Momentum, that is. SciFact hasn’t acquired a large gentleman called Maurice. Although there might be a future animated character in that :).

The week-two board pack is ready (scroll down for the slides). The board remains me, my wife and our one-year-old, and there’s more to be excited about: more views, more comments and signs that some people are coming back.

There’s also a slightly less glamorous development, more arduous, and distinctive un-Andy…

I need a routine.

Another week, another little hit

Last week, I wrote about the excitement of seeing people respond to something I’d made. That hasn’t worn off.

The week-two report records 67,459 views across YouTube, Instagram and TikTok, compared with 51,026 in the first report. Up about 32%.

YouTube contributed 36,255 views and Instagram 29,377. Both grew. TikTok went the other way, falling from 4,990 to 1,827.

So the overall direction is pleasing, but there are things to understand underneath it.

The usual small print applies: these are views, rather than different people, and they include views on older videos. YouTube’s week runs from 5–11 October; Instagram and TikTok cover the seven days between the reporting snapshots on 7 and 14 October. Even the first reporting windows weren’t identical.

I’m happy to enjoy the progress without pretending this is a perfectly controlled experiment.

YouTube also added another 95 net subscribers, taking the total since launch to 202. Comments rose from 77 in the first report to 114.

Those are still small numbers in the grand scheme of the internet.

They don’t feel small when it’s your thing.

Week two is already week three

There’s a slightly strange time problem with these updates.

You launch. A week passes. You review it.

By the time you’re writing about week one, you’re already doing week two. This week-two review arrives while I’m working through week three, and most of my headspace is somewhere in weeks four, five and six.

The report is looking backwards. The work is happening now. The ideas are already wandering off ahead.

Which is why routine has become a big thing for me this week.

The launch had that lovely burst of energy: an idea, a creative direction, enough tools to make it happen and an urge to get it out into the world.

Keeping it going means repeatedly finding questions worth answering, working out an inviting way into them, getting the videos produced and making sure they reach people.

The week-two pack records 74 new YouTube Shorts, with mostly the same videos shared to Instagram and TikTok. That’s a lot of questions, explanations and little creative decisions.

I still find it exciting. I also need a rhythm that helps me keep doing it.

Brainstorming needs a place in the week. Production needs a place. So does checking what happened after I pressed publish.

Turns out the next episode doesn’t make itself merely because I’m enthusiastic about having a channel.

The comments need a routine too

Some of the nicest things in this week’s pack are things people have said.

One viewer wrote:

“This channel oddly comforts me”

I love that.

SciFact started with the idea of soft, playful ways into a subject. Someone finding comfort in it feels connected to what I wanted to make.

Another comment said:

“finally a video that really explains it the way i can understand”

That viewer appeared on two different Shorts. There’s something particularly pleasing about recognising that someone has encountered the channel more than once and is still joining in.

Responding to those people is part of the work.

There are questions to answer, suggestions to think about and conversations to continue. I want to give that attention rather than let it become something I squeeze in whenever I happen to notice a notification.

The viewer-question videos are getting watched too. The report shows more than a thousand views each for the vinyl and CD explainers, with the car air conditioning video close behind.

That gives me a reason to keep listening. People are helping me find questions I can do something with.

The £15 experiment that ate two evenings

There have been frustrations as well.

I tried to boost a video on Instagram. Just £15, to see what would happen.

A small experiment. Modest ambitions.

It ended up taking two evenings, and I still didn’t manage to get it working.

So I can’t tell you whether £15 of promotion helped SciFact reach a useful audience. I can tell you that trying to arrange it used up rather more of my time than I intended.

That belongs in this update too.

It’s easy to show a chart going upwards. The chart doesn’t show the evenings spent trying to get a fairly ordinary task to work.

But those evenings are part of building the thing. They affect your energy, your time and what else you manage to do.

I wanted a little advertising experiment. I got an administrative subplot.

My AI agents would like to change the creative

The board pack comes with recommendations from my AI agents.

This week, they want changes to the openings: rework the first second on five Shorts and try an opening specifically for TikTok.

There’s evidence behind the suggestion. The report puts engaged views for the new YouTube Shorts at 35.5%, compared with 37.1% for the previous batch. On one TikTok video about retro games, 65% of viewers had gone within a second.

That’s worth paying attention to.

I’m still not sure about changing the creative.

I’ve come to like it the way I like it. The pace, the feeling, the way somebody is invited into the explanation. There’s a character to these videos that I’m attached to.

