Category: Technology & The Internet

  • I Miss Being Bored

    I Miss Being Bored

    Why having nothing to do might have been surprisingly good for us and why there are benefits of boredom.

    I Used to Be Bored

    I remember being bored. Properly bored. Not “there’s nothing interesting on Netflix” bored, while holding a device containing almost the entirety of human knowledge, entertainment, and argument in one hand. I mean nothing to do bored.

    Waiting for a bus. Sitting in the back of a car on a long journey. Standing in a queue. Waiting for someone who was late. Sitting in a doctor’s waiting room with a three-year-old copy of Country Life. Being on the toilet and discovering there was nothing within reach to read except the back of a shampoo bottle. There were these little gaps in the day when nothing happened and, because nothing happened, you had to find something to do with your own brain.

    I’m beginning to wonder whether we accidentally got rid of something useful.

    We Fixed Boredom

    We have become extraordinarily good at eliminating those gaps. There is almost no situation now in which I need to be alone with my thoughts. Bus late? Phone. Queue? Phone. Someone goes to the toilet in the pub? Phone. Lift taking eleven seconds to arrive? Apparently eleven seconds is now an intolerable amount of time to spend simply existing, so: phone.

    And I don’t say this from a position of superiority. I do it too. My phone is a remarkably effective anti-boredom machine and, as I’ve written before, the internet itself has become remarkably good at deciding what I should look at next. It contains books, music, podcasts, games, messages, photographs, news, and an effectively infinite quantity of people being wrong about things. The slightest hint of an unoccupied moment appears and I can kill it instantly.

    Which sounds like progress. Except I’m increasingly suspicious of what else we’re killing.

    The Strange Things Boredom Made Us Do

    Boredom used to make me do things, although not necessarily useful things. I would take things apart, read things I wouldn’t otherwise have read, stare at maps, look through cupboards, pick up books because they happened to be there, notice odd details about places, and follow trains of thought that didn’t have any obvious destination.

    I even remember being so bored once that I started reading the dictionary. Not looking something up, actually reading it. Probably explains my love of words!

    I’ve already established that I have something of a weakness for learning things I’ll probably never need, although apparently this started earlier than I realised.

    I personally remember a time when, as a bored 12-year-old, I decided to take apart some of my electrical and electronic devices to “see how they worked”. Erm, yeah. Very soon afterwards, “see how they worked” became “see why they no longer work!”

    And I’m not just talking about small things. I’m talking about taking apart an old and dilapidated CRT television. I mean, these things were dangerous, and I really didn’t know enough about electronics to know any better. But I did it anyway.

    My parents were furious, naturally, but afterwards I was enrolled in the school electronics club. At least now I could take things apart AND put them back together again afterwards.

    Would I have developed my interest in electronics if I hadn’t been bored that day? Quite possibly. But the boredom certainly accelerated things.

    Some of those things went nowhere. But some of my longest-running interests probably began precisely because nobody had yet invented an algorithm capable of noticing I was bored and immediately showing me a thirty-second video. Boredom wasn’t necessarily an absence of stimulation. Sometimes it was the point at which I had to start generating my own.

    The Asimov Story I Keep Thinking About

    There is an Isaac Asimov short story called Writing Time, part of his Azazel stories. Azazel is a tiny demon, about two centimetres tall, whom a character called George can summon and persuade to grant favours to people. Like most stories involving wishes, the favour tends to work perfectly while producing a result the recipient hadn’t quite anticipated.

    In Writing Time, the recipient is a writer called Mordecai Sims. Sims is infuriated by how much of his life he wastes waiting. Waiting for buses, waiting for taxis, waiting to be served, waiting for people. All that wasted time when he could be writing. So Azazel fixes it. Sims no longer has to wait. The bus arrives when he does, taxis appear, service is immediate, and all those annoying little delays disappear. He gets exactly what he wanted: more time to write.

    And then he can’t write.

    It turns out that all those periods when Sims thought he was wasting time weren’t really wasted at all. That was his writing time. Not the time when his fingers were putting words onto paper, but the time when his brain was quietly assembling the things that would eventually become words.

