I am working on redesigning the deep-level setting for my Rynn series. The idea is to have several thousand (up to a hundred thousand) interconnected Places, spread out among a hundred or so different physical worlds, a significant number of which may be normally causally disconnected and obeying slightly (or wildly) different temporal flows.

The basic idea is that the portals are the energy and information pipelines of a (possibly transcended) civilization, and the story-focus users of the portals are merely infinitesimal parasitic motes of dust (aka humans and the like) that get carried along in the exchange, even though this was not even remotely the purpose behind the system's creation. Now given the temporal mechanics in play, the current users may well have had millenia to understand and customize their interfaces to this portal system. Heck, minor civilizations may have arisen, built such interfaces, and wiped themselves out long ago (or perhaps they are yet to arise, it's really rather confusing).

Now the portals need not be shimmering mirrors, need not be on 100% of the time. There are no real constraints on shape or transit structure. Transit is instantaneous from the perspective of the object being transported, although timelines (and even time linearity) might vary dramatically across systems. Remember, this is not a system designed for transport, it's just a part of a truly massive system used by tiny marginal powers for this purpose.

Now, if I had a Place (or series of Places) that served no other purpose other than to be Interchanges (central nodes for travel), how would they be designed? I'm thinking maximizing defensive potential by the controlling power (some sort of IFF system?), practicality, ease-of-use, volume capacity, possibly in that order.

  • $\begingroup$ If the Interchanges exist specifically to make travelling easier, why would there be any consideration other than traffic flow? Is/was there some entity(ies) the creators are/were concerned about? $\endgroup$
    – Frostfyre
    May 13 '15 at 0:12
  • $\begingroup$ The interchanges are hacks built on top of the transdimensional pipelines by (comparatively to the pipe creators) primitive civilizations. Might still be aeons ahead of us today, but primitive. The interchange controllers (seem to be gone, too now) may still have choosen to close some portals, limit flow from others, favor more closely temporally aligned destinations etc. $\endgroup$ May 13 '15 at 0:14
  • 4
    $\begingroup$ The Stargate network seems to fulfill all your criteria: created by an Ascended predecessor race; 100,000+ gates, each on a different planet; not continuously operating; instantaneous transport; "aftermarket" IFF system (iris/Atlantis gate shield); ease of use is comparable to dialing a telephone; no limit on volume, except the speed you can shove stuff through. $\endgroup$ May 13 '15 at 0:38
  • $\begingroup$ @2012rcampion unless you're the wrong size, don't know the correct "dial-in" codes in advance, etc. The idea is to have the interchanges more easily reached by the lesser races than the pipe conduit system itself. $\endgroup$ May 13 '15 at 0:54
  • $\begingroup$ Easy: if the 'lesser races' are the ones constructing the gate system (piggybacking on top of the underlying wormhole network) then they simply distribute a directory with each DHD. (The gate-builders had terrible UX designers: the DHD has knowledge of every gate, but no way to access/display those addresses. In your world you could just add a nice touchscreen interface...) Also, we've seen (via Orlin and the Ori) that you can make gates of any size, as large or small as you want. $\endgroup$ May 13 '15 at 1:00

This turned out to be a little long-winded, but it was fun. This is probably one of my favorite parts of building new worlds.

TL;DR: A collection of intelligent, formless interdimensional zones that form a constellation around Places and a line-of-sight network across the dimensions and through space.

The first things I thought of when reading your question were the Transitive Planes from Dungeons & Dragons lore and an interdimensional area that links magical doorways in Dragon Age: Inquisition. 2012rcampion mentions the Stargate series in the comments. My answer will pull from elements of all these.


The Transitive Planes are the Astral, Ethereal, and Shadow Planes. They're pretty much the glue that connects the other planes in the D&D cosmos. Like the Astral Plane, your Interchanges connect Places to each other. The other two planes link locations on the Material Plane, which, for your purposes, would be locations in the same Place.

In Dragon Age, Eluvians are elven portals that take a traveler into a nexus. The portals can only be opened if the traveler has a certain token, with each token keyed to a certain portal(s).