That isn’t me declaring that everything is perfect. Or that I’ll never change it.

But not yet.

I want to understand what I’d be changing and why. Would a different opening help someone find their way into the video? Would it preserve the feeling I’m trying to create? Would I be making a better introduction, or simply a louder one?

Those are creative decisions. The numbers can help me ask the questions, but I still have to make the decisions.

Last week I wrote about being led by insight. I haven’t abandoned that because another spreadsheet has arrived.

And one recommendation didn’t happen

There’s an awkward little line in the review of last week’s proposed changes.

The recommendation was to reduce TikTok publishing to one to three posts a day.

The week-two report records 72 posts.

The verdict in the deck is wonderfully direct: “Not done.”

Fair enough.

I can’t claim to have tested a lower posting frequency when I haven’t actually done it. And I can’t use this week’s results to decide whether that proposed change worked.

It’s another reason the routine matters. A recommendation in a presentation needs to become an actual decision, then an actual part of the work, before there’s anything meaningful to review.

Some recommendations I’m questioning. This one simply wasn’t carried through.

Both belong in the account.

Still excited. Getting into a rhythm.

Week two has given me plenty to feel good about.

More people are watching. Some are returning. Questions are turning into videos. Someone finds the channel comforting, which is a lovely thing to hear about something you’ve made.

It’s also brought the work of keeping it going into sharper focus: finding a routine, making time for the community, dealing with frustrating bits of administration and deciding which advice deserves action.

This is exactly the point of #buildingInPublic for me.

I want to show the progress and the work around it. The decisions I’m making, the things I haven’t managed and the recommendations I’m still arguing with.

The big mo keeps rolling.

Now I’m working out a routine that can keep up with it.

You can watch SciFact on YouTube. And if you’re building something yourself—or have thoughts on any of this—come and find me on LinkedIn or X. I’d love to compare notes.

I’ll not bore you with the log files and gateway restarts. It’s the interesting bit that came next.

The detour that paid for itself

While I was geetting DeepSeek v4 Flash (my favourite model at the moment in Hermes), my agent noticed something else was broken: vision. The tool for reading images was pointing at a model that no longer existed on the provider. It returned a 404, quietly, the way these things do.

I asked it to find a working one. It went digging through the provider’s live catalogue and came back with something better than a fix — it came back with a free one.

Nous carries genuinely free models. Not free-trial, not free-until-the-end-of-the-month. Free. And the free vision model worked perfectly – but whist the agent was working through the logic of finding the free model, it was pinging around reading price lists of different provider APIs.

That’s when the actual idea arrived.

“Am I on the best model?”

So by default I have some credits in Nous and Merge. I also think that Nous can route through Open Router, but whenever I’ve tried to do that before, I’ve got frustated and gave up.

So Hermes is logicizing as a good agent would and then I stopped it:

That’s interesting that Merge and Nous expose their model prices. Can you set up a skill and a cron job to email me the list of the cheapest once a week so that I know I’m on the best model? Or maybe actually that list would be too long — I don’t need any Anthropic or OpenAI or Gemini info because I have monthly plans for all of those. It’s really where is the cheapest DeepSeek model and if there are any new models at a good price to try.

That last bit is the whole thing. A list of every cheap model is noise. A list that says here’s the cheapest DeepSeek, and here’s anything new worth trying is a decision. And the filter that makes it useful is the one only I could give: I already pay flat monthly rates for Anthropic, OpenAI and Google. Their prices are irrelevant to me. Showing them would be filling the page to look thorough.

So: two providers I actually route to (Nous and Merge), cheapest-first, with the flat-rate vendors stripped out.

Then we hit the small problem

I’d asked for it by email. My agent went looking and found that email isn’t set up on this machine at all. No mail client, no SMTP credentials, nothing. A perfectly good idea, blocked on a delivery route.

What I liked about this bit is that it didn’t just stop and ask me to configure email. It went and found what was already wired up. Which helped me to also suggest an option:

Me: Slack?

Slack. Already connected, already delivering, already where I’d see it. The channel was the easy part; I’d just assumed email because that’s what I’d asked for.

The build

What came out the other side was three things:

• A script that queries both providers’ live catalogues, reads the real per-token prices, and prints a short digest. No AI in the loop for the data — it’s deterministic, so the numbers can’t drift or get “summarised”.
• A skill — a note to my agent about how the whole thing works, so next time it doesn’t have to rediscover that one provider quotes prices per token and the other per million, or that a listed route sometimes 404s on the real call.
• A cron job — every Monday at 8am, straight into my Slack.

