161: Summer

June was Floors 16 and 17. July was Floor 18. August was Floor 18 again, and the third time we cleared it in 3 hours 8 minutes after the first two attempts had failed at the Phase 3 trigger.

The summer had the quality of sustained momentum. Not urgent — sustained. The three AM sessions continued on their schedule. TwilightTide's analyses came on their regular Thursday cadence, each one incorporating the week's session data into the formation adjustments. Wenqing tracked everything and updated everything. Old Wolf walked through boss encounters with the same deliberate pace he brought to everything — the kind of player who never rushed a briefing and never wasted a session, and who had the specific patience that comes from understanding that patience isn't passive. Zhu Yuhan was in the Iron Hills four mornings per week and the east wing every time, running her Priest protocols with the consistency that had become its own form of anchor.

The Phase 3 trigger on Floor 18's first two attempts had produced a mechanic that hadn't appeared in any server documentation — a reverse-aggro pulse that converted the formation's positioning from offensive to defensive in a single event, requiring a full formation reconfiguration in the four seconds before the next damage wave. It was a mechanic that could only be countered by preparation, not reaction, because four seconds was not enough reaction time for 85 people. TwilightTide had found it in the second attempt's session log and built the counter before Saturday. The third attempt used it. The floor fell in 3 hours and 8 minutes.

The guild was 118 members in June. 127 in July. 133 in August. The growth was controlled — we'd been selective since March, taking in only the players whose skills and working habits passed the council's vetting. Quality over count. Old Wolf's phrase, from a council session in April, and he'd meant it as a standing policy.

***

Wanqing was in Hangzhou more — the joint seminars ran through the summer, and she was at the east courtyard bench on Tuesday and Thursday mornings with the regularity that a fixed academic schedule produces. By the third week of June the Tuesday bench had become a fact rather than an event, the kind of thing you note when it changes rather than when it happens.

I stopped noticing the thermos and started noticing the problem set.

She switched problem sets four times over the summer, which was the rhythm of her research schedule — each topic replacing the previous one with the practiced ease of someone who'd been doing this long enough to know how their own concentration worked and what it needed. The transition between topics was not the break it might appear to be from the outside. She didn't stop working — the problem set changed but the depth of attention didn't. She arrived at the bench with the same quality of focus whether it was variational methods or the Japanese monograph, which meant the focus was hers and the topic was just the current direction it was pointed in. The topics advanced: variational methods in June, then asymptotic analysis in July, then a two-week stretch in early August where she was reading a Japanese monograph on functional equations that had been translated in the fifties and hadn't been updated since. She was reading it with the specific critical attention that suggested she was looking for something she wasn't certain was there.

She didn't tell me what she was looking for. I didn't ask. This was the pattern. It had been the pattern since February 2015 and the Sunday train and the poetry collection she'd been reading when I first saw her.

You don't ask people what they're looking for in the middle of looking.

The summer bench had a quality distinct from the winter one or the spring one. Full shade by mid-morning — the maple's summer canopy complete, no gaps, the light only getting through at the edges in brief strips. The stone arm warmer than in any other season; warm to the touch by nine AM if the sun had hit it at dawn. The campus ambient sound in June and July was emptier than the semester sound — fewer people, the particular quietness of a campus between sessions that still had its full physical presence but was operating at half its usual occupancy. By August the incoming students had started their orientation programs and the sound had shifted back toward the semester register. I'd sat at this bench through another full cycle now. The first full summer cycle at the bench with Wanqing across from me at the Tuesday-Thursday frequency that the joint seminars imposed. It had become structural fast, the way structural things do: gradually and then completely.

***

Father's recovery progressed through the summer on the immunosuppressant calibration schedule. Doctor Yan had said the first year's fatigue would be the primary adjustment factor, and it was: Father was tired by early afternoon on most days. But it was a different kind of tired from the pre-surgery illness tiredness. Before the surgery the tiredness had been losing-ground tired — the specific quality of a body fighting something it was slowly losing to. The post-surgery fatigue was working-against-a-fixed-resistance tired. The resistance was real, but it wasn't gaining on him.

He called on Tuesdays. Brief calls — ten minutes, sometimes twelve, the kind that covered the functional updates and one or two non-functional details that he'd decided were worth saying. The soup timing. The supply chain for the lacquerwork materials. Once, in July, he said that he'd been sleeping better since the surgery — not a lot better, a specific small amount better, enough to notice. I wrote it down in the call log I kept. I had a call log because Wenqing had suggested it in March, noting that keeping a record of the Tuesday calls would make it easier to track the recovery's progress over time. He'd been right. The July sleep note was different from the June sleep note, which was different from the May sleep note. The record showed something the individual calls didn't: a direction.

He was in the kitchen every morning before the tiredness set in.

By July he'd worked through the remaining soup variations and moved on to the workshop preparation — the lacquerwork tools and the water-pattern dyeing setup and the embroidery framing equipment, which he'd ordered from suppliers he'd worked with for twenty years and which had been arriving at the shop since June. He was methodical about the preparation the same way he was methodical about everything: systematic, unhurried, completing each category before beginning the next.

The shop was partially given over to storage and preparation while he was in the recovery period. This didn't bother him. The shop had been the right thing for thirty years; the workshop was the right thing for the next thirty. The transition made sense.

***

Xiaoyu called me on a Sunday in July to tell me she'd been accepted into the city's best high school on academic merit. She'd known since the spring entrance exams and hadn't told me until the official notification letter arrived, which was her way — she waited for the thing to be confirmed before she reported it.

