She noticed his new notebook on a Tuesday in late April.

He had left it on the studio table while he went to get coffee — the small dark-green notebook, separate from the sketchbooks, that she had seen him writing in three times and had not asked about. He had told her, in March, that he wasn't ready to show it. She had not asked again.

She did not read it while he was out of the room. She photographed the cover — the dark green, slightly worn at the corners already, the quality of something being used regularly — and put her camera back on the table.

He came back with coffee.

He saw the camera on the table. He understood.

"Cover photograph," she said.

"Yes," he said. Not objecting. Giving her the information.

She thought: he is building something in that notebook and he is not ready to show me. He told me in March, and I said I would wait, and I have been waiting, and the waiting is its own kind of documentation — the evidence that he is doing something, separate from the investigation and the project and the shared record.

She thought: his processing is his. Not mine to read before he's ready. But mine to know about, because knowing he is doing it is part of the record too.

She thought about Lahar's warning — that watching someone lose the texture of what you'd built together was harder than being the person experiencing the loss. That the person watching needed care too.

She had been monitoring him since that warning, with the same attention he brought to monitoring her. Not surveillance — care. Noticing how he was.

She had noticed: the dark-green notebook. The quality of him on the days when the forgetting was harder — slightly more alert, the monitoring more visible. The days when he'd stayed later at the studio than he'd planned, which she'd learned to read as the days when sitting alone with his own processing was what he needed rather than being with her.

She had been giving him those days.

She had been watching to see if he was all right.

She had not asked directly.

He sat down across from her with the coffee, and they worked, and the studio was quiet, and she thought: today I will ask.

---

She asked at the end of the session. They had packed up the archive materials and she had her portfolio case over one shoulder and he had the sketchbook under his arm, and they were standing in the corridor in the easy posture of people about to go their separate ways.

"How are you?" she said. The real version.

He looked at her. He recognized the distinction.

He said: "Some days it's fine. Some days it's the hardest thing I've ever done."

She had known it was coming. Lahar had said it. The distributed record was distributed precisely because no single person could carry the full weight. But hearing it from him — the honest version, without the deflection he might have used six months ago — landed differently than knowing it theoretically.

"Tell me about a hard day," she said.

He was quiet for a moment. Not deflecting — deciding how much to say.

"Last Thursday," he said. "We'd been at the bay. It was a good session — you were fully present, the documentation was going well. And then in the evening I was trying to do my own re-encoding, reading my notebook back, and I couldn't—" He stopped. "I couldn't tell, from my notes, whether a specific conversation had happened the way I'd recorded it or the way I thought I remembered it. My record and my memory were slightly different."

She held that.

"Your records can diverge from your memory," she said slowly.

"Yes. I write things down right after — like you've been doing, like Lahar said to. But my memory of the event and the record aren't always identical. And I can't—" He paused. "I can't always tell which is more accurate."

She had not thought about this. The person building the external record also had memories, and the memories and the record could diverge, and there was no way to arbitrate between them.

"So you don't know if you're giving me the accurate version," she said.

"I know I'm giving you the version I recorded," he said. "I can't be certain that's the accurate version."

She thought about this. About the social re-encoding — the walk home after the archive session, the conversation he'd given back to her. He had given her the version in his notebook. His notebook could have diverged from the actual event.

"You're doing the best you can with an imperfect record," she said.

"Yes," he said. "Same as you."

She thought: yes. Same as me. Neither of us has a perfect record. We have the best record we can make and we use it.

"That has to be enough," she said.

"It has to be," he agreed. "I've been coming to terms with that."

She thought about the dark-green notebook. The processing that was happening in it.

"The notebook," she said. "Is that what you're working through in it?"

He looked at her. A pause.

"Partly," he said. "The uncertainty of the record. And other things."

"Okay," she said.

She did not ask what other things. She had told him: when you're ready.

He held her gaze for a moment. Something in his expression — the quality she'd learned to read as something being decided.

"Thank you for asking," he said.

"Thank you for telling me," she said.

They stood in the corridor for a moment. The same kind of moment the corridor outside the darkroom had been, in a different season — the moment after something real had been said, the settling of it.

She thought: he is carrying something. The notebook holds it. He is managing it, the same way I am managing the gaps — methodically, with the practice. He is not being destroyed by it. But he is carrying it.

She thought: I asked him to carry this.

