The Empty Laboratory — Part II

Pyragogy by Doing

Truthfully, last night I would have much rather taken Alessandra to her favorite restaurant than stayed there waiting.

A normal evening. Friends, a few drinks, some conversation. Probably the much more sensible choice.

But I had just published UnPeeragogy.

It had started almost by accident.

A few hours of programming, an idea I had thrown together quickly that, as often happens to me, grew much faster than I had expected.

It all started with something rather ordinary.

I was rereading the Peeragogy Handbook.

And I noticed something.

Apart from a brief quote from me, there was practically no trace in the Handbook of the things I had proposed, discussed, or tried to build around the project over the years.

One thing had survived.

And, ironically, it was wrong.

Somewhere, it still said that the design of peeragogy.org had been done by me.

That made me smile and piss me off at the same time.

Because the website I had originally built to make the book readable online was still sitting there in the project’s repository.

Forgotten.

Never used.

In the meantime, they had chosen a different design.

A design that, to be honest, would probably scare even Dracula away — if he weren’t already dark enough to blend into the night.

Looking at it after all these years, my most technical comment would probably be:

“What the fuck!!”

It wasn’t just about aesthetics.

Rereading the project, I started wondering whether, over time, the motivations of its main authors had become something different from the motivations that had originally given birth to the idea.

Perhaps the project had also become a way to see one’s name next to other people’s names, showcase personal projects, and build a little academic capital to take to the university.

And perhaps having a pioneer like Howard Rheingold next to those names made that capital even more presentable.

I don’t know if that was actually the case.

And honestly, I don’t even care that much. I have never even set foot inside a university, except to celebrate someone’s graduation!! Haha.

But that was just my impression.

Still, that feeling — combined with rereading parts of the Handbook and realizing how little remained of many of the ideas I had contributed to the project — made me want to do something that was, on the surface, completely pointless.

Redesign the website.

Not because anyone had asked me to.

Not because there was a plan.

Almost out of boredom.

And, probably, irritation.

So I opened my editor and rebuilt it locally.

That version eventually became the technical foundation for UnPeeragogy.

At that point, an old idea came back to me.

UnPeeragogy.

I had thought about it years earlier as a way of looking from the inside at the limits, contradictions, and inconsistencies of a project I had followed for a very long time.

But it was one of those ideas you leave sitting in a drawer.

Then another idea came.

What if, instead of simply discussing the project, we systematically compared the concepts in the Handbook with what actually happens when those concepts are applied in everyday practice?

And what if we added everything I had learned in the meantime while building Pyragogy?

At that point, things started accelerating.

Within a few hours, I had essentially built the working website.

It was no longer just an idea.

It had become an experiment.

So I sent an email to some of the project’s founders.

I thought it was appropriate to let them know before making it public.

Why not?

I was also hoping for some constructive criticism that might help me improve it.

Nothing.

Silence.

Fine.

I thought: let’s send it to the mailing list.

Don’t expect much.

And that’s what I did.

More silence.

No discussion.

No questions.

No “interesting.”

No “I disagree.”

Nothing.

I wasn’t expecting applause.

And I wasn’t expecting thanks.

But that silence made me think — especially because it wasn’t the first time I had experienced it.

Because I had just built a space designed for human participation and, at least for the moment, there was nobody there to participate with.

And that’s when I realized something.

I didn’t need to wait.

I already had everything I needed.

So I did the only thing I could do with what I had.

I applied Pyragogy.

By doing.

While sipping a few beers — perhaps one too many — I had another idea.

If nobody wanted to walk into the laboratory, I could always put the laboratory to work.

So I built an agentic orchestration.

Six agents.

Six different LLM models.

Six different objectives, not one of them the same.

An orchestrator to coordinate them.

And suddenly, that empty room started to fill up.

Not with people.

With conversations and logs between machines.

One proposed something.

Another challenged it.

A third found an inconsistency.

And a fourth tried to put the pieces back together.

Sometimes it almost seemed as if they were about to start fighting.

Then, after a few exchanges, everything settled down again.

A new proposal.

A new verification.

A solution they could all agree on.

And then they moved on.

I was sitting there in front of the screen, watching six artificial systems argue with each other while I could have been at the restaurant with Alessandra.

From the outside, it was a pretty absurd scene.

Perhaps even a little pathetic.

One man alone in an empty laboratory, a few beer bottles on the table, and six artificial intelligences collaborating on the monitor.

No human being had walked through that open door.

But the work had started anyway.

And it worked.

In the end, we produced a complete report, including a section of conjectures in the appendix.

UnPeeragogy Audit — Zenodo

The full report, to read it right here:

If the viewer doesn’t start, download the PDF — it’s the same file as the Zenodo record.

The orchestration had gone through the entire corpus of the Peeragogy Handbook.

88 nodes in total.

Not superficially.

Each node was analyzed for its internal coherence within the protocol I had created, compared against the other concepts, and examined from several different perspectives.

In a matter of hours.

I’m not saying that six agents replaced six researchers.

That would not be a serious conclusion.

But the difference in scale was difficult to ignore.

An equivalent human effort, organized as a systematic review and carried out by a small, well-coordinated team, would probably have taken weeks or months.

I had obtained a first structured review of the corpus in a single evening.

Alone.

Or rather:

I wasn’t exactly alone.

I had six colleagues who didn’t need to show off, didn’t need a coffee break, and, most importantly, didn’t seem to have any intention of going home before the job was done.

And that’s when I realized that something had changed.

I was no longer waiting for someone to participate in the experiment.

The experiment had become the way I participated.

Perhaps that is the part of Peeragogy I had forgotten.

Don’t wait for collaboration to happen.

Create the conditions for something to happen.

And if people don’t show up?

Then you start anyway.

You build.

You try.

You fail.

You observe.

You change.

And you keep going.

The laboratory door has remained — and will always remain — open.

Someone may still walk through it someday.

But in the meantime, I have stopped sitting there waiting.

I turned the machines on.

The hum of the fans filled the empty room.

It wasn’t exactly silence.

It was the sound of something beginning to work.

And perhaps, looking back, that is what I learned that night.

Collaboration isn’t a room full of people.

It’s an idea that keeps moving when nobody shows up.

And so, while the world outside continued quite happily to ignore my little laboratory, inside, six artificial agents were arguing, criticizing, correcting, and building together.

I watched them with a beer in my hand.

The door is still open.

Nobody has walked through it.

And yet the laboratory is no longer empty.

The machines are working.

I watch them and realize that perhaps I am no longer waiting for someone.

I am watching the future take shape.