    I read that story years ago, and it has stuck with me. Asimov wrote it long before we started carrying devices specifically designed to fill every spare moment, but I can’t help thinking about it now.

    We’ve all summoned Azazel, but now he’s in the disguise of a smartphone.

    When I’m Doing Nothing, My Brain Isn’t

    I’ve noticed this with my own ideas. Ideas don’t generally arrive when I sit down and command myself to have one. Right. It is 10:00. I have allocated the next 45 minutes to creative thought. Begin.

    Nothing.

    The useful thought arrives in the shower, or while I’m walking somewhere, driving, lying in bed, or staring out of a train window. The common factor seems to be that I’m doing something which occupies just enough of my brain to stop me consciously trying to think, while leaving enough of it unoccupied to wander around unsupervised.

    This very article was, in fact, inspired by a recent period of boredom. I awoke at silly o’clock, ridiculously early, and my wife was still asleep. So I got up, went downstairs, and had to occupy myself until I was tired enough to return to bed, but also without disturbing her slumber.

    I picked up my phone, put it down, picked up my Kindle, put it down, perused the contents of the fridge, sat down, picked up my phone again, returned to the fridge, decided I still wasn’t hungry, and put my phone away again. I remember thinking, “God, I’m bored and I’ve nothing to do!”

    Which, in turn, reminded me of the times when I was bored and FOUND things to do.

    The seed of this essay took root that morning.

    That wandering is where quite a lot seems to happen. One thought bumps into another. Something I read three weeks ago collides with something somebody said yesterday. An old memory wanders past. Two completely unrelated ideas turn out to have something in common, or at least enough in common for me to spend the next twenty minutes wondering whether they do.

    Somewhere in that mess, occasionally, there’s an idea.

    The problem is that I’ve become very good at preventing this state from occurring. If there is a spare minute, I fill it. If I’m walking, I can listen to a podcast. If I’m on a train, I can read the news. If I’m waiting for someone, there will be something on my phone demanding attention. None of this feels particularly significant while I’m doing it. I’m just filling time.

    But that was Sims’s mistake too. He thought he was just filling time.

    Consuming Instead of Generating

    If I take my phone out while waiting for a bus, my brain doesn’t have to produce anything because something else immediately supplies the thoughts. A news story, a message, a video, someone’s opinion, someone else’s response to someone’s opinion, another video, another message. There is always something available and, increasingly, something specifically selected because a machine has calculated that I am quite likely to look at it.

    None of these things is inherently bad. I like reading things. I like hearing what other people think. I have spent an unreasonable amount of my life actively seeking out information and have no intention of stopping now. The problem is simply that there is never a gap.

    There is a difference between having nothing to think about and having nothing given to you to think about. I’m not sure we make that distinction often enough.

    Boredom Wasn’t Always Good

    I don’t want to get too nostalgic about this because boredom was, by definition, boring. Long car journeys were tedious. Waiting forty minutes for a bus in the rain wasn’t secretly a mindfulness retreat. Sitting in a waiting room didn’t automatically turn you into Isaac Newton.

    Smartphones have made those moments immeasurably better in lots of ways. I can read a book, learn something, talk to somebody on the other side of the world, listen to music, or find out where the bloody bus actually is. I’m not suggesting we throw our phones into the sea and return to an idyllic age of staring at wallpaper.

    I just wonder whether eliminating every moment of boredom was an accidental overcorrection.

    I Might Need to Be More Bored

    I’m not planning a digital detox. I know myself well enough to suspect that any grand declaration about dramatically changing my relationship with technology would last until roughly lunchtime.

    I could leave the phone in my pocket occasionally, though. Walk somewhere without listening to anything. Wait for the kettle without checking something. Sit on a train and look out of the window. Allow a queue simply to be a queue. Give my brain a few minutes in which nobody else gets to decide what enters it.

    I suspect I’ll find this surprisingly difficult, because boredom now feels uncomfortable remarkably quickly. There is an almost automatic movement towards the pocket as soon as nothing is happening. I don’t even consciously decide I’m bored. My hand apparently notices before I do.

