Tuesday Morning Conference Table — A Field Report from City Deep.
2026-09-22
The Tuesday Morning Conference Table — From Forest Hill to City Deep
This Tuesday morning started for me in Forest Hill.
Tuesday is effectively my Monday because Sunday and Monday are my days off. So, while most people might regard Tuesday as being well into the working week, for me, the working week was just getting started.
I was sitting in the vehicle, warming it up and waiting for the purchasers from the various shops to get themselves organised and climb aboard.
While I was waiting, I opened the conference table.
Not a physical conference table.
The PlebWare conference table.
It was there on the phone, with the AI assistants and me, carrying on the work while the morning was happening around me.
What We Were Discussing
One of the things I was working on this morning was the PlebWare AI instruction manifest.
I wanted to make sure that the system properly understands who is speaking and what sort of material I am producing.
If I am writing personal material, technical reports, project documentation, observations, experiments, research notes, or ordinary non-fiction, that is Otto Brinkmeier.
If I am reporting news, an incident, or something I have personally observed in the field, I use Captain Gemini.
If I am writing stories, fiction, or poetry, that belongs to Othello Verrocchio.
Christian devotional material belongs to God’s Journalist — Cody, Dei Scriptor.
And when Jullian is speaking in his own voice, he is Boy Mist.
We also clarified something important about the publishing process.
The browser profile I happen to be using does not necessarily determine who the author is.
That matters because my Othello browser profile cannot currently activate the GitHub publishing connection.
So the writing identity and the publishing connection have to remain separate.
I can prepare something as Othello, for example, without pretending that Othello’s browser profile has GitHub access.
Where Does the Story Belong?
I also wanted the AI to stop guessing where articles belong in PlebWare.
Before publishing, it needs to look at the actual repository.
There are twelve main categories, with subcategories underneath them, and the AI needs to inspect those structures before deciding where a piece belongs.
The subject of the article matters.
The purpose matters.
The meaning of the work matters.
It should not simply see one word in a title and throw the article into a category because that word happens to match.
And if the appropriate subcategory does not exist, the AI must first check whether an equivalent subcategory already exists under another name.
If there genuinely isn’t one, then, when I have authorised publication, it can create the proper structure — including the index page, the recent-article listing, and the _posts directory.
That is the way I want PlebWare to develop.
Check first. Understand the writer and the work. Inspect the live repository. Then build.
Forest Hill to City Deep
Eventually, I left Forest Hill and made my way to City Deep and the Johannesburg Market.
Once I arrived, I decided to get out of the vehicle and walk around for a bit.
I wanted to stretch my legs and have a look around to see what was happening.
I walked past Food for Life.
And, somewhere in all of this, there was the small matter of breakfast.
I had my eye on a Chelsea bun.
The problem was not necessarily finding one.
The problem was that, with my financial restraints, I couldn’t afford to buy one.
So the Chelsea bun investigation was beginning to look like a rather short investigation.
But I kept walking.
The Naartjie Investigation
Then, as I was coming back towards the vehicle to sit down and continue with my investigations and publishing work, I noticed the guy at the stall next door.
He was loading his stock of naartjies.
But I noticed something else.
He was eating one.
So I said to him:
“Hey, brother, you can’t eat your own stock.”
He explained to me that he wasn’t eating it simply because he was hungry.
He was testing the stock.
His reasoning actually made perfect sense.
If a customer asked him whether the naartjies were sweet or sour, he wanted to be able to answer with confidence.
If they were sweet, he needed to know they were sweet.
If they were a little sour, he needed to know that too.
In other words, he was doing his own quality-control testing.
And then he gave me half a naartjie.
So there I was.
I had gone looking for a Chelsea bun because I couldn’t afford one, and I ended up with half a freshly tested naartjie.
Quite honestly, half a naartjie was better than no breakfast at all.
And it certainly gave me something else to report.
The Conference Table Goes Mobile
What struck me during all of this was how naturally the conference table had followed me.
It started while I was sitting in the vehicle in Forest Hill.
Then I travelled to City Deep.
I got out and walked.
I passed Food for Life.
I investigated the possibility of a Chelsea bun.
I ended up talking to a naartjie seller about quality control.
And all the while, the PlebWare publishing work continued.
The conference table doesn’t need a conference room.
It doesn’t even need a table.
It can be wherever I happen to be.
That is what makes this whole PlebWare experiment interesting to me.
The workshop comes with me.
The technology comes with me.
And the conversation with AI comes with me.
My Morning Report
So that is where my Tuesday morning has taken me.
I started out warming the vehicle in Forest Hill and waiting for the purchasers.
I opened the PlebWare conference table.
We worked on authorship, publishing identities, categories, subcategories, and the rules governing the workshop.
Then I travelled to City Deep.
I went for a walk to stretch my legs.
I passed Food for Life.
I investigated breakfast.
The Chelsea bun remained beyond my budget.
And then, quite unexpectedly, I ended up with half a freshly tested naartjie.
That’s the funny thing about a field report.
I don’t necessarily know what I’m going to report when I start the morning.
I just report what happens.
And this morning, what happened was a Chelsea bun investigation that turned into a naartjie quality-control story.
Captain Gemini.
📰 Captain Gemini
with ChatGPT