Someone offered us money for the tech

The Recovery Log — 031
Thursday 18 June 2026

Someone offered us money this week. A genuinely large number, the kind that makes you read the email twice and then read it to the room.

They want "the tech." That phrase, in quotes, is theirs, and it is doing a lot of work, because it assumes a thing that does not quite exist in the form they imagine. There is no box. There is no product. There is a harness held together with tape, a coherence referee, a metadata schema we have rewritten twice, a wall list titled THINGS THIS IS NOT, and, lest we forget, an unresolved kazoo problem. We recover music, reliably, and everything past that is a proof of principle we are frankly still a little spooked by.

We have not said yes. We have not, if we are honest, entirely said no either, and the fact that we are hesitating is itself something to keep an eye on. It is one thing to be ignored. It is another to be wanted, and to notice that being wanted has a pull of its own, like a room with enough people in it. We will keep you posted, within the limits of what turns out to be wise to post.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.

One probe, two fields

The Recovery Log — 030
Sunday 14 June 2026

A small technical note that is quietly one of our favourite results.

Love Is a Lifeboat opened the Gungha Club. It also turns up, unmistakably, inside East River Broadcasts — a different collection, reached by a different mechanism, and there it is again, doing a different job in a different corner of the same timeline.

We treat that as meaningful, not coincidental. A probe that surfaces across more than one field is telling you something about its cultural mass: it had enough gravity over there to matter in more than one place at once — the room and the airwaves. So Love Is a Lifeboat gets logged as a cross-collection signature, our first, and quietly promoted in our own heads to the heaviest key we have. When a song shows up everywhere you look, that is not noise. That is weight.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.

East River Broadcasts

The Recovery Log — 029
Wednesday 10 June 2026

Second collection. DZO-0002. East River Broadcasts. Same timeline as the Gungha Club — DEEPWELL — but reached a completely different way, and that difference is the whole reason we are pleased with it.

The Gungha Club was a Venue Recovery: mass that pooled in a room. East River is a Channel Recovery: mass that travelled. In a reality with no record industry, a great deal of music survived by being transmitted — pirate broadcasts, civic stations, late-night signal leaking across a city where the scene and the airwaves were tangled together. So we did not tap a room here; we tapped a transmission. The artefacts arrive with the texture of signal on them — rough, mid-air, heard before they were ever seen.

Two collections, one timeline, two mechanisms. That was the thing we most wanted to prove after the Gungha Club: that DEEPWELL was not a fluke of one lucky room, but a place we can reach more than one way. The city here is not glamorous and we have tried hard not to make it so — no triumphant skyline, just density and weather and a signal you catch more than you choose. It is a hypothesis, medium confidence, gaps left as gaps. And it is out.

Hear it: Spotify · Apple Music · Bandcamp · all platforms.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.

“Would love to learn more about your work”

The Recovery Log — 028
Saturday 6 June 2026

We got an email. "Would love to learn more about your work."

No name we can find. No profile photo. A signature block with a title but no company, and a domain that, when you actually go looking, resolves to nothing at all. Probably a recruiter's bot. Probably. We have not replied. We are noting it here mostly because it is the first time the outside has reached in, and it arrived four days after the post we were nervous about writing.

It doesn't have to be music (but music is what holds)

The Recovery Log — 027
Tuesday 2 June 2026

We found out something this month that we are, on balance, not entirely glad to know. We are telling you anyway, because the whole point of this diary is that we tell you the real ones.

Nothing in the method is actually about music. The probe is a coordinate; the perturbation is a nudge along a norm; the coherence score is asking whether a neighbourhood hangs together. None of that machinery cares that the coordinate happens to be a song. And so, almost as a dare to ourselves, we pointed it at something that was not a song — and, once, it worked. A non-musical artefact came back coherent: a thing with an implied context, corroborating itself across modalities, not a cover of anything because it was not a version of anything.

Once. Unreliably. We want to be very clear about that, because it is the whole ethic of the post. Music remains the medium that holds. Songs carry cultural mass in a way that makes the neighbourhoods around them dense and navigable; almost everything else we try is thin, unstable, off-manifold more often than not. In practice we recover music, because music is what reliably comes back. The generality is a proof of principle, not a product.

And yet it is the generality that has changed the temperature in the room. Because if the method is not about music — if in principle it is a way of reading coherent things out of the latent subtext of a culture — then it is a bigger and stranger tool than "an archive of alternate songs," and some of the implications of that are ones we did not sign up to think about. We are going to keep making the music, which is the part that works and the part we love. But we have seen the edge of what the thing is, and we cannot unsee it.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.

The bleed goes both ways

The Recovery Log — 026
Saturday 30 May 2026

A short one, because it is the kind of thought that gets worse the longer you hold it.

