable methodology. De Bono’s later contribution was synthesizing existing thinking into a simple, memorable structure that could be practically applied. The tragedy wasn’t the absence of knowledge, but the absence of a framework that made comprehensive value analysis both routine and unavoidable.
Counterfactual: Had value scanning been standard practice, we might have anticipated and mitigated: surveillance capitalism, attention addiction, digital divides, misinformation amplification, and democratic destabilization. The estimated cost of not conducting systematic value scans in the 1990s-2000s is measured in trillions of dollars and immeasurable social damage.2
Pete: Gee Grandpa, your bot uses a lot of big words. Did you understand what it was saying?
Grandpa: Sure. Brainiac just confirmed what we were saying: That people in 1995 didn’t have the tools, or use the tools, to separate what is fun from what is useful. As a result, Brainiac says trillions of dollars was wasted and there was major damage to people and relationships.
Pete: That’s why we get taught to use tools like Six Value Medals in school?
Grandpa: Yes, that’s why.
Pete: But surely the talk between Gates and Letterman was a rare example of people in the past being hopeless?
Grandpa: No Pete. Everywhere you looked there were examples.
Pete: We’re home Grandpa. Good talk.
Grandpa: Yes. Let’s do it again soon.
Phil Bachmann, 25 Dec 2025
Bill Gates Explains the Internet to Dave (1995) | Letterman
See timestamp: 3:09
Taken directly from a chat with Claude Sonnet 4.5