Stargate features ring-shaped devices built by the Ancients that create wormholes between them. The Ancients used force fields to keep gates secure, while modern races rely on more mundane methods such as Earth's metal iris and the Goa'uld's use of soldiers.


These Interchanges are nebulous, interdimensional points in an effectively limitless network. Rather than allowing an infinite number of dimensional tunnels to form a hazardous web between the dimensions, gateways lead from a Place to one of these Interchanges. The Interchanges exist in a sort of orbit, not really anchored to the gateways of the Place they orbit as much as the Place itself. This ever-shifting constellation of Interchanges around each Place means gateways can become inactive as it loses line of sight with one Interchange and before it gains line of sight with the next. If there are two Interchanges with line of sight, the gateway remains closed as a safety precaution, to prevent duplication or splicing of travelers.

Whereas gateways can only be used to travel from a Place to an Interchange, Interchanges can be used to travel immense distances or to hop from one dimension to another. Each Interchange has a Master Gateway. A traveler in possession of a Key can access a specific subset of addresses for other Interchanges unique to that Key or Key type. The Master Gateways hold the addresses of every other Master Gateway in the entire network. With a Key, a traveler can use a Master Gateway to create a connection to any other Master Gateway, but only if they have line of sight to each other. This makes dimension hopping more difficult as both the dimensions and the gateways have to be in proper alignment. Master Gateways are intelligent in their own right, allowing them to build a multi-stage tunnel that links any number of them to build a single continuous tunnel, allowing a traveler to move from one Interchange to another so long as some number of intermediary Master Gateways have line of sight to each other; the traveler need never know how many steps were taken to reach her destination.

The gateways in each Place are protected from unwanted access by Tokens. Without the proper Token in her possession, a traveler is refused entry to her destination. The Interchanges collect these wayward travelers if they find themselves rejected in transit, preventing their unwitting destruction. The Interchanges are designed to never reject a traveler, but someone who loses their access rights winds up trapped in the limitless network no matter what they try.

Tokens can be anything: a birthmark, a tattoo, a piece of jewelry, a physiological trait (e.g., red hair). A Place's gateways can be reprogrammed with the proper means, allowing each gateway to have different Tokens over time. This programming also includes such factors as maximum transfer rate, which can be used to ensure that no more than ten people can come through each minute, or maybe hundred, a thousand, or just one per day.

The gateways of each Place are formless. They don't embody any solid form. Instead, they are defined by the space in which they occur. The Master Gateways, being intelligent, can manipulate the dimensions of the gateways that open to their respective Interchanges. This allows arbitrarily large or small creatures, objects, and groups to use gateways at their leisure. Like the local gateways, the Master Gateways can be reprogrammed to limit flow and reject connections. Likewise, Master Gateways can refuse to participate in continuous tunnels that involve banned Interchanges.

The Interchanges themselves are formless. They are controlled by the intelligence of their resident Master Gateway. Their environment is sufficient to support any traveler inhabiting the Interchange. Master Gateways communicate with each other to establish viable conditions at destinations for travelers. Each Interchange can have a limitless number of connections to the local Place, but only those gateways that are active can be interacted with (e.g., seen, touched, used). The intelligence of the Master Gateway can also conceal or reveal active and inactive gateways, depending on specific programming and the Tokens in a traveler's possession; of course, it limits this to the overlapping subset if multiple travelers have differing Tokens.


I'm not sure how science-based your story is. And you could ignore this entirely, if you decide it doesn't apply to your world. But I think it actually adds some interesting depth, so I'll post it anyway.

The problem you're going to run into is your portals are, basically, FTL. And under relativity, an FTL portal is only safe from a causal standpoint if both ends are exactly at rest to each other, and each experiences time at precisely the same rate. In other words... never. Which means that if you had unrestricted two-way portals, they would allow time travel.

To see this, consider the following thought experiment (ignore if you're already familiar with the paradox issues). Consider a two-way portal with one end on earth, and the other in a ship travelling away. The ship is moving at a high fraction of the speed of light, such that time dilation occurs at a 1:2 ratio. But the problem is that time dilation is symmetrical. So the ship is seeing time go by on earth at 50% of normal, and earth is seeing time go by on the ship at 50% of normal. So if I leave the ship at 10 minutes through the portal, I arrive back on earth at 5 minutes (ship time), and I can then jump back to the ship at 2.5 minutes... you can see how it doesn't really work. Even trying to describe when I arrive at each end is suspect.

