Some Old Spaceship Models

Working on models of spacecraft modules this week reminded me that I once was actually quite into this kind of thing, so I dug out some old POV files of spacecraft models stored on an external harddrive from my childhood (the files, not the harddrive--that's only a few years old), and I had quite a lot of stuff saved that I had forgotten about. Apparently, I was at one point working on a model of a universal docking adapter, which I never finished:

My old spacecraft models have a number of other repeated components--engines, fuel tanks, etc.--but most of them do not have their own separate model files yet. A few which do include these rocket engine exhaust plumes:

And this RCS flywheel housing:

And it seems that I got most of the way through a decent recreation of the forest domes from the Valley Forge, from the old sci-fi film Silent Running:

And it seems I have one incomplete spacecraft design loosely based on the Valley Forge aesthetic:

And I've got a partial section of a spar intended for supporting a fission reactor, which I think was just an experiment in trying to get gold foil surface texturing to look right:

In terms of complete spacecraft models, the Minos is not fully detailed but it is a whole ship, which I used for TTRPG purposes in high school:

Here's a model of a rotating dumbell station, with the crew module at one end connected to a central airlock, and a counterweight for equipment:

Next we have a nuclear rocket design, with the engine markedly far away from the crew quarters and other ship components on a long spar:

A small asteroid tugboat:

Which is carried by a bussard ramjet:

And I played around a little bit with schematics for possibly slightly more realistic magscoops / magsails:

Here we have a large interstellar probe with instrument booms:

A gigantic tanker with three engines and a tiny little crew module up front:

And a final design vaguely modelled off the Orion concept, but not actually using an Orion pusher plate, which I called the Pilgrim:

The key feature of this last ship is the rotating passenger sections supported by cabling on winches, which swings and and out to keep the passenger modules aligned with apparent gravity as the ship switches between thrust and spin.

I've put all of these old models, my new tiling modules, and some other cruft that needs organizing, in a GitHub repository, so y'all can follow along as I clean stuff up and get those old components integrated into my new spacecraft design system.

Dense Fractal Space Construction

Suppose we want to design a series of basic reusable modules that can serve as the basis of a design aesthetic for some spacefaring civilization.

Let's start with a sphere, for maximum strength and material economy. It should be as easy as possible to expand stations or ships built out of these modules over time, and we'd like subsequently-added sphere modules to pack together as tightly as possible to minimize travel distances within the structure. That means each sphere should be able to connect to 12 others (the 3 dimensional kissing number), so each basic module should have twelve cylindrical attachment points (unused ones can be capped off with windows or airlocks or sphere-section bulkheads). There are two ways to arrange those--with stacked layers either aligned with each second layer over, or shifted by one lattic position--but the aligned arrangements, corresponding to placing connections at the vertices of a cuboctahedron (or on the faces of a rhombic dodecahedron) maximizes symmetry and maximizes the number of straight line paths from one attachment point through the center of the sphere to an opposite attachment point, which makes it easy to build straight-line structures with individual modules in many different orientations.

Standardized modules and standardized attachment points, resulting in standard exterior spacing between components, also suggests that these modules should come with standardized exterior anchor points for EVA, which means nicely standardized greeblies completing the design aesthetic. While doing a high ropes course once, I came across an ingenious design for a passive Continuous Belay System (an system which allows you to move around a structure relatively freely without ever becoming disconnected from a safety line)--a special C-clip is used which has an opening large enough for flat plates to pass through, which allows passing by mounting points for cables, but small enough that it cannot come off the cable itself. Specially-designed intersection plates also allow sliding the clip from one cable to another without ever becoming detached. This is slightly more restrictive than two-carabiner systems, where you move one at a time while keeping the other attached, but also much more idiot-proof, and provides a functional set of greeblies in terms of the anchor and intersection plates. The details of how the plates are designed are not easily visible when zoomed out to look at a whole module, but they do provide nice greebling.

Now, you can just densely tile space with the standard minimum-sized modules. But a collection of 13 basic modules forming a "raspberry" has the 12 exterior modules arranged such that each one has one of its connection points exactly at the vertex of a larger cuboctahedron.

Thus, these raspberry units can be used to tile space as well, in exactly the same pattern as the base units. (Also note that the doubled anchor-rings around each intermodule connection provide multiple "lanes" for workers to pass each other on EVA.) One might choose to do so, leaving a lot of cluster-internal attachment points unused, so as to set up a recursive city-like structure of local neighborhoods, larger districts, and whole cities. But additionally, these can be used to define a scale for larger modules that will fit into the same system. For example, a large sphere can enclose a raspberry except where the 12 external attachments poke through, providing double-hull protection for the interior lower-scale modules, or permitting leaving out some members of the full cluster while maintaining a standard external interface. In this following image (which took forever to render), 13 raspberries are each encased in a glass secondary hull and used to form a second-scale raspberry structure.