It also keeps a list of every model it has seen, so from week two onward it can tell me what’s new rather than re-reading the whole catalogue at me. Small detail. But that’s the difference between a report and a notification.

The first thing the radar found was me

Here’s the bit I didn’t expect. I asked it to check the cheapest DeepSeek route, assuming the answer would be “the one you’re on, well done”. It wasn’t.

Same model, different route, and the price gap was silly:

Merge’s input price is higher. Its output price is eleven times cheaper. Which one wins depends entirely on how you actually use the thing — and I didn’t know that, so I asked it not to guess either. It went to my real logged usage and did the maths on my own numbers, not a hypothetical.

Merge came out about 57% cheaper for me. The reason is that my usage is dominated by cached reads — the same long context being re-read over and over — and Merge charges a fraction of what Nous does for exactly that.

But a price on a page isn’t proof. So it made one live call, and the listed price matched the actual bill to the cent. Then it switched me over and checked the gateway came back up on the right model.

Me: Yes, make the switch please

Done. Same model, roughly half the price, verified rather than assumed.

What I’m taking from it

Silent failures are the expensive ones. A model that stops answering tells you immediately. A model that quietly downgrades itself tells you nothing, and you carry on for weeks wondering why the thinking feels thinner. If a setting matters, test that it works, not that it’s set.

The useful version of “monitor prices” is “answer one question”. I didn’t want a market report. I wanted to know whether I was on the cheapest DeepSeek and whether anything good had shown up. Everything else was decoration.

Give the constraint, not the spec. “I have flat plans for those three” is the most valuable sentence I gave it. That’s the bit a generic tool can’t know and can’t guess — and it’s the bit that turned a long list into a decision.

The best fixes are the ones that keep working. I could have asked “what’s the cheapest model?” and got an answer that was stale within a fortnight. Instead I’ve got a robot that checks every Monday, and a note to my agent that means it won’t have to learn the same lessons twice. The answer is the small part. The habit is the useful part.

And it did, in the end, find itself a cheaper model. I just had to ask.

If you think this would be useful, you can have it. The script and the skill are on GitHub — point them at your own providers and they’ll do the same job for you:
github.com/andygauk/model-price-radar

The board currently consists of me, my wife and our one-year-old. Governance is still developing.

But there’s something quite exciting about having enough to put in a board pack. A week ago, this was a creative idea I wanted to get out of my head. Now there are videos, viewers, comments and actual things to discuss.

When I launched SciFact, the fanfare was confined to my own office. This week, people outside it started responding.

That feels super good!!!

Oh. People are watching.

The first report adds up to just over 51,000 views across YouTube, Instagram and TikTok, with YouTube accounting for nearly 30,000.

Those are views, rather than 51,000 different people. The reporting windows also differ slightly: YouTube’s figures run to 4 October, while Instagram and TikTok run to the morning of 7 October. This is a snapshot of the launch, with all the untidiness that comes with it.

Still. Fifty-one thousand is kind of wild for me. And I know it’s not 10 million subscribers, and Saturday Night Live isn’t going to ask me to host any time soon… but, week one!!

I know enough about marketing to understand why I shouldn’t get carried away with a view count. I’m also the person who made the thing, and I’m quite happy to admit that seeing the numbers move gives me a little dopamine hit.

We should probably make a video about what dopamine is. For research purposes, obviously.

Then somebody comments, and there’s another little hit. Someone has watched, thought about it and bothered to respond.

The best part is when they ask a question.

The audience has started joining in

SciFact is built around simple questions about everyday things. A soft, playful way into a subject you might otherwise never think about.

Why does popcorn pop? How do submarines sink and float? Why are bubbles round?

Now some of those questions are coming from viewers.

Five have already become videos, including why oily puddles have rainbows, how vinyl records hold music and how car air conditioning works.

I love that.

Someone’s curiosity has become the starting point for something I make. The next video can answer them, while also opening the subject up to everyone else.

There’s a lovely connection there with the original idea for SciFact: make a concept feel approachable enough that someone wants to come closer. A friendly face on a modem was one way into that. A question in the comments is another.

That’s the kind of response I want to pay attention to.

The slide I’m arguing with

The board pack ends with recommendations from my “team”, Al and Inca. Or is that Sal and Inca now, after Donald Trump’s meeting at the White House? Haha.