So, what's the solution?

Limited/one-way portals.

I've outlined my conceptual thinking about this in more depth here. But the key is that you only hit causality issues with unlimited any-way travel. If you restrict your portals, you can be safe. And I think this is where you could get an interesting plot point.

The way it would work, imagine you jump from point A 100 light years to the nexus (point B). Because of the potential paradox, this locks you out of returning immediately - the portal going back has to close. Depending on the exact timing, it might be up to 100 years, but a much shorter period - say, 1-2 years - is more likely, because not much in the universe moves at truly relativistic speeds. But keep in mind it has to close for everyone, because otherwise you could create a communication paradox - you go to B, tell Alice something, she goes back to A immediately.

Now, presumably your transcended civilization knows how to calculate this sort of thing, and they would design their system accordingly. Portals would close, change direction, and re-open based on an extraordinarily complex schedule involving the relativistic speeds of all objects in the network, plus all objects that might interact with the network. But your civilization that's using this network probably doesn't understand it fully - they probably have worked out schedules, and they might know - roughly - why things are happening, but overall it would be extremely unpredictable. It's like modern humans trying to predict weather patterns - we know roughly what will happen, but there's a lot of variation in that for things to change or unexpected actions.


So what does this mean for you? The main thing is that there will always be a degree of uncertainty about portals. They might close early, they might open late. They might flicker off for no apparent reason for a couple of hours, then turn back on. People in the interchanges could work out schedules and predictions, but they would be just that - predictions. And it's possible that things would shift and change significantly over time, causing trade centers and transportation networks to shift and evolve based on forces beyond our comprehension.

Another side effect is that communication would be restricted for long periods. So A can send news to B, but A is locked out of news from B for a long period. You might get 1 year of A -> B, then 1 year of a closed portal (to allow potential paradoxes to close up), then B -> re-opens, and repeat. Which means occasionally up to a 3-year gap.


If we assume that the network was created for energy distribution, we can assume the network to be directional graph with tree topology and real world nodes being leaves that do not directly connect to other real world portals.

Further we can assume that at the roots of the tree are power sources powering the entire network. This follows from assuming that the purpose of the network is to safely access such energy sources from the real world. We can assume there is a several such roots and that every leaf can access several since otherwise portal and everything relying on the energy it conveys could fail catastrophically at any time.

We can also assume that the roots can't be directly accessed due to being inimical to lifeforms, otherwise the portal system would be unnecessary. Although the builders of the network might have transcended that limitation and currently reside at the roots and possibly have the ability to monitor and control the network. Root access for the win...

As for the intermediate nodes, I propose a simple system with two types of nodes. Let us call them source nodes and sink nodes. Every source node connects to one power source portal root node and multiple sink nodes. Every sink node connects to one real world portal leaf node and multiple source nodes.

A failure of the power node would be dealt automatically with source node and connections to it powering down and sink nodes spreading the load to other source node connections. You'd want to make sure that sink nodes have as little as possible overlap in source nodes they connect to so that the increase in load is distributed over as many of the remaining power sources as possible.

While it probably was not a factor for the builders that same system would work in reverse to deal with lost portals. If you lose control of a portal you can close the portal at the sink node and the source nodes will send the excess power to their other sink nodes. If you lose control of the sink node somehow, you can close portals to at all source nodes you control.

If two powers control source nodes connected to the same sink node, a scenario not envisioned by the builders, attacks become possible. Although the defender would probably have a heavy advantage. The stargate solution of building the best fast closing gate you can directly in front of the gate and opening it when you control the link should work well enough. In stargate they used radio to request opening the iris when dialing from outside, but magic communication or morse code by throwing stones at the gate should work well enough.