    It might be interesting to resist it occasionally, if only to find out what my brain does when I stop giving it things to do.

    Nothing Happened

    There used to be parts of the day when nothing happened. At the time, I thought those moments were empty. Looking back, I’m not so sure they were. They were when things I’d read resurfaced, when ridiculous questions occurred to me, when I’d notice something outside the window, or when one idea had enough time to bump accidentally into another one.

    Mordecai Sims thought waiting was stealing time from his writing, then discovered it was part of the writing. I think that’s why the story has stayed with me for so long. I recognised the irritation with waiting when I first read it. These days, I recognise Sims’s mistake as well.

    I wonder how much time I’ve managed to save myself in the last fifteen or so smartphone enabled years.

    And what I used to do with it.

  • The Internet Was More Interesting Before Everything Became a Platform

    The Internet Was More Interesting Before Everything Became a Platform

    I remember the early days of the internet.

    I remember the terrible websites with flashing green text on red backgrounds, animated GIFs in abundance, JavaScript visitor counters at the bottom of the page, MIDI tunes that started playing when you accessed the site, and entire websites dedicated to someone’s hamster.

    (Guilty as charged, m’lud.)

    And you know what?

    It was GLORIOUS!

    Don’t worry, this isn’t an “old man yells at cloud infrastructure” post.

    The old internet wasn’t perfect, it wasn’t ideal, and it certainly wasn’t always safe.

    However, what it was, was quintessentially human.

    THE INTERNET USED TO FEEL GEOGRAPHICAL

    The actual internet was a highway.

    It wasn’t the destination, but the means of travelling between weird towns, villages, and hamlets, each of which specialised in… well… something.

    Much like the towns in the Mad Max universe, where you have Bartertown, Gastown, Bullet Farm, and Tomorrow-morrow Land, the internet had Hamsterville, Catopolis, Conspiracy Creek, and Anosmia Acres.

    I was a serial website owner.

    I ran an anosmia support forum which, somehow, got me on television for my fifteen minutes of fame.

    Its design may well have made visitors’ eyes bleed, granted, but as it was a support group for people with no sense of smell, it was perhaps appropriate that my design stank. (See featured image from The Internet archive above!)

    I ran a joke conspiracy PHPBB forum called the UK Conspiracy Nexus, where every “member” was another character I had created and all the posts, arguments, and discussions were written by me.

    I deliberately invented ridiculous conspiracy theories, such as the connection between Toblerone and Illuminati mind-control dentistry.

    It’s the triangular shape, you see.

    It was all good fun… until I noticed my posts being shared non-ironically on more “serious” conspiracy sites.

    Dear God.

    What did I spawn?

    I also created a maths and general nerdiness site dedicated to things geeks might find “sexy” (that is to say, cool).

    In hindsight, calling it Sex4Geeks was perhaps inviting exactly the sort of dubious attention it received.

    That site was short-lived.

    It did, however, end up becoming a joke in my best man’s wedding speech.

    Pete?

    Naive much?

    Heck yes.

    But the internet was full of things like this.

    BBSs.

    Usenet.

    Forums.

    IRC.

    Personal webpages written in hand-crafted HTML.

    It wasn’t one place.

    It was thousands of interconnected little places, each with their own personality, culture, and rules.

    You didn’t live there.

    You visited.

    You travelled between them.

    You discovered them.

    You got lost.

    Wonderfully so.

    NOBODY NEEDED A BUSINESS PLAN

    The majority of the internet back then existed for one reason.

    Someone cared about something enough to make something about it.

    Not because they wanted followers, subscribers, sponsorships, or monetisation.

    It was because somebody cared enough about model railways, their collection of Tamagotchis, or the history of the China tea trade.

    They had passion and they wanted to share it.

    There was a website for every niche interest and most of them looked as though they had been assembled during an environmental catastrophe and built entirely in the dark.

    THE GREAT CONSOLIDATION

    Of course, initially seen as a fad, the internet proved it wasn’t going anywhere.

    People realised its potential.

    You could now order a pizza online.