Some videos attracted more attention than others. Hair and body myths did well. Kitchen questions and everyday physics showed promise. Sharing was low. TikTok went very quiet.

There are sensible suggestions attached to those observations: make more of certain topics, sharpen the opening, give people a reason to share, reduce the posting frequency.

And this is where I start arguing with my own presentation. But for me, this is what #buildInPublic should be about: showing the workings, the successes and the failures.

A week of results can suggest a direction worth exploring. It can’t tell me everything about why something worked, or whether doing more of it will make SciFact better.

A video about hats and baldness travelled across all three platforms. Does that mean people particularly like body myths? Was the question unusually tempting? Did the opening work well? Did it simply get a better opportunity to be seen?

I don’t know!

“Make more of these” is a reasonable experiment. It becomes much less reasonable if I quietly turn it into “this is what the channel must now be”.

This thing is my (other) baby. I’m excited about what it could become, and I care about the character it’s developing. I’m not going to change that character every time a chart points somewhere interesting.

I’ve heard “led by the data” before

In previous roles, I worked with people who would say things like “we’ll be led by the data” or “the strategy is based on data”.

It frustrated me.

Data matters. Of course it does. But strategy and creative direction need to be led by insight.

You can use data to support loads of different arguments. Choose a different metric, a different time period or a different comparison, and you can find another apparently convincing story.

The hard work is understanding what the evidence means.

For example, SciFact was conceived for teens. In this first YouTube report, 21.5% of measured viewers were aged 13–17, while 51.6% were aged 25–44. Those age figures cover signed-in viewers, so they’re only part of the picture.

Are adults enjoying the same accessible explanations I intended for teenagers? Is the playful approach working across ages? Or is there something else drawing them in?

Those questions matter because a larger adult audience doesn’t automatically require a different creative direction. It might mean the original direction has wider appeal than I expected.

That’s an interpretation to investigate. Changing the brief is a decision.

Common wisdom needs a question mark too

There’s plenty of advice about growing a channel. Sharper hooks. More calls to action. Fewer posts. More posts. Follow the format that’s working.

Some of it will be useful.

But if we all follow the same advice, we’ll all end up making versions of the same thing.

Take sharing. The report records 61 shares across those 51,026 views and recommends ending every Short with a reason to send it to someone.

I can see the logic. I also want to know what that would feel like in the video.

Would it add something playful and relevant? Would someone think, “My mate would enjoy that”? Or would I be bolting a marketing instruction onto a satisfying little explanation?

The execution matters. So does knowing when to leave it alone.

TikTok’s sudden drop deserves investigation too. Trying a slower publishing pace seems sensible. But the figures alone don’t prove that posting frequently caused the drop.

I want to test an idea and understand what happens. That gives me something useful to learn from.

This is what building in public looks like

There were 215 posts across the three platforms in the launch report. That includes the same videos appearing in different places; it isn’t 215 separate films.

There were corrections too: three YouTube Shorts were replaced with corrected uploads. On one, a viewer commented that the plane was moonwalking because the plane sprite was going backwards in the video. Red-faced moment. But we fixed it, told the viewer and thanked them for letting us know.

I want those details in the open alongside the pleasing numbers. And I want to share the decisions while I’m still thinking about them, when there’s something for other people to question or contribute to.

So week two will involve making, listening and trying a few things. Some suggestions from the board pack may earn their place. Others may need more thought.

I’m delighted people are watching. I’m even more delighted they’re asking questions.

Now I want to understand what’s connecting with them, and use that insight to make something I’m proud of.

You can watch SciFact on YouTube. And if you’re building something yourself—or have thoughts on any of this—come and find me on LinkedIn or X. I’d love to compare notes.

By this morning, it was a playable demo. So, in the spirit of building in public, I’m sharing it while there’s still plenty to work out.

Think Wordle for debating: one daily motion, a side to defend, and three short rounds to make your case. You can type or speak your argument, but the countdown keeps you honest. Being assigned a position you don’t agree with is part of the fun.

The idea centres on two distinct AI roles: a Debater committed to the opposing position, and an independent Chair judging both your reasoning and how well you express it.

This is explicitly a demo. The opponent is currently scripted and the verdict is illustrative. It tests the format; the live AI roles are still to come. You can reset and explore three sample motions.

I’d love to know whether it makes you want another go—and what you’d need to trust the Chair’s judgement.

Try the Debate Society demo

At about half past one in the morning, I decided I might like to make a science channel.

By lunchtime, I’d launched one.

