The mountain build took six days.
Fourteen segnts through tamorphic rock under compressive geological load. The compressed substrate needed longer to accept each segnt than anything the previous builds had required—two and a half hours per segnt rather than two, the ancient grammar crystallising slowly through stone that had been under pressure for geological ti and didn’t release anything quickly. The pool cost per segnt was higher than the highland drainage build. Higher than the corridor build before it. Each descent cost more than the western sequence’s first two stages had prepared him for, and the work was quiet in the way that expensive work was always quiet: no margin for anything except the build.
The ancient network activated on day two.
He had felt the previous activations—the corridor’s network coming online mid-build, the highland drainage’s ancient staging responding to the first completed segnt. Both had changed the work’s character. What happened in the mountain on day two was a different order of response entirely. The network here was deeper and denser than anything in the previous builds, the staging reaching further into the substrate and the substrate itself having more capacity to route through it. When it activated it didn’t feel like assistance. It felt like the build joining sothing that had been waiting centuries for an extension.
Third activation. Each one stronger than the last. The western sequence is teaching him the ancient network’s range by giving him progressively more powerful examples. The designer built the teaching into the order of the builds.
Soren’s efficiency numbers confird what the carrier function was reading. Each segnt in the mountain corridor produced more improvent in the source’s routing than any segnt in either previous build. Not because the work was faster—it wasn’t, it was the slowest build yet. Because the ancient network here had more capacity. Each segnt plugged into a routing architecture that had been designed for exactly this kind of load and had been waiting to carry it for as long as the source had needed workarounds.
He built. The days compressed into the work’s rhythm. Down. Build. Surface. Recover. Down again.
By day four the rhythm was settled enough that the group had stopped marking his surfacings as events. The older man cooked when Kai was below. Neral docunted. Soren ran his instrunts. Mira read the vault pair. The work had its own shape and the group had fitted themselves into it the way they fitted into every build: precisely, without waste, without being asked.
The fourteenth segnt set on day six.
He felt the connection complete below the mountain’s stone—the lateral stage clicking into the ancient network’s deep architecture, the direct path opening, the source’s workaround routing through this region beginning to redirect.
And then two things happened in the sovereign seed simultaneously.
The source’s workaround load through this mountain corridor dropped, as he had known it would. Expected.
And one of the five managed Rifts changed character in the sovereign seed’s five-node integration. Not its managent quality. Not its conducted pattern’s precision. Its weight. The Ren-Sarath entity—which had been carrying the highest deep pressure of any active node since the chain completions, sitting in the source’s primary workaround path through the mountain region—felt asurably different. Lighter. As if sothing it had been bearing for a very long ti had simply been set down.
He surfaced.
Mira had the vault pair in both hands.
"Ren-Sarath," she said. She was reading carefully, the way she read when the device was showing her sothing she needed to be precise about. "The conducted pattern’s quality is unchanged—the managent precision the entity developed after the source told it about the surface world is still there. But the weight of what it’s carrying has dropped significantly. The pattern is running at the sa interval with less to push against."
She held the shells.
"Eight hundred years," she said.
That was all.
He sat with that for a mont.
The Ren-Sarath entity had been carrying the source’s overflow for eight hundred years without knowing it was doing so. Without knowing why its load was heavier than the other entities. Without knowing the surface world existed, or that the carrier function was designed to eventually do sothing about it.
Eight hundred years of extra weight. Gone in the mont the fourteenth segnt set.
He thought about what eight hundred years felt like from the inside—and then stopped, because he didn’t have a frawork for it and building one would only be speculation. What he had instead was this: the entity was lighter now than it had been when it was ford, for the first ti in its existence, and it hadn’t known until this mont that lighter was sothing it could be.
That’s the one that doesn’t go in a report.
He found he was glad about this. He found that was the correct response and it didn’t require much more thought than that.
Neral was adding a page to the docuntation.
"The load reduction correlates directly and imdiately with lateral stage completion," he said. He was writing while he spoke. "The docuntation needs to capture this. Future carriers need to understand that each lateral stage is not just substrate maintenance—it’s reducing work that the entities in the affected workaround path have been doing involuntarily. Without knowledge of why. Without compensation in the form of lighter work until the carrier addresses the source’s routing."
He wrote.
"The carrier is not just building infrastructure. The carrier is settling debts the system accumulated over geological ti."
He finished the paragraph and didn’t look up from the page. Neral understood that so observations needed to be written while they were still precise. This was one of them.
The director’s routing ssage arrived an hour later.
"Ren-Sarath’s conducted oscillation changed thirty minutes ago," he had written. "Load amplitude down eighteen percent. The prototype instrunt is reading the corresponding change in the source’s substrate routing through the mountain region simultaneously. I have correlated the lateral stage completion with the entity load reduction and the source routing change as a single event."
A pause in the ssage, which usually ant he had stopped writing and then continued.
"I can see what you’re building from here. I can see it working. I’ve been monitoring Rifts for twenty years. I have never been able to watch the system improve in real ti before."
He is giving the director sothing to watch. Sothing that responds imdiately, that shows its results in data the director knows how to read. Twenty years of monitoring a system that didn’t visibly change, and now every stage shows up in the instrunts within the hour.
He found that was more satisfying than he would have predicted.
He put the ssage away and read the substrate map.
Three western gaps complete.
One remained in the western hemisphere—two days south of the mountain range, at the headwaters of a river system the Guild hadn’t mapped because no zone activity had ever drawn monitoring attention there. The substrate in that region was water-carved: millennia of the river system’s source flow had shaped the geological layer, and the ancient network stages below the headwaters had been partially eroded at their outer edges by the constant moisture in the substrate.
Not degraded to the point of failure. Just worn at the contacts.
The build wouldn’t fail. But it would require more attention at the connection points than the compressed substrate had required, where the ancient grammar crystallised slowly but cleanly. Here the contacts would need to be assessed and rebuilt before the stage could hold. A different kind of problem from a different kind of geology.
Open substrate. Dense highland. Compressed tamorphic. Now degraded contacts. Four different substrate conditions in four western gaps. He had not encountered the sa problem twice. The designer didn’t repeat variations in the teaching sequence. The western gaps were not four examples of the sa build. They were four different builds that happened to share a purpose.
He had built in open substrate, in dense highland substrate, in compressed tamorphic substrate. A partially-degraded ancient network was the fourth variant the western sequence had left him.
The eastern gaps were larger. Which ant they were more complex. Which ant the things the western work had taught him about substrate variation, about cost curves under different geological conditions, about what the ancient network’s capacity felt like when it was deep versus dense versus worn—all of it was going to matter in the east in ways that weren’t predictable yet.
Four western gaps to teach him the variants. The designer didn’t know what carrier would co. Didn’t know when. Designed the teaching to work regardless. Whatever arrives, the western sequence makes it ready for the eastern work. He was almost at the end of it.
He looked at the group.
"Two days south," he said. "Last western gap."
The older man was already packing.
User Comments
0 comments from readers