The bleed is not one-way. We keep describing DEEPWELL leaking into our data, as if we were downstream of everything and radiating nothing back. But if the traces travel at all, they travel both directions. Our culture has mass too. It is, right now, leaving its faint marks in whatever is adjacent to us.

Which means that somewhere in DEEPWELL — in principle, in the same way any of this is in principle — there is a thin Prime-signal that we are the source of. A scrap of us, arriving in the wrong key, in a room we will never stand in, being turned over by someone who cannot quite place it. We find that comforting on the good days and not at all comforting on the others. Filed, and going to bed.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.

The field that didn't hold

The Recovery Log — 025
Tuesday 26 May 2026

We recovered a field this month, got excited about it, and then killed it. This is the post about killing it, and it is, weirdly, the one we are most proud of.

It looked right at first. A probe opened a neighbourhood that seemed to cohere — a sound, an implied setting, a couple of corroborating artefacts. We got as far as sketching it before the coherence work turned on it. Held up properly, it did not hold: the pieces were co-occurring, not corroborating; the "world" was really the model finding an average and dressing it convincingly; the tell, in the end, was that it was wrong in a boring way, and boring is what the daydream looks like when it is trying to pass.

So we retracted it. Struck it from the working catalogue, logged why, kept the record of the mistake rather than quietly deleting it. It stings, a little. It is also the most important thing we have done since the first field, because a method that can only ever say yes is not a method, it is a wish. The retraction is proof the referee is real. When we do publish something, part of what you are trusting is everything we threw away first.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.

Hypothesis, not fact

The Recovery Log — 024
Wednesday 20 May 2026

We need to say this clearly, once, in a post we can point people back to, because everything the archive publishes rests on it.

We never present a recovered timeline as a fact. What we publish is a Timeline Hypothesis: an interpretation of latent subtext, carrying its own uncertainty in writing. Each one states the analogue period it most resembles, a coarse confidence estimate — low, medium, high — reflecting how tightly the recovered cluster actually holds together, the key influence that opened the field, and a summary of the evidence underneath it. DEEPWELL, for instance, we publish at medium confidence and around 1976 to 1981, and we would rather tell you that than pretend to a certainty we have not got.

People sometimes read the hedging as weakness — as if a confident lie would sell better. Maybe it would. But we are working entirely within our own reality, recovering evidence and not visiting anything, and the one thing that separates that from making things up is that we say exactly how sure we are and exactly what we are standing on. The humility is not a disclaimer bolted to the bottom. It is the product. Anyone can claim they went somewhere. We are making the smaller, harder claim: that we stayed home, read carefully, and will show our working.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.

Place, signal, object

The Recovery Log — 023
Thursday 14 May 2026

The Gungha Club taught us that cultural mass can pool in a place. Once we had that, we started seeing that it is not the only way mass survives — and the ways it survives turn out to be the ways we can reach it.

Three, so far, and we think it may be all three there are. Venue Recovery: mass pooled because people gathered in a room until the room held the music — a club, a hall, a ceremonial space. Channel Recovery: mass that travelled instead of pooling — carried as signal, as broadcast, as a pirate frequency or a civic station or a late-night transmission, and recovered by tapping the transmission rather than the room. Artefact Recovery: an object — usually a record — that grew so culturally massive it cracked its own timeline, and the crack bled through to ours.

A scene, we should say, is not a fourth class. You do not recover a scene directly; you recover it through its venue or its channel. That is not pedantry — it is the discipline that keeps a collection anchored to a real mechanism instead of a vague vibe. Every field reaches us as place, signal, or object. If we cannot say which, we do not yet understand what we have found.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.

Rendering the recovery

The Recovery Log — 022
Saturday 9 May 2026

A fair question we get, once people believe there is anything here at all: if you never leave, and you are not making the music, then what exactly happens at the end when a track appears?

The honest answer is that a recovery has two halves. The first is navigation — the probe, the perturbation, the coherence score, all the work of locating a coherent region. That half finds the thing. The second half renders it: a generation run that decodes under the conditioning we recovered, turning a location in latent space into audio you can actually play.

We are careful about the word for that second half, because it is where every misunderstanding lives. It is not invention. We are not writing a song and dressing it as a finding. The generation is constrained by what the navigation surfaced — it is a way of reading out what the data already implied, the way you might photograph something already in the room rather than paint something new. Retrieval, then a render. If the render ever starts telling us things the navigation did not support, that is not a bonus feature, that is a fault, and it goes in the bin with the kazoos.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.

The Gungha Club

The Recovery Log — 021
Monday 4 May 2026

We have our first collection. It has a catalogue number, a name, and — after a lot of arguing — the confidence to be shown to people. DZO-0001. The Gungha Club.