    Such a technological breakthrough is something we take entirely for granted nowadays.

    Companies with ideas and dollar signs in their eyes jumped in.

    Slowly, giants emerged from the chaos.

    Where once there had been millions of random amateur websites, suddenly you had Facebook, Twitter, and MySpace.

    Then YouTube, Reddit, Instagram, and TikTok.

    These places acted like magnets, drawing in the weird from the satellite towns, villages, and hamlets, and housing them instead in vast digital metropolises.

    THE ALGORITHMS THAT ATE DISCOVERY

    Finding things used to be an adventure.

    There were a few basic search engines, reciprocal links, blogrolls, and web rings connecting like-minded sites together.

    You would wander around them feeling as though you were forging your own path through a strange and fascinating landscape.

    You discovered what people made.

    You sought out things that interested you.

    Now we no longer discover what people made.

    We discover what the algorithms, working on behalf of the platforms, want us to see.

    Want your site to appear first in search results?

    Just pay lots of money.

    It’s even worse now with AI-generated summaries appearing above the sites that originally created the content.

    In the pursuit of making things easier, we lost something human along the way.

    THE DEATH OF WEIRD

    The old internet was eccentric, amateur, and personal.

    Websites now look polished, optimised, branded, and laden with “smart” features.

    The old internet may have looked like a terrible hodgepodge mess, but it looked like somebody.

    It was human.

    Much of today’s internet looks like a corporation.

    A polished, focus-grouped, metrics-driven corporation.

    A CONFESSION

    Nostalgia is a terrible historian.

    As Baz Luhrmann famously said:

    “Nostalgia is a way of fishing the past from the disposal, wiping it off, painting over the ugly parts, and recycling it for more than it’s worth.”

    The old internet had its fair share of problems.

    Awful design.

    Scams.

    Misinformation.

    Toxicity.

    Every village conspiracy nut could find a like-minded community and build little echo chambers where the worst ideas fed off one another and grew.

    I’m not arguing that the old internet was perfect.

    I’m simply saying it was more human.

    We still have scams, misinformation, and toxicity.

    Now it’s just run by a handful of billionaire puppet masters.

    WE LOST MORE THAN WEBSITES

    The real loss wasn’t the technology.

    That grew.

    Who would have imagined we’d one day carry around access to the sum total of human knowledge on a small device that fits in the palm of our hand?

    We lost individuality.

    Experimentation.

    Curiosity.

    And yes, weirdness.

    The internet became a consumer space for selling content rather than a creative space for sharing ideas.

    IN CONCLUSION

    Perhaps that’s one of the reasons I reincarnated this blog.

    Not because I expect thousands of readers.

    Not because I have a monetisation strategy.

    Not because some marketing consultant told me to build my personal brand.

    No.

    Because I miss people having little corners of the internet that belonged to them.

    Places where we could be knowledgeable, enthusiastic, eccentric, or even completely wrong about something.

    Places that felt human.

    The internet may have become a platform, but there’s still room for the occasional strange little website.

    Or should I say, in my case, the occasional strange Little website.

    This is my own weird nest on the web.

  • This Post Was Written by AI

    This Post Was Written by AI

    Wait, just wait…

    By “AI” I do, of course, mean an Actual Idiot.

    Namely, myself.

    Obvious clickbait title, eh? Thought that’d catch you!

    Let me elucidate.

    Actually, no. Let me explain.

    Words like elucidate are what got me into this mess in the first place.

    Despite my best intentions, I apparently sound artificial.

    I’m not offended by this. In fact, I’m quite fascinated by it.

    I’d noticed that my writing scored embarrassingly highly on various online “Was this written by AI?” tests. At first I couldn’t understand why.

    Well, that’s not entirely true.

    I had a fair-to-middling idea why.

    It’s my writing style.

    It’s too… measured.

    Too formal.

    Too structured.

    Too fond of words like fair-to-middling.

    I grew up with a fairly severe stammer. (See my previous post about that.)

    As a result, I spoke slowly and used small, simple words. But just because I spoke slowly, chose my words carefully and delivered them in a flat, measured manner didn’t mean I lacked a vocabulary.