I did sleep in between. Before this turns into one of those posts about how much you can achieve if you simply stop having normal bodily requirements.

The channel is SciFactShort: animated science and tech explainers, around 45 seconds to a minute long, offering soft, playful ways into everyday questions.

Why does a microwave spin your food? How heavy is a cloud? Why does your phone battery give up when it gets cold?

And, for a younger audience who may find this difficult to believe: what was dial-up internet?

Yes, we used to phone the internet. It screamed at us while connecting, occupied the household telephone line and occasionally caused a family argument.

I’m talking as though most of my readers don’t remember engaged telephone lines and images downloading line by line. Sorry. We’ve all aged beautifully.

1:35am: I could make something with this

The starting point was a PBS Space Time video about perpetual motion machines. It got me thinking about science education and whether I could make something easier to approach.

The audience I had in mind has never known a time when you couldn’t click a button and reach an extraordinary amount of information. Having access to an explanation is almost taken for granted.

But you still need a reason to become curious. And sometimes you need a friendly place to start.

That was the feeling I wanted to follow.

One question. Something familiar enough to recognise, with something hiding inside it that you might never have considered.

You’ve seen a cloud. You’ve probably never tried to weigh one. You use a microwave. You might never have wondered why your dinner goes round.

There’s a small invitation in each of those questions. You already know enough to join in.

I wanted to turn that invitation into a video.

Before bed: give the modem a face

The first episode was “What was dial-up internet?”

For many of us, a modem was part of our first experience of going online. A beige box that connected the computer in your house to an enormous world outside it.

You heard it doing the work. The dialling, the screeching, the negotiation. Going online was an event, sometimes accompanied by somebody asking how long you were going to be.

For the audience I’m making these videos for, that experience is unfamiliar. Their internet is simply there.

So the modem in the animation got a friendly face.

I liked the connection in that. This little object had once helped us find our way onto the internet. Now it could help someone find their way into understanding how that happened.

The face gave the viewer someone to follow. You could watch the modem trying to make a connection, rather than start with a technical description of what a modem does.

Through those small scenes, an unfamiliar system became something you could recognise: a conversation, a wait, an interruption.

That was the creative idea I wanted to keep exploring. Give people something comfortable to approach, then help them discover what’s going on.

The finished animation was 43 seconds long. By about quarter to two, I had a video and a list of possible follow-ups.

Then I went to bed.

The next morning: follow the invitation

At around 10:24, I picked two more questions: why microwaves spin your food, and how heavy a cloud is.

One was more for the item-in-your-house-but-I-don’t-know-how-it-works explainer videos. The other was a genuine science-surprise: a cloud is as heavy as a 100 elephants! What?!?

A cloud looks light. What happens when you put its weight on a set of scales and bring in some elephants?

The subject is science. Making the video is a creative act.

I’m choosing what someone encounters first, what gives them confidence to keep watching and which comparison might help an explanation click.

A friendly character can do that. So can a visual joke, a familiar situation or a question you can’t quite leave unanswered.

There won’t necessarily be a face on everything. I’m not working towards a universe in which every household appliance requires eyes.

The feeling is more consistent than that: soft, playful, easy to enter.

The look belongs to the idea

That thinking carried into the visual direction.

Deep indigo backgrounds, rounded shapes, yellow, teal and pink. Clear captions, with the important words picked out. Movement that gives you something to follow.

I liked the warmth of flat animation and the possibilities it gave me for explaining things you can’t ordinarily see. You can go inside a battery, give a cloud some weight or let two modems have a conversation.

I wanted each episode to offer a small, satisfying piece of understanding. The sort that leaves you thinking “ohhh”, then perhaps asking another question.

Forty-five seconds won’t tell you everything about clouds. It can make you look at one differently.

That feels like a worthwhile thing to make.

11:22am: the atom was too cute

By late morning, it had a name: SciFactShort.

The episodes were beginning to belong together. Then came the logo.

One option was an atom mascot. I liked it, but the face was too cute.

Soft and friendly still needs some personality. I wanted curiosity with a little mischief in it. A raised eyebrow. The suggestion that this character had an experiment underway and probably hadn’t mentioned it to the appropriate authorities.

Two expressions emerged: the Schemer and the Mad Scientist.

The Mad Scientist won. Goggles pushed up, mismatched eyes, enormous grin. There was somebody who looked pleased to have discovered something and keen to show you.

I emailed it to a friend I knew would love it.