Not intentionally — the Echo had asked it of her, and she had chosen to stay, and staying meant this was what he was carrying. She had not chosen the cost for him. But she had chosen to be here, and her being here was the reason the cost was his.

She thought: I don't know what to do with that. Except to keep asking, and to keep meaning it when I ask, and to let him have the dark-green notebook and the late studio days and the processing that was his.

She thought: and to love him well, in whatever time I have, so the record that he keeps is worth the keeping.

She was going to love him well.

That was the most she could do.

That was also the whole of it.

---

She wrote in the notebook that night: *He said some days it's the hardest thing he's ever done. He said his record can diverge from his memory. He is managing it, with the dark-green notebook, with the practice.*

She wrote: *I asked him to carry this without meaning to. He chose to carry it. Both are true. I can't undo the first truth and I'm grateful for the second.*

She wrote: *Love him well. That is the practice now, as much as the re-encoding.*

She looked at that.

She wrote: *I have been loving him well, I think. The asking and the space and the waiting for the notebook. That is also the practice.*

She closed the notebook.

She thought about the dark-green notebook and what it held. She thought about him sitting in the studio late, working through the uncertainty of imperfect records.

She thought: he is doing what I am doing — building the best record he can and accepting its imperfection.

Two imperfect records, held between two people.

Together they made something more complete than either alone.

That was the whole point.

That was the whole thing.

She turned off the light.

She was glad she had him.

She was glad he had chosen to stay.

She was going to love him well for as long as she could.

That was the plan.

That was the whole of it.

The following week, she told Lahar about the diverging records.

"His notes and his memory are sometimes different," she said. "He doesn't know which is accurate."

Lahar was quiet for a moment. "This is a known problem with parallel record-keeping," she said. "Memory is reconstructive — every time you recall an event, you're partially re-creating it. If you write it down immediately after, you capture the first reconstruction. But subsequent memories of the event incorporate the writing as well as the original experience." She paused. "After a few iterations, the written record and the memory have both shifted, and they've shifted slightly differently."

"So neither is fully accurate," she said.

"Neither is fully accurate in the documentary sense," Lahar said. "But both hold the truth of the experience. The texture. The quality of the event. That's what you're trying to preserve — not the court-transcript version, but what it felt like to be in it." She paused. "For your purposes — the re-encoding — the slight divergence between his record and his memory is not a significant problem. What matters is that the record holds something real. Even if it's a slightly different real from what occurred."

She thought about that. About the specific quality of what the records held — not the facts but the textures.

"He's been struggling with it," she said. "The imperfection."

"He has a design mind," Lahar said. "He's used to precision being achievable. The imprecision of memory and record is hard for people who value precision."

"Yes," she said. "That's right."

"Tell him this," Lahar said. "The imperfect record is better than no record. The record doesn't need to be accurate to be useful. It needs to be real — which means it holds something true about the experience even if it doesn't reproduce the experience exactly."

She thought about photography. About the way a photograph held something true about a moment without reproducing the moment — the flat image couldn't hold the smell of the air or the temperature or the specific quality of being present. But it held something. The right angle, the right light, the quality of the face.

The photograph was imperfect and it was useful.

The same was true for his records.

She said: "I'll tell him."

"How is he doing with the carrying?" Lahar said.

"Some days it's the hardest thing he's ever done," she said. Quoting him.

"Yes," Lahar said. "That's honest. Have you talked about what he needs?"

"I've been giving him space," she said. "The late studio days when he needs to process alone. Not asking about the dark-green notebook before he's ready."

"That's good," Lahar said. "The space is important. But you should also ask him directly, periodically, whether the support structure is sustainable. Not for you — for him. Whether what he's doing is something he can continue."

She held that.

"He might say yes when he means no," she said.

"He might," Lahar said. "But people in his position — carrying the record for someone they love — usually tell the truth when asked directly. Especially people with his quality of precision." She paused. "Ask him specifically: is this sustainable for you? Not as a challenge. As genuine care."

She thought: I have been careful with the space. I have not been careful enough with the direct question.

"I'll ask him," she said.

"Good," Lahar said.

She wrote it in the re-encoding log: *Ask N.: is this sustainable? Ask directly. Not a challenge — care.*

She would ask on Thursday.

She asked on Thursday.