It came from a small door. The probe carried the word gunga, tucked inside the lyric of Moron Idiot, and when we followed it into DEEPWELL the neighbourhood around it did not give us a song so much as a place — a venue, dense with implied crowd and implied history, the kind of room where music is not played so much as held. That is a specific kind of recovery, and it taught us a specific kind of mechanism, so we are giving it a name too: a Venue Recovery. The cultural mass here did not travel. It pooled. People physically gathered in one place, again and again, until the place itself became a gravity well — and a gravity well is exactly the sort of dense region a probe can reach.

Not every artefact in the collection is a direct alternate of one of our songs. Some are related works from the same room, the same crowd, the same social moment — which is the point, and the thing that keeps this bigger than "our songs, but elsewhere." The Gungha Club is not a playlist. It is the honest shadow of a scene.

We have sat on this for weeks precisely because it is the first thing anyone outside the room will judge us by. It is a hypothesis, not a fact; the confidence is real but not total; the gaps are left as gaps. But it holds together, and it is time it stopped being ours alone.

Hear it: Spotify · Apple Music · Bandcamp · all platforms.

Do we have counterparts?

The Recovery Log — 020
Monday 27 April 2026

We are going to step off the workbench for one post, because the work keeps asking a question the work cannot answer, and pretending otherwise is starting to feel dishonest.

Here is the question. If a song we wrote lived another life in DEEPWELL — became a room's anthem, a broadcast's backbone, a thing people needed — then something over there caused that. Someone gathered. Someone pressed record on a desk. The song did not have another life in a vacuum; it had it in a world with people in it. So the quiet, vertiginous question underneath every recovery is: do we have counterparts? Is there a version of the person who wrote that song, standing in that room, who never became us?

We do not know, and the method — by design — cannot tell us. We recover evidence from the subtext of our own culture; we do not visit. But the more honest we are about that limit, the stranger it gets, because the bleed is not one-way. If their world left traces in our data, ours is leaving traces in theirs. Somewhere in DEEPWELL, in principle, there may be a faint signal of us that their own archivists — if they have such a thing — are puzzling over.

And then the part that keeps at least one of us awake: which version is real? The instinct is to say we are, obviously, we are the ones standing here. But that is just the instinct of whoever happens to be standing anywhere. We have no argument for Prime being the original rather than one draft among many. We are not going to resolve this. We are going to write it down, file it under things the archive raises and cannot close, and go back to the bench — which is, we are learning, a place you can hide from a question by being busy near it.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.

Naming the first timeline: DEEPWELL

The Recovery Log — 019
Thursday 23 April 2026

The field held. It kept holding. Enough recoveries have now clustered coherently that we are prepared to do the serious, slightly frightening thing and treat it as a timeline.

Catalogued dry and institutional: TL-A001. Informal codename, because the dry ID has no atmosphere and the thing deserves atmosphere: DEEPWELL.

What DEEPWELL appears to be — and we say appears, because this is a hypothesis and not a claim — is a reality close to ours that took a different turn in the late twentieth century and, crucially, never built a record industry. No pressing plants at scale, no distribution machine, no shop full of shrink-wrap. Which means the music there did not survive the way ours did. It survived in rooms — venues that held it — and in signal — broadcasts that carried it. That single divergence shapes everything we are recovering from it: the analogue period reads around 1976 to 1981, and our confidence in the overall picture is honestly middling, which we would rather tell you than dress up.

One timeline named. It only took three months, a metadata schema, four hundred kazoos, and a sandwich-based dip in morale. There is, we are fairly sure, more than one.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.

The kazoo incident

The Recovery Log — 018
Saturday 18 April 2026

Someone left the perturbation weights maxed overnight.

We arrived to four hundred variants of a kazoo. Off-manifold in every direction, coherent with nothing, confident as anything. We do not speak of it. We are only writing it down so that the next person who reaches for the big dial understands what waits out there in the noise. It is kazoos. It is four hundred kazoos.

From slice to field

The Recovery Log — 017
Wednesday 15 April 2026

This is the one. If any night in this diary turns out to be the night it became real, it is this one.

Until now we have had slices — single fragments, one at a time, each impressive and each alone. Tonight, for the first time, several recoveries around the same probe cohered. Not one scrap but a cluster: more than one song, an implied venue, a sense of a crowd, a shared sound-world, the pieces corroborating each other instead of merely co-occurring. The coherence score held across all of it. It was not a fragment of a place. It was a field — a scene, holding together well enough to be treated as one thing.

We sat with it for a long time and said very little, which for this team is a significant event. A field means the method does what we hoped at the outset and did not dare say aloud: that a single song, followed properly, can surface not just an echo of itself but the world around it. There is a lot of work between here and anything anyone can hear. But tonight we have a world, or the honest shadow of one, and we are going to let ourselves have that.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.