    Far from it.

    I was a voracious reader.

    If there was something to read, I read it.

    Books.

    Magazines.

    Newspapers.

    Encyclopedias.

    Dictionaries.

    The backs of cereal boxes.

    If it contained words, it was fair game.

    My nose was almost permanently buried in a book.

    I developed a particular fondness for older literature. Poe, Lovecraft, Shelley, Dumas, Dickens and many others filled my head with strange vocabulary and elaborate sentence structures.

    Reading dictionaries and encyclopedias gave me a phenomenal vocabulary. Positively Brobdingnagian, one might say.

    (I use that word rather a lot. It’s one of my favourites. Perhaps I should stick with “Big” going forward?)

    My irritating near-photographic memory meant I accumulated a vast store of facts and trivia. So much so that my parents’ nickname for me was “The Professor”, because I was always explaining how something worked, who invented it or why some obscure historical event mattered.

    In public, however, I was deeply introverted.

    I embodied the old instruction:

    “Speak when you’re spoken to.”

    And even then, I generally replied with the minimum amount necessary to remain polite.

    I developed an inferiority complex that never entirely left me.

    I’m fifty-six years old.

    I work for a large multinational IT company and support some fairly significant clients.

    Think military.

    Think government.

    Think organisations with enough bureaucracy to make Kafka and Orwell ask everyone to calm down.

    And yet, in my head, I’m still just a code monkey.

    Most people my age have become team leaders, managers, senior managers or directors.

    Not me.

    I never particularly wanted promotion.

    I lack anything resembling a healthy ego.

    I’m just as happy writing a flagship application or major piece of software as I am testing somebody else’s code or filling out administrative paperwork.

    I’ve never been good at putting myself forward.

    I rarely assume excellence.

    I rarely even claim competence.

    Which is ridiculous, of course.

    I know I’m good at what I do.

    I find that difficult to say.

    In fact, if I’m being painfully honest, I’m very good at what I do.

    There. I’ve said it.

    Please don’t make me do it again.

    At university, the expected trajectory by my peers was a PhD.

    Instead I dropped out at the end of my BSc before my Masters.

    Why?

    Because I couldn’t picture myself as a doctor.

    (Albeit a Doctor of Mathematics rather than the Gallifreyan variety.)

    Me?

    A stammering working-class lad from a family of labourers and cooks?

    I knew my place.

    And in my mind, it certainly wasn’t “Doctor”.

    The imposter syndrome hit me big time.

    Looking back, that says rather more about me than it does about academia.

    My lack of ambition and reluctance to promote myself influences my writing too.

    I minimise myself.

    I soften statements.

    I hedge.

    I qualify.

    I write passively because active statements feel uncomfortably close to boasting.

    Even writing this article feels slightly unnatural.

    I’ve deliberately filled it with “I am”, “I did” and “I think” statements.

    Normally I’d be restructuring sentences, removing myself from them and quietly disappearing into the background.

    The urge to do so is considerable.

    I’m resisting it for the sake of the experiment.

    So perhaps that’s why AI detectors think I’m an AI.

    Not because I write like a machine.

    But because I write in a style that is structured, measured and oddly formal.

    The irony is that I was probably sounding like this long before anybody took artificial intelligence seriously.

    For the record, I’m not particularly an AI enthusiast.

    My mathematical background means I understand the statistical prediction and pattern-matching taking place beneath the surface.

    My IT background helps me understand the engineering behind it.

    I find the technology fascinating.

    I find many of its applications considerably less so.

    Personally, I wouldn’t trust a large language model to write a postcard, let alone a blog or novel.

    Speaking of novels, I have one.

    Actually, I have ten.

    Although, technically, they’re ten versions of the same novel.

    None of them will ever be published because none of them will ever be finished.

    I keep discovering ways to improve them.

    Or ruin them.

    The distinction becomes increasingly blurry after version seven.

    As for AI-generated writing?

    I find much of it technically impressive but emotionally hollow.

    Competent.

    Coherent.

    Soulless.