"The school has a Japanese exchange program," she said. "I can apply for the exchange in second year."

"Osaka."

"Yes. The same partner school as the middle school sister program. It would give me formal certification for the translation work, which would make the contracts more consistent in the third and fourth year."

"Are you going to apply."

"Yes." She said it with the same certainty she brought to everything she'd decided to do — not aggressive certainty, just the clear factual version. "The university fund covers the application fees. I've been tracking the cost structure since February."

Since February of this year — while Father was in hospital. She'd been planning her second-year program while tracking the surgery prep costs.

"You've been planning this since February," I said.

"Since March, formally. February I was just watching the numbers." A pause that carried the specific quality of someone who had something else to say and was deciding whether it was the right moment. "The craft workshop opens when Father is ready. I asked him last week what the opening plan was. He said October, tentatively."

"October."

"He's been building the material inventory since June. The booking system is ready — I helped him set it up in May. I built it in a spreadsheet with a three-month forward view and he's been updating it himself since. He's very precise with spreadsheets. Faster than I expected, honestly."

"You helped him set up the booking system in May," I said.

"It's not complicated. He said he needed something he could update himself without needing to call someone. So I built it simple and showed him the interface twice. He had it by the second day." A pause. "He said the water-pattern dyeing instruction is the session he wants to lead first. He's been testing the instruction sequence on Mother."

I looked at the July campus through the lab window. "Mother is the first student."

"She's been very patient," Xiaoyu said. "She said she already knows how to do the water-pattern dyeing but she's been pretending she doesn't so he can practice the instruction sequence."

I sat with that for a moment.

"She said it was good instruction," Xiaoyu added. "She said she learned things she hadn't known. He's been doing this for thirty years — even the people who think they know the technique apparently have gaps."

"Even though she already knew."

"He's been doing this for thirty years. She said she realized she'd been doing parts of it slightly wrong for twenty." A pause that carried what might have been a smile in audio. "She told me that privately. She didn't tell him."

"Why."

"Because he already knows the instruction is good. He doesn't need to know about her twenty years of slight errors. Those are two different kinds of information."

She was right. She was almost always right about the social layer of things, which was a capability I'd developed late and she'd had since she was small. Xiaoyu had learned early that different kinds of information served different people in different ways, and she filed them accordingly. The twenty years of slight errors was filed under: information that would not benefit Father to know. The information that had benefited Father was the craft workshop pricing research, which she'd filed there. The distinction was not strategy — it was care, which was a different thing.

Father's daughter.

***

The guild's server standing by August: Rank 4 on the Tianlong server, which was the highest independent guild ranking since the server had opened. Tianxia Coalition was distributed across three entries in the top twenty, none above rank 6. Wang Jian's restructuring after CW I had produced a leaner coalition — smaller core, higher per-member quality — and the three entries represented different branches of that structure.

Wenqing monitored it on the weekly report. The CW II bracket opened in November and registration in October. Nothing was imminent.

TwilightTide's council recommendation had produced one addition since June: Cloudrift, a strategic analyst who ran parallel analysis work. He'd been in TwilightTide's secondary analysis pipeline since January — she'd been quietly evaluating him for eight months before recommending him. The council addition vote had been unanimous after her recommendation.

Old Wolf had asked her before the vote: "You're recommending a second analyst."

"I'm recommending the right person for the council. The fact that he runs analysis doesn't make him redundant — it makes the analysis have a second check. Two analysts can find each other's errors. One analyst can only find the errors they're capable of finding."

"You trust him."

"I trust his methodology. I've been working from his secondary analysis for four months and I've found no errors in his core approaches. He found two errors in mine. Both were small. I corrected them."

Old Wolf had looked at her for a long moment with the expression he used when someone had said something he hadn't expected. "Four months of secondary analysis without errors."

"Four months of secondary analysis that I couldn't find errors in," she said. "That's the correct framing. Absence of evidence isn't evidence of absence. I've been looking carefully."

He'd voted yes.

By August, the guild was running with two analysts, a dedicated logistics specialist, and an Iron Hills three AM window that had four regular members on varying schedules. The formation analysis quality had improved every month since September 2015. The kill record stood at 14 consecutive server-firsts on Floors 7 through 20.

Pioneer's Path: cycle 128 as of mid-August.

Thirty-two cycles to the next Heritage Fragment node. October at current rate. Maybe November if the summer sessions ran shorter than planned, which they wouldn't.

I ran cycle 129 on a Thursday morning in the third week of August while Wanqing worked on the Japanese monograph at the bench and the summer cicadas ran their seasonal register in the campus trees. She didn't look up. The monograph had something in it she was still looking for.

The summer cicadas had a specific quality at this point in August — still at full volume, not yet at the diminishing end-of-summer register, but the sound was denser in the early mornings and slightly thinner by midday in a way that had started two weeks ago. The season's own countdown, audible if you were paying attention. I'd been paying enough attention this summer that I could track the progression week to week. Something about the slower pace of the post-surgery months — the load shifted, the urgency recalibrated — had made the sensory layer more available. I'd been noticing things I had been too focused to notice when everything was oriented toward the fund and the timeline. The cicadas. The problem set progression. The specific color of the maple in high summer. The way the bench's stone arm was warmer on Thursday mornings than Tuesday ones, for reasons that had to do with the angle of morning light through the adjacent building's gap. Small facts. The summer accumulating them.

The summer continued.