Alternately and probably easier you could simply use a separate portal for each connection. At that point you should be able to disconnect links to nodes you do not control and probably reconnect to unused portals in nodes you do control. The network would have the management system to support this as ability to reconfigure the network topology would have one of the original design specifications. This would result in each power having an essentially fully secure network of its own, possibly with otherwise low value border nodes connecting to friendly powers for trade and diplomacy.

Even in this scenario you'd probably build gates in front of the portals, just in case.

As for topology, I think it is reasonable to assume 1 to 6 topology for nodes. This would result each sink node getting power from six source nodes and power sources. This is bit low, but would be sufficient since the system would have been built with ability to relink to new nodes in case of failure.

With near optimal topology each world would then be one hop from its own sink node, two hops from six source nodes, three hops from 30 sink nodes, four hops from 30 worlds and 150 source nodes, five hops from 750 sink nodes, six hops from 750 worlds and 3750 source nodes, seven hops from 18750 sink nodes, eight hops from 18750 worlds and enough source nodes that the network is large enough by ten hops to fulfil the size condition in the question. In reality such optimal growth would not happen since maintenance after the builders left would have been uncoordinated and locally optimized.

For defenses we can assume that in addition to the gates at portals them selves there would be fats close gates to prevent assault in from one gate and out from another. I think a topology resembling a seven pointed star would be good. Getting out from a gate would require getting thru two gate houses with guards. Getting out from another would require bypassing two other gatehouses. This would leave time to either evacuate and disconnect from lost node or get reinforcements and crush the attack as needed. The number of troops you can get thru by surprise is limited, so unless you have some extremely powerful individuals attacking, you are safe enough.

There would obviously be guards doing custom checks at any entry points to your network also. With magic it might be practical to but any people not working for the system administration and all goods in stasis for the transit thru the system. In any case I doubt the system would have the capacity for large numbers of people or cargo, if it was originally developed for energy and information. So it is actually reasonable to limit travel to employees and people specifically authorized by government. Cargo could be quite limited as well. Gold, silver, spices, magic items and reagents, supplies for expeditions and new colonies.


It sounds like these interchanges form what is essentially an inter-dimensional internet, with nodes connected by data lines.
In any kind of inter connected network you'd want to have a lot of redundancy, just in case one path was shut down.
Since it's inter-dimensional these nodes don't need to exist anywhere physical like we'd understand physical. It could just be a tiny knot of 4 dimensional space inside of a nth dimensional construct.
Each node would connect to several other nodes and/or end points (planets). Each node and end point would have an address, (like a MAC address) so if you had a map then traveling from node to node to destination would be pretty easy. The problem is knowing where you were going. You could map out the whole network, but depending on size it could take a long time to figure out all the connections. There is a lot of redundancy in the connections, so it is entirely possible to travel in a circle if you aren't paying attention.

I picture it like this. Each node is encoded on an atomic level with its unique address, and has several tunnels leading off of it. Each tunnel is encoded with the address of the node it's connected to. These node connections are static, but because of the nature of multi-dimensional space they sometimes will close or constrict due to interference, and then reopen when conditions improve. In these cases another route would need to be taken, which is why there are so many redundant connections.
Some past (or future?) races have mapped out well traveled and resilient routes using physical signs, but there are shortcuts if you have a map and the paths are open.

Most end points (planets) have only one node connecting them into the network, though it is not impossible or unheard of for popular/central planets to have more than one node connection.

Network map

In this example you could get from Red to Brown in as few as 6 hops, or a lot more if there was a lot of path closures happening.


Depending on the mechanics of your portals, you may not need any physical space for your interchange(s). I'm going to make the assumption that portals are a fixed in place on a galactic scale (from Planet A to Planet B, from Planet B to Planet C, but not from Planet A to Planet C), but on a local scale are not fixed.

If you can superpose portal exits and entrances (put them in the exact same place), then you could create a route from any point in your network to any other point. Think of it like a train station, where instead of commuters changing trains to go further afield, the whole train changes tracks to suit the commuter. This gives you the effect of a direct A-C connection, but only if interchange B routes you there.

This itself is a good defence mechanism. Route your enemy back to where they started. If you're less compassionate, route them to an icy hell-hole of a planet that has no entrance portals.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.