    Perhaps that’s why I’m amused when people accuse me of writing like an AI.

    I’ve spent a lifetime trying to figure out how to sound more human.

    It would be deeply ironic if I finally succeeded just as the machines started sounding like me.

    BEEP BOOP.. END OF POST ERROR : REFILL #C0FFEE.

  • Small Talk Is TCP Handshaking for Humans

    Small Talk Is TCP Handshaking for Humans

    I have long suspected that small talk is not really about talking at all. I mean, obviously it’s about talking, but it’s not about imparting information or saying anything of actual interest.

    It is, I think, more accurately described as a kind of human protocol negotiation. A lightweight exchange whose purpose is not to transfer meaningful information, but to confirm that both parties are present, functional, and broadly willing to proceed without causing a scene.

    In computer networking terms, it feels rather a lot like a TCP three-way handshake.

    That may sound like an over-complicated way to describe “Morning” and “Busy day?” but I maintain it is exactly the right amount of over-complication.

    The analogy works on several levels at once.

    TCP handshakeHuman small talk
    SYN“Morning.”
    SYN/ACK“Morning. Busy today?”
    ACK“Yeah, not too bad.”
    Connection establishedActual conversation may now proceed

    The actual content is almost irrelevant.

    The handshake exists to establish availability, willingness, compatibility, safety, and communication parameters. Small talk does much the same thing. It is not there to win prizes for insight. It is there to say, in the safest possible way:

    I see you. You seem non-threatening. We may now continue.

    That is quite a useful invention, really.

    It is also, for some of us, faintly bewildering.

    Because my brain has a tendency to assume conversations should do one of three things:

    1. exchange useful information.
    2. explore an interesting idea.
    3. contain at least one genuinely terrible joke.

    Small talk often does none of these things.

    Which is not to say it is pointless. Far from it. It is just that its point is not always the same as the point my brain is looking for.

    That can make it feel a bit like packet loss.

    Some of the signal gets through. Some of it doesn’t. A few words arrive intact, but the important meaning seems to have been dropped somewhere between the intention and the execution.

    The weather is discussed.

    The weekend is mentioned.

    A vague inquiry is made about one’s wellbeing.

    And yet the actual data being transmitted is something like:

    I am safe. You are safe. We may continue this interaction without immediate risk of social collapse.

    That is a rather elegant system when it works properly.

    It is also a system with some spectacular failure modes.

    Failed Handshake

    Sometimes the handshake fails immediately.

    You offer the smallest possible opening.

    “Morning.”

    And the other person replies:

    “Is it? I mean the way the world’s going it might as well be the end of days. The planet’s going to hell in a hand basket and we’re all doomed.”

    At that point the connection has not merely failed. It has been forcibly terminated, the cable has been unplugged, and the router has probably started smoking.

    Connection reset by peer.

    Timeout

    Then there is the timeout problem.

    Someone asks:

    “How are you?”

    and the answer requires a socially acceptable level of detail, emotional calibration, and acceptable honesty.

    My brain, meanwhile, is attempting to calculate whether “fine” is technically accurate, whether “surviving” sounds too bleak, whether “not bad” is too British to be believable (see below!), and whether any response longer than six words will prompt a follow-up I do not have the energy to sustain.

    By the time I have finished this internal analysis, the moment has passed.

    Connection timed out.

    Packet Flooding

    Some people go in the opposite direction.

    You offer:

    “Morning.”

    And they respond:

    “Morning! Did you see that interesting show on TV last night about medieval bridge taxation? Made me think about my current view on plug in solar panels, my advancing years, my neighbour’s cat, and the collapse of the world’s democratic politics systems..”

    That is not a handshake.

    That is packet flooding.

    ADHD can do this rather beautifully. One tiny conversational spark and suddenly a completely unrelated archive of information is bursting out of the firewall at speed.

    Which, to be fair, can sometimes be entertaining.

    It just isn’t small talk anymore. There’s nothing small about it.

    Encrypted Traffic

    Us British people add another layer of complexity, because many of us speak in what is essentially encrypted traffic.

    “Not bad.”