That’s a good feeling: making something that amuses you, then immediately wanting somebody else to see it. Little tip for anyone who has a creative friend: don’t show too much interest, because they’ll ask for your reaction to every expression tweak, background graphic, and video adjustment… To that friend who shall we remain nameless, whoops and sorry :).

By lunchtime, it existed

The videos went into Google Drive. The ideas, publishing dates and descriptions went into Notion. There was a YouTube banner and a home on my website.

The organisation of knowing where to put the files and where to put copy and images was important to me. I’m not a very organised person so having a clear view of files goes here and text goes there was important. The other bit about the organisation was that organisation doesn’t have to be perfect! Notion might not have been the best place and Google Drive might not have been the best place. But it worked. And in this case working is better than perfect.

By roughly half past twelve, I’d launched the channel.

YouTube Shorts, Instagram Reels and TikTok suited the idea. A question can meet someone while they’re scrolling. They don’t need to have decided to study science or search for an explanation.

Perhaps they stop because they recognise the object. Perhaps the question surprises them. Perhaps there’s a beige modem making an unreasonable noise.

The opening only needs to earn enough curiosity for them to stay.

I like the possibility of that encounter: a small explanation arriving in the middle of someone’s day and giving them something to take away.

I kept making. Teletext. Cold phone batteries. VHS tapes. Why you can’t tickle yourself. Why the sky is blue. How your phone knows where you are.

By early afternoon, nine episodes had been produced, with more to release over the following days.

In the first three hours I had a little bit of traction, a few viewers but nothing to write home about. I didn’t do anything to promote the videos. As I’ve always said to clients: you have to find ways to tell people that your content exists. Promotion is of course the next part of the cycle for me. Then analysis of what’s starting to work. Then more content.

The first videos used captions, synthesised effects and a soft synth music bed, with voiceover still to come. There were things to add and things to improve.

There is a tendancy to want to go back over the old videos and ‘fix’ things that you discover during the creative process. My advice: that is almost always a bad idea. Keep moving forward and evolve what you’ve got. If that idea of evolving your creative live gets you scared, I suggest you go back and look at the very first episodes of The Simpsons. They’ve been going for 36 years and, wow, have they evolved over the years!

I didn’t want to perfect the enthusiasm out of it

I could have spent weeks preparing this.

I know enough about the creative process and marketing to generate a substantial amount of preparation if left unsupervised. Audience research. More visual routes. More naming options. A slightly different eyebrow.

Some of that might have improved the channel. Some might simply have used up the energy that made me want to create it.

I didn’t want to spend a hundred-million hours perfecting the possibility.

I had a feeling. The moment told me there was something worth following, and I had enough in front of me to act on it.

I was making judgements all the way through. The atom changed because its personality wasn’t right. The questions earned their place because I could see an inviting way into them. The episodes developed a shared look because they belonged to the same idea.

I could keep making those decisions while also letting the work exist.

Twelve hours, plus everything that came before

There’s a large qualification to the speed of this.

I’m luckier than most in the tools I have at my disposal. Creative experience, design, AI expertise, marketing. I can move between having an idea, making it and getting it in front of people.

I also grew up alongside the internet. Some of the early episodes draw on things I remember using, waiting for and occasionally shouting at.

That experience gives me material. The tools give me ways to do something with it.

AI made the production speed possible. My experience helped me recognise the direction I wanted and make the decisions that kept it moving.

So, yes, about twelve hours elapsed between the idea and the launch, including sleep. A few hours of actual work.

But I brought rather more than twelve hours of experience to it.

What excites me is how close an idea can now be to something I can actually make while I still feel the urge to make it.

A little closer to Monster House

This also feels like a precursor to Monster House, my children’s book series.

There’s a connection in the way a character can invite you into something. You become interested in who they are, follow what happens to them and find yourself exploring a world along the way.

SciFactShort gives me another place to work with that instinct. How do you make something approachable? What makes somebody curious? How much personality can a small expression carry?

A face on a modem is one answer. There will be others.

For now, making this has got that part of me moving. I’ve followed a creative feeling far enough for other people to encounter it.

Now people get to see it

You can find SciFactShort on YouTube, TikTok and Instagram.

I’m interested to see which questions catch people, what they remember and what they ask next.

At half past one in the morning, I had a feeling about a science channel. By lunchtime, a little beige modem was helping explain the internet.

It used to be our way in. I quite like giving it that job again.

Fancy a conversation about this? Find me on LinkedIn or X, or drop a comment below.