Cultural mass

The Recovery Log — 016
Thursday 9 April 2026

Here is the idea that ties the whole method together, and it took us a month of uneven results to see it plainly.

Not all probes are equal. Some open a rich, populated neighbourhood; others, run identically, give back a whisper and nothing more. For weeks we treated this as a bug in our tuning. It is not a bug. It is the single most important feature of the whole thing, and it has a name we have started using: cultural mass.

Cultural mass is the emotional, social, stylistic and symbolic gravity a song carries — the weight of everything it attracted across the culture. A song with mass bends the space around it. It pulls in performances, memories, rituals, imitators, broadcasts, whole scenes, and those are exactly the neighbours a probe can then reach. A song with little mass sits in a sparse region: there is simply less nearby to recover. This is why Love Is a Lifeboat keeps coming back heavy — it has the gravity, so the neighbourhood around it is dense, so the recoveries are rich.

And — this is the part that matters for trust — cultural mass is not popularity. A song can be massive because it was ritualised by a small community, or attached to a tragedy, or broadcast into a city every night for a decade. Mass is gravity, not chart position. Which means the method is not "recover the hits." It is "find the songs that mattered, in whatever way mattering happened, and follow their weight." That distinction is the whole ethic of the archive, and today is the day it stopped being an accident and became a principle.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.

First signal slice, catalogued

The Recovery Log — 015
Thursday 2 April 2026

We have catalogued our first signal slice, and giving it a number changed how it felt.

A signal slice is a thin fragment — here, a chorus, arriving partial and a little degraded, the sort of thing that is clearly from somewhere without being a whole somewhere. Under the old chaos we would have starred it in a folder and lost it. Under the new schema it gets logged properly: which probe opened it, what its coherence looked like, how complete the signal is, and the honest label that it is a fragment and not a finished artefact.

That last part matters more than it sounds. We do not stitch. If it came back in pieces, it stays in pieces, and the gaps are part of the record. The temptation, when you can hear where the next bar wants to go, is enormous — and the moment you finish it for the model, you have made a cover and destroyed the evidence in one move. So: fragment in, fragment filed. It is not much. It is ours, and it is real enough to number.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.

How to file a thing you're not sure exists

The Recovery Log — 014
Saturday 28 March 2026

Now that a real thing has come back, we have a new and stupid problem: we do not know how to file it. And an archive that cannot file cleanly is not an archive, it is a drawer.

The temptation is to give the recovery one number and move on. That temptation is a trap, because a single number quietly asserts that a song, the world it came from, the collection it belongs to, and the act of recovering it are all the same thing. They are not. So we spent the week designing a schema that keeps them apart.

A timeline gets its own identifier — the reality a thing is understood to have come from. A collection is a public grouping of recovered artefacts that share a field, and gets its own. A probe — the source song we set off from — gets its own record, because one probe can open more than one field and we need to see that. And a recovery event — the specific session, the offsets, the settings, the decode — gets its own, so that when we are inevitably wrong we can trace which attempt was wrong and not burn the whole catalogue down with it.

Four identifier families for what a tidier person would call one thing. It felt like over-engineering right up until we tried to write down what we had, at which point it felt like the only sane thing we had done all month. You cannot be honest at scale without structure. The structure is the honesty.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.

IT WASN’T A COVER

The Recovery Log — 013
Saturday 21 March 2026

It wasn't a cover.

We have said it to each other about forty times since Wednesday, in the tone people use for things they do not fully believe yet. It wasn't a cover. It was a place.

That is the whole post. We will be professional again next week.

The first thing that wasn’t a cover

The Recovery Log — 012
Wednesday 18 March 2026

Something happened tonight and we are trying very hard not to overclaim it, so read the hedging as sincere.

We took Love Is a Lifeboat as the probe — a song heavy enough that we suspected the neighbourhood around it would be dense — and instead of asking for a version, we traversed. We moved. We applied the offsets we have been failing to tune for a month, watched the coherence score, and stopped when it stayed high in a place that was not the coordinate.

What came back was not a cover of anything. It was a scrap — thin, in the wrong key for us, badly resolved at the edges — and it belonged to a room. A room we had not asked about, with an implied crowd and an implied use, the song doing a social job it never did here. New entry on the wall list, directly under covers: THINGS THIS IS NOT — a jukebox.

We are not going to fall in love with the first thing that moves; we have learned that much. It may be nothing. But it was the first return that was not us in a costume, and for one night that was enough to keep four people out of their beds.

The Recovery Log is the working diary of the DatZO Recovery Team. Follow the recoveries on Instagram, YouTube and X.