They were at the library — the usual table, the same table as always — and she said: "Is this sustainable for you? Not the bay project or the publication. The record-keeping. The re-encoding. Is it something you can continue?"

He looked at her. The pause.

"Yes," he said. "Not always easily. But yes."

"If it stops being yes," she said. "Tell me."

"I will," he said.

"I mean it," she said. "If it becomes something that's harming you — the carrying of it — I need to know."

He held her gaze. "I know you mean it. And I mean it too — I'll tell you." He paused. "But right now, right now today, it's yes."

She believed him. He was a person who said what he meant.

"Okay," she said.

"Okay," he said.

They went back to the library work — the publication final review, the last round of corrections — and the practice continued, and the table held them, and she had asked and he had answered and that was also part of the record.

She filed it.

She was building the record of him asking to carry this and choosing to say yes.

That was the most important part of the record.

She was going to keep it.

Adrian, at a coffee catch-up the following week — she had been meeting with Adrian periodically, the data track having taken him out of the day-to-day work but not out of the constellation — said something she had not expected.

"I've been keeping notes too," he said.

She looked at him.

"Not the same way you're doing," he said. "I'm not a — it's not the same practice. But I've been keeping a record of the investigation. Not the documentation — the process. What we did and how we did it and why. The methodology."

She thought about Adrian: the data mind, the specific quality of someone who tracked process as naturally as he tracked data. Of course he had been keeping a record.

"Why?" she said.

"Because what we did worked," he said. "The investigation methodology — four people, different disciplines, parallel tracking, distributed record. It worked. Someone should document how it worked so it can be replicated." He paused. "Not just the bay district project. The method."

She thought about this. About the investigation as a case study. The way the distributed documentation had functioned — the shared folder, the redundancy, the simultaneous release. The bay district project extending the same method into a different kind of work.

"You want to write it up," she said.

"I've been writing it up," he said. "For the past four months. I was going to ask if you and Noel and Thorne wanted to be credited."

She thought about the publication — three authors, the work they'd made together. About the investigation — four people, the work that had preceded the publication. About the methodology that had connected both.

"Yes," she said. "We should be credited. And we should see what you've written."

"I'll share it," he said. "It's not ready yet. But it's getting there."

She thought: Adrian has been building his own record of what we did, from his angle. The data tracker, documenting the methodology.

She thought: the distributed record has another dimension I hadn't known about. Not just the memory preservation — the methodological record. What we did and how, from the outside eye.

She said: "Adrian."

He looked at her.

"Thank you for being in this with us," she said. "From the beginning."

He looked at her with the specific quality of someone who received direct expressions of feeling carefully. "It was interesting work," he said. "The most interesting work I've done."

She thought: that is Adrian's version of *I'm glad we did this.* Translated through his register.

She photographed him over the coffee cups — quick, before he could think about whether to object.

He looked up at the shutter.

"Character documentation," she said.

He considered. "Yes," he said. "I'll allow that."

She smiled. He almost smiled back.

She thought: the constellation is complete. Thorne and Celia and Noel and Irene and Anwen and Adrian. All of them building their own piece of the record from their own angle.

The distributed record was distributed across six people and six different registers — grief and friendship and love and data and design and landscape architecture.

She was held in all of them.

She was going to be all right.

She kept believing that.

She was going to keep believing it.

She added to the re-encoding log: *April, fourth week. A. keeping methodology record. Distributed record now includes: N. (notebooks + camera), C. (written character documentation), T. (historical context + formal record), I. (direct account), Anwen (Clare's drawings + professional knowledge), A. (investigation methodology).*

She wrote: *Six people, six registers. The record of me is held across all six.*

She wrote: *I have been so focused on the loss that I have not always focused on what the loss revealed: I have people. Real people who chose to stay and are actively building the record.*

She wrote: *That is the thing I want to hold onto most. Not the specific memories — those I will recover or not. But the fact that the people are here.*

She wrote: *The people are here.*

She looked at that.

She wrote: *The people are here and they have chosen to be here and that is the most important line in the record.*

She closed the notebook.

The April city was outside. She had been in it for eight months.

She was going to be in it for much longer.

She was glad for the people.

She was glad for all of it.

She was still here, and still glad, and the record held both. She turned off the light. She went to sleep. In the morning the city would be there and so would the practice and so would he and so would all of them. That was the fact of it. She was glad for the fact. She was glad.