And, once larger-scale modules are established as A Thing, you can just have monolithic modules of that size which have whatever internal configuration you want, for holding stuff that needs more space (say, gravity centrifuges, or nuclear reactors, or whatever), and smaller scale modules will fit nicely in between them.

Non-spherical modules can also be added to the system as long as they have attachment points that align with the lattice points for their specified scale. So, if you have, e.g., a long mass driver or something, you don't need to waste a ton of sphere space on it--you can make a cylindrical module with attachment points spaced appropriately to tesselate more standard modules around it. More prosaically, however, they may be some utility in having polyhedral modules.

This cuboctahedral modules, with connectors at each vertex, has less interior volume lower inherent pressure-hull capacity, but provides the convenience of flat surface for mounting equipment on the exterior, and more room between modules for maintenance access. As a raspberry cluster, it looks like this:

Meanwhile, this rhombic dodecahedral module, with connection points in the center of each face, does a better job of efficiently filling space, leaving less wasted space betwee modules. 

That's no good for external maintenance access, but when working with larger scale modules, where there would be more wasted space between spheres, this may be a good option, at least for low-pressure environments. As a raspberry cluster, it looks like this:

Truly Alien Languages

A few weeks ago, I watched Avatar: The Way of Water, and I was rather disappointed at the reduction in linguistic worlbuilding effort compared to the original Avatar. I'll do a proper review on that at some point when I am not in the middle of an interstate move with three children, but in the meantime I want to talk a bit about properly alien languages--the sort that cannot be accurately pronounced, or produced at all, by unassisted human actors, because they use sounds that humans can't make, or signs that we can't perform, or completely inaccessible modalities. Specifically, I want to talk about how these show up in movies and TV shows--or rather, how they don't show up.

The Star Wars universe numerous aliens that produce non-human-like speech--most notably, Wookies--but as portrayed on screen, it's all gibberish. The Prawns from District 9 (<- Amazon affiliate link) have non-human-like speech... but again, it's just gibberish. Star Trek: Picard, season 3 features a few lines of subtitled non-humanoid dialog, in a receptive-multilingualism situation in conversations between Vadic and her crew... but it's just gibberish, despite Star Trek being a franchise that has already shown willingness to invest in conlangs for Vulcan and Klingon. And Star Trek: Discovery, season 4 (proper review coming later) features a modally-alien language encoded in a combination of spatially-and-temporally distributed visual signals and chemical messaging... but it's just gibberish!

Clearly, modern TV and movie productions are willing to invest in producing proper conlangs, and they are willing to pay audio and visual artists to produce asemic representations of Truly Alien Languages; so why have we not yet seen these two things combined? Even the Avatar franchise, which introduced the Na'vi language, seems to have completely punted on the language of the Tulkun, which to all appearance is whalesong-esque (you guessed it!) asemic gibberish!

Two plausible reasons have been suggested to me by other members of the Language Creation Society: 

I fear studios wouldn't be interested in commissioning a conlang they wouldn't be able to use for promotional purposes, unlike, for instance, High Valyrian.

--Baptiste Faussat

I interpret "promotional purposes" in this sense as meaning "that fans can learn"--but most of the conlangs that, e.g., David Peterson has been comissioned for have never developed huge fan speaker communities, or any fan speaker communities, and there frequently isn't enough public material for that to even be possible, yet such commissions continue.

I feel like a non-human-pronounceable conlang in a movie would be a nightmare of going back and forth with sound designers and such.

--George Corley

As opposed to just telling the designers to go wild once, and not requiring any back-and-forth with multiple people. That seems... unfortunately plausible. It's also entirely plausible that it has just never crossed the mind of any producer or director that anyone would ever care, and it probably won't until there is some initial breakthrough production that demonstrates it. I had hoped that Avatar: The Way of Water would turn out to be such a production, but it failed me. So, the opportunity to be the first is still open!

At the moment, I am putting my hopes in Project Hail Mary, which is currently in pre-production to star Ryan Gosling as Ryland Grace. (See also my previous linguistic review of the book, along with other works by Andy Weir.) Decoding the Truly Alien Language in this story is in fact a significant feature of the plot, so I rather hope that they will see fit to actually produce a language to decode! But given my disappointment with Avatar: The Way of Water, I kinda worry that they'll make a hash of it after all unless someone pushes for them to do it right. So, here's my public plea to Phil Lord, Christopher Miller, Ryan Gosling, Andy Weir, and anyone else involved in the production: please, hire a conlanger for Eridian!

(Incidentally, I'm available! Contact me! But soliciting applications through the LCS is also an excellent option!)

Stay tuned for a future blog post where I start working out details of a canon-compatible Eridian language, after I dig my copy of Project Hail Mary-the-book out of whichever box it ended up in so I can re-extract all the details that I didn't quote in my last review. I can probably even build off my previous work on Fysh A and Tjugem to produce synthesized speech samples for Rocky, though I admit I have no idea whether that would be considered helpful or annoying by any potential film sound design department that one might end up going back-and-forth-with....

If you liked this post, please consider making a small donation!

The Linguistically Interesting Media Index