    Not only is this British understatement at its very best, this can mean:

    • life is actually pretty good actually!
    • life is acceptable.
    • life is mildly annoying and could be better.
    • life is a bit of a struggle but let us not make a fuss.
    • a bottomless abyss has opened beneath me, but the kettle still works, so we are continuing cup in hand and trying our best to avoid the hordes of demons spewing forth from this hellish portal.

    That is an extraordinary amount of information to pack into two words.

    It is also one of the reasons small talk can be so difficult to decode. The literal words are only half the message. Tone, timing, context, and cultural expectation carry the rest.

    The protocol is not just words.

    It is inference, and unfortunately, not something I’m particularly good at.

    UDP People

    Then there are people who skip the handshake entirely.

    They are the conversational equivalent of UDP.

    “HELLO STRANGER HERE IS MY INFODUMP OF TRAUMAS, PROBLEMS AND CONCERNS!!.”

    No connection setup.

    No reliability guarantees.

    No orderly exchange of packets.

    Just pure, immediate data transfer with absolutely no regard for conventional protocol.

    This can be alarming, but occasionally refreshing.

    Stateful Firewall

    Autistic masking probably deserves its own category here.

    It can feel a bit like trying to pass through a stateful firewall while manually monitoring every tiny aspect of your own behaviour.

    Eye contact? Adjusted.

    Tone? Checked.

    Facial expression? Calibrated.

    Volume? Sensible.

    Response time? Natural enough to avoid suspicion.

    Internal panic? Well hidden, hopefully.

    The result is not always elegant, but it is functional.. barely.

    NAT Traversal

    Trying to socialise at parties is also very much like NAT traversal.

    You know there is a connection somewhere in there.

    You know the other person is a perfectly ordinary human being.

    You know you are supposed to be able to reach them.

    And yet the routing is somehow wrong, the ports are blocked, and everyone is standing in awkward little clusters pretending not to notice that all communication has become mysteriously more difficult than it should be.

    And when you do get into a cluster, the least said about the actual handshaking the better. (Especially Dave with the sweaty palms!)

    So Why Does Small Talk Exist At All?

    Because it is doing a job.

    A very human job.

    Small talk reduces uncertainty.

    It establishes connection.

    It signals intent.

    It gives both parties a safe place to stand before deciding whether to move into more meaningful territory.

    It is a social compatibility check.

    A “yes, we can talk” exchange.

    A way of saying:

    I am friendly.

    You are safe.

    We may proceed.

    That is actually rather beautiful, when you think about it.

    Even if the words themselves are often about absolutely nothing of consequence.

    Why It Feels So Hard

    For many neurodivergent people, the difficulty is not that small talk is meaningless.

    The difficulty is that its meaning is often indirect.

    It is a social ritual disguised as conversation.

    A lot of the work is happening in the subtext.

    And subtext is a slippery little creature.

    If your brain prefers clarity, precision, and explicit information, small talk can feel like being asked to decode a message written in invisible ink.

    Which, admittedly, is not very helpful.

    The Human Side of the Protocol

    And yet…

    And yet, for all my cantankerous grumbling, I do think there is something lovely about it.

    We are, after all, social animals operating a remarkably fragile but surprisingly effective communication system. Small talk is one of the tiny pieces of software that keeps the whole thing from crashing immediately.

    It is how we ease into each other’s company.

    It is how we test for warmth.

    It is how we say, with minimal risk and maximum politeness, that we are willing to share a moment. You’re lowering your shields and allowing the Klingons to beam aboard.

    That is not nothing.

    In Closing

    So yes, perhaps small talk is the human equivalent of a TCP handshake.

    A few neat little exchanges whose real purpose is not the content, but the connection.

    Sometimes it works perfectly.

    Sometimes it fails immediately.

    Sometimes the packets arrive out of order.

    Sometimes somebody starts talking about medieval bridge taxation or python programming before the connection is even established.

    But when it does work, it lets two strangers become, very briefly, not strangers at all.

    And really, that is quite a clever trick for a few essentially meaningless words and a friendly look